WO2022141121A1 - Image transmission method, mobile platform, remote control device, system, and storage medium - Google Patents

Abstract

An image transmission method, a mobile platform, a remote control device, a system, and a storage medium. The method comprises: decoding an encoded image frame sent by an encoding end to obtain a target image frame, and determining first feature information of the target image frame (S101); acquiring second feature information of a reconstructed frame sent by the encoding end, the reconstructed frame being obtained by decoding the encoded image frame by the encoding end (S102); and if the deviation between the first feature information and the second feature information is greater than a preset threshold, sending error feedback information to the encoding end, the error feedback information being used for instructing the encoding end to transmit an error tolerant frame (S103). The method can improve the reliability of wireless image transmission.

Description

Image transmission method, movable platform, remote control device, system and storage medium technical field

The present application relates to the technical field of wireless image transmission, and in particular, to an image transmission method, a movable platform, a remote control device, a system and a storage medium.

Background technique

At present, the encoding end of the wireless image transmission system can transmit the encoded video code stream to the decoding end through the wireless channel, and the decoding end decodes and displays the received video code stream. In the process of video transmission, due to the deterioration of the channel quality of the wireless channel, the video code stream sent by the encoding end may not be sent to the decoding end in a timely manner with high quality. A common practice at present is that the decoding end judges whether an error occurs in the video stream based on a cyclic redundancy check code (CRC), and requests an error-tolerant frame from the mobile platform after judging the error. However, this method of judging whether the image is transmitted correctly based on the correctness of the video stream will have false detection. If the error is not detected correctly and timely and the error is corrected based on the error-tolerant frame retransmitted by the encoding end, the decoding end will not be able to operate normally. Decoding the video may even cause this error to continue to spread, causing the video displayed on the decoding end to appear blurry, and the user experience is not good.

SUMMARY OF THE INVENTION

Based on this, the embodiments of the present application provide an image transmission method, a movable platform, a remote control device, a system and a storage medium, which aim to accurately and timely find errors in the image transmission process and improve the reliability of wireless image transmission.

In the first aspect, the embodiment of the present application provides an image transmission method, which is applied to a decoding end, and the decoding end is communicatively connected to the encoding end, and the method includes:

Decoding the encoded image frame sent by the encoding end to obtain a target image frame, and determining the first feature information of the target image frame;

Obtain the second feature information of the reconstructed frame sent by the encoding end, wherein the reconstructed frame is obtained by the encoding end decoding the encoded image frame;

If the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the encoder, where the error feedback information is used to instruct the encoder to transmit an error-tolerant frame.

In the second aspect, the embodiments of the present application also provide an image transmission method, which is applied to an encoding end, and the encoding end is communicatively connected to a decoding end, and the method includes:

acquiring second feature information of the reconstructed frame corresponding to the encoded image frame;

Send the encoded image frame and the second feature information to the decoding end, so that the decoding end decodes the encoded image frame to obtain the target image frame, and determines the target image frame. First feature information, if the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the encoding end, where the error feedback information is used to indicate the encoding The terminal transmits error-tolerant frames.

In a third aspect, the embodiments of the present application also provide an image transmission method, which is applied to an encoding end, and the encoding end is communicatively connected to a decoding end, and the method includes:

Send the encoded image frame to the decoding end, so that the decoding end decodes the encoded image frame to obtain the target image frame and determine the first feature information of the target image frame;

obtaining the second feature information of the reconstructed frame corresponding to the encoded image frame, and obtaining the first feature information sent by the decoding end;

If the deviation between the first feature information and the second feature information is greater than a preset threshold, an error-tolerant frame is sent to the decoding end for the decoding end to decode the error-tolerant frame.

In a fourth aspect, an embodiment of the present application further provides a remote control device, the remote control device is communicatively connected to the movable platform, and the remote control device includes a memory and a processor;

the memory is used to store computer programs;

The processor is configured to execute the computer program and implement the following steps when executing the computer program:

Decoding the encoded image frame sent by the movable platform to obtain a target image frame, and determining the first feature information of the target image frame;

acquiring second feature information of the reconstructed frame sent by the movable platform, wherein the reconstructed frame is obtained by decoding the encoded image frame by the movable platform;

If the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the movable platform, where the error feedback information is used to instruct the movable platform to transmit an error-tolerant frame.

In a fifth aspect, an embodiment of the present application further provides a movable platform, the movable platform communicates with a remote control device, and the movable platform includes a memory and a processor;

the memory is used to store computer programs;

The processor is configured to execute the computer program and implement the following steps when executing the computer program:

acquiring second feature information of the reconstructed frame corresponding to the encoded image frame;

Send the encoded image frame and the second feature information to the remote control device, so that the remote control device can decode the encoded image frame to obtain a target image frame, and determine the target image frame. First feature information, if the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the movable platform, where the error feedback information is used to indicate the The removable platform transmits fault-tolerant frames.

In a sixth aspect, an embodiment of the present application further provides a movable platform, the movable platform communicates with a remote control device, and the movable platform includes a memory and a processor;

the memory is used to store computer programs;

The processor is configured to execute the computer program and implement the following steps when executing the computer program:

Send the encoded image frame to the remote control device, so that the remote control device decodes the encoded image frame to obtain a target image frame and determine the first feature information of the target image frame;

acquiring the second feature information of the reconstructed frame corresponding to the encoded image frame, and acquiring the first feature information sent by the remote control device;

If the deviation between the first feature information and the second feature information is greater than a preset threshold, an error-tolerant frame is sent to the remote control device for the remote control device to decode the error-tolerant frame.

In a seventh aspect, an embodiment of the present application further provides a control system, the control system includes the above-mentioned remote control device and the above-mentioned movable platform, wherein the remote control device is communicatively connected to the movable platform, so The remote control device is used to control the movable platform.

In an eighth aspect, an embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the above-mentioned The steps of the image transmission method.

The embodiments of the present application provide an image transmission method, a movable platform, a remote control device, a system, and a storage medium. The target image frame is obtained by decoding the encoded image frame sent by the encoding end, and the first image frame of the target image frame is determined. One characteristic information, and then obtain the second characteristic information of the reconstructed frame corresponding to the encoded image frame sent by the encoding end, if the deviation between the first characteristic information and the second characteristic information is greater than the preset threshold, then send error feedback information to the encoding end , to instruct the encoder to transmit error-tolerant frames, so that errors in the image transmission process can be accurately and timely detected, and at the same time instruct the encoder to transmit error-tolerant frames. transmission reliability.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not limiting of the present application.

Description of drawings

In order to explain the technical solutions of the embodiments of the present application more clearly, the following briefly introduces the accompanying drawings used in the description of the embodiments. For those of ordinary skill, other drawings can also be obtained from these drawings without any creative effort.

FIG. 1 is a schematic diagram of a scene for implementing the image transmission method provided by the embodiment of the present application;

2 is a schematic flowchart of steps of an image transmission method provided by an embodiment of the present application;

3 is a schematic flowchart of steps of another image transmission method provided by an embodiment of the present application;

4 is a schematic flowchart of steps of another image transmission method provided by an embodiment of the present application;

5 is a schematic flowchart of steps of another image transmission method provided by an embodiment of the present application;

6 is a schematic flowchart of steps of another image transmission method provided by an embodiment of the present application;

7 is a schematic flowchart of steps of another image transmission method provided by an embodiment of the present application;

8 is a schematic block diagram of the structure of a remote control device provided by an embodiment of the present application;

FIG. 9 is a schematic block diagram of the structure of a movable platform provided by an embodiment of the present application;

FIG. 10 is a schematic structural block diagram of a control system provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

The flowcharts shown in the figures are for illustration only, and do not necessarily include all contents and operations/steps, nor do they have to be performed in the order described. For example, some operations/steps can also be decomposed, combined or partially combined, so the actual execution order may be changed according to the actual situation.

Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments described below and features in the embodiments may be combined with each other without conflict.

At present, the encoding end of the wireless image transmission system can transmit the encoded video code stream to the decoding end through the wireless channel, and the decoding end decodes and displays the received video code stream. In the process of video transmission, due to the deterioration of the channel quality of the wireless channel, the video code stream sent by the encoding end may not be sent to the decoding end in a timely manner with high quality. A common practice at present is that the decoding end judges whether an error occurs in the video stream based on a cyclic redundancy check code (CRC), and requests an error-tolerant frame from the mobile platform after judging the error. However, this method of judging whether the image is transmitted correctly based on the correctness of the video stream will have false detection. If the error is not detected correctly and timely and the error is corrected based on the error-tolerant frame retransmitted by the encoding end, the decoding end will not be able to operate normally. Decoding the video may even cause this error to continue to spread, causing the video displayed on the decoding end to appear blurry, and the user experience is not good.

In order to solve the above problems, the embodiments of the present application provide an image transmission method, a movable platform, a remote control device, a system and a storage medium. The first feature information of the target image frame, and then the second feature information of the reconstructed frame corresponding to the encoded image frame sent by the encoding end is obtained. If the deviation between the first feature information and the second feature information is greater than the preset threshold The end sends error feedback information to instruct the encoder to transmit error-tolerant frames, so that errors in the image transmission process can be accurately and timely detected, and at the same time instruct the encoder to transmit error-tolerant frames. The decoder can recover errors based on the error-tolerant frames retransmitted by the encoder. , which improves the reliability of wireless image transmission.

Wherein, the encoding end may include a movable platform, and the decoding end may include a remote control device. Please refer to FIG. 1. FIG. 1 is a schematic diagram of a scene for implementing the image transmission method provided by the embodiment of the present application. As shown in FIG. 1 , the scene includes a movable platform 100 and a remote control device 200 , the movable platform 100 is connected in communication with the remote control device 200 , and the remote control device 200 is used to control the movable platform 100 . The movable platform 100 includes a platform body 110 , a power system 120 provided on the platform body 110 , an image acquisition device 130 and a control system (not shown in FIG. 1 ), and the power system 120 is used to provide moving power for the movable platform 100 , the image acquisition device 130 is used to acquire an initial image frame.

The power system 120 may include one or more propellers 121 , one or more motors 122 corresponding to the one or more propellers, and one or more electronic governors (referred to as ESCs for short). Wherein, the motor 122 is connected between the electronic governor and the propeller 121, and the motor 122 and the propeller 121 are arranged on the platform body 110 of the movable platform 100; the electronic governor is used for receiving the driving signal generated by the control system, and according to the driving signal A driving current is provided to the motor 122 to control the rotational speed of the motor 122 . The motor 122 is used to drive the propeller 121 to rotate, thereby providing power for the movement of the movable platform 100, and the power enables the movable platform 100 to achieve one or more degrees of freedom movement. In certain embodiments, the movable platform 100 may rotate about one or more axes of rotation. For example, the above-mentioned rotation axes may include a roll axis, a yaw axis, and a pitch axis. It should be understood that the motor 122 may be a DC motor or an AC motor. In addition, the motor 122 may be a brushless motor or a brushed motor.

Wherein, the control system may include a controller and a sensing system. The sensing system is used to measure the attitude information of the movable platform, that is, the position information and state information of the movable platform 100 in space, such as 3D position, 3D angle, 3D velocity, 3D acceleration and 3D angular velocity. For example, the sensing system may include at least one of a gyroscope, an ultrasonic sensor, an electronic compass, an inertial measurement unit (Inertial Measurement Unit, IMU), a visual sensor, a global navigation satellite system, a barometer, and other sensors. For example, the global navigation satellite system may be the Global Positioning System (GPS). The controller is used to control the movement of the movable platform 100, for example, the movement of the movable platform 100 can be controlled according to the attitude information measured by the sensing system. It should be understood that the controller can control the movable platform 100 according to pre-programmed instructions.

In one embodiment, the controller is further configured to acquire the second feature information of the reconstructed frame corresponding to the encoded image frame; send the encoded image frame and the second feature information to the remote control device 200, so that the remote control device 200 can encode the encoded image frame and the second feature information. Decode the image frame obtained after obtaining the target image frame, and determine the first feature information of the target image frame, if the deviation between the first feature information and the second feature information is greater than the preset threshold, send error feedback information to the movable platform , the error feedback information is used to instruct the movable platform to transmit error-tolerant frames. Therefore, the remote control device 200 can accurately and timely find errors in the image transmission process, and at the same time instruct the mobile platform 100 to transmit error-tolerant frames. Reliability of image transmission.

The remote control device 200 includes a display device 210, and the remote control device 200 displays the initial image frame sent by the movable platform 100 through the display device 210 for the user to view. It should be noted that the display device 210 includes a display screen disposed on the remote control device 200 or a display independent of the remote control device 200, and the display independent of the remote control device 200 may include a mobile phone, a tablet computer, a personal computer, etc. Other electronic equipment with a display screen. Among them, the display screen includes an LED display screen, an OLED display screen, an LCD display screen, and the like.

In one embodiment, the remote control device 200 is further configured to decode the encoded image frame sent by the movable platform 100 to obtain the target image frame, and determine the first feature information of the target image frame; The second feature information of the reconstructed frame, wherein the reconstructed frame is obtained by decoding the encoded image frame by the movable platform 100; if the deviation between the first feature information and the second feature information is greater than a preset threshold Platform 100 sends error feedback information for instructing movable platform 100 to transmit error-tolerant frames.

The movable platform 100 includes drones, unmanned vehicles and mobile robots, and the drones include rotary-wing drones, such as quad-rotor drones, hexa-rotor drones, and octa-rotor drones. It is a fixed-wing UAV, or a combination of a rotary-wing type and a fixed-wing UAV, which is not limited here. Remote control device 200 may include, but is not limited to, smartphones/mobile phones, tablet computers, personal digital assistants (PDAs), desktop computers, media content players, video game stations/systems, virtual reality systems, augmented reality systems, wearable devices (eg, watches, glasses, gloves, headwear (eg, hats, helmets, virtual reality headsets, augmented reality headsets, head mounted devices (HMDs), headbands), pendants, armbands, leg loops, shoes, vest), gesture recognition device, microphone, any electronic device capable of providing or rendering image data, or any other type of device. The remote control device 200 may be a handheld terminal, and the remote control device 200 may be portable. The remote control device 200 may be carried by a human user. In some cases, the remote control 200 may be remote from the human user, and the user may control the remote control 200 using wireless and/or wired communications.

Hereinafter, the image transmission method provided by the embodiments of the present application will be described in detail with reference to the scene in FIG. 1 . It should be noted that the scene in FIG. 1 is only used to explain the image transmission method provided by the embodiment of the present application, but does not constitute a limitation on the application scene of the image transmission method provided by the embodiment of the present application.

Please refer to FIG. 2, which is a schematic flowchart of steps of an image transmission method provided by an embodiment of the present application. The image transmission method can be applied to a decoding end, the decoding end is connected to the encoding end in communication, the encoding end can include a movable platform, and the decoding end can include a remote control device.

As shown in FIG. 2 , the image transmission method includes steps S101 to S103.

Step S101: Decode the encoded image frame sent by the encoding end to obtain a target image frame, and determine first feature information of the target image frame.

Exemplarily, the encoding end may include a movable platform, the decoding end may include a remote control device, the movable platform includes an image acquisition device, and the image acquisition device is used to acquire an initial image frame, and the movable platform can encode the initial image frame and send it to the The remote control device sends the encoded image frame, and the remote control device decodes the encoded image frame to obtain the target image frame. The movable platform may encode the initial image frame by means of inter-frame coding and/or intra-frame coding, which is not specifically limited in this embodiment of the present application.

In one embodiment, the first feature information of the target image frame includes at least one of the following: grayscale feature information of the target image frame, pixel information of a local area of the target image frame, hash string of the target image frame, target image The brightness information and contrast information of the frame, the image structure information of the target image frame, and the feature information extracted from the target image frame based on the preset feature extraction network. The preset feature extraction network is a pre-trained neural network, and the local area of the target image frame may include the central area, the upper left area, the lower left area, the lower right area, and the upper right area of the target image frame.

In one embodiment, the method of determining the first feature information of the target image frame may be: performing hash processing on the target image frame based on a preset hash algorithm to obtain a hash string of the target image frame, or performing hash processing on the target image frame. Perform discrete cosine transform processing, and perform hash processing on the target image frame after the discrete cosine transform processing based on a preset hash algorithm to obtain a hash string of the target image frame, wherein the preset hash algorithm includes an average hash Any of Algorithms, Difference Value Hash Algorithms, and Perceptual Hash Algorithms.

In one embodiment, the feature information extracted from the target image frame based on the preset feature extraction network may include the texture feature of the target image frame, and the training method of the neural network may be: acquiring a plurality of sample data, the sample data including image frame and the texture feature corresponding to the image frame; iteratively train the neural network according to multiple sample data until the neural network converges or the number of training times reaches a preset number of training times, and a preset feature extraction network is obtained. The neural network may include a convolutional neural network, a recurrent neural network, a deep neural network, and a recurrent convolutional neural network.

In one embodiment, the method of determining the first feature information of the target image frame may be: according to the gray value of each pixel in the target image frame, determine each image row, image column or image in the target image frame. The target gray value of the block; the gray feature information of the target image frame is determined according to the target gray value of each image row, image column or image block in the target image frame. The target gray value includes an average gray value or a total gray value determined based on the gray value of each pixel in the image row, image column or image block.

Exemplarily, the target grayscale values of each image row, image column or image block are arranged to obtain a grayscale feature vector whose dimension is the image height, image width or the number of image blocks of the target image frame; the grayscale The feature vector is normalized, and the normalized gray feature vector is rounded to obtain gray feature information of the target image frame.

For example, if the target image frame includes N1 image lines, the target grayscale values _of the N1 image lines are arranged to obtain _an N1 _- dimensional grayscale feature vector. For another example, if the target image frame includes M ₁ image columns, then by arranging the target grayscale values of the M ₁ image columns, an M ₁ -dimensional grayscale feature vector can be obtained. For another example, if the target image frame includes P ₁ image blocks, the target gray values of the P image blocks are arranged to obtain a P ₁ -dimensional grayscale feature vector, where N ₁ , M ₁ and P ₁ are all positive integers .

Step S102: Acquire second feature information of the reconstructed frame sent by the encoding end, where the reconstructed frame is obtained by the encoding end decoding the encoded image frame.

Exemplarily, the movable platform decodes the encoded image frame to obtain a reconstructed frame corresponding to the encoded image frame, then determines the second feature information of the reconstructed frame, and sends the second feature information of the reconstructed frame to the remote control device. , the remote control device receives the second feature information of the reconstructed frame sent by the movable platform. Wherein, the second feature information includes grayscale feature information of the reconstructed frame, pixel information of a local area of the reconstructed frame, hash string of the reconstructed frame, brightness information and contrast information of the reconstructed frame, image structure information of the reconstructed frame, pre- Suppose the feature extraction network extracts the feature information from the reconstructed frame, the preset feature extraction network is a pre-trained neural network, and the local area of the reconstructed frame can include the central area, the upper left area, the lower left area, and the lower right corner of the reconstructed frame. area, upper right area.

In one embodiment, the movable platform inserts the second feature information into the Supplemental Enhancement Information (SEI) field of the encoded image frame, so that the encoded image frame carries the second feature information, and sends the information to the remote control. The device sends the encoded image frame carrying the second feature information; the remote control device obtains the second feature information of the reconstructed frame from the supplementary enhancement information SEI field of the encoded image frame sent by the movable platform. By inserting the second feature information into the SEI field of the encoded image frame, the second feature information can be transmitted to the remote control device together with the code stream, so that the remote control device can obtain the second feature information in time.

In one embodiment, the movable platform sends the encoded image frame to the remote control device through the broadband wireless channel, and sends the second characteristic information of the reconstructed frame to the remote control device through the narrowband wireless channel; the remote control device obtains the movable platform through the narrowband wireless channel. The second feature information sent. Due to the high reliability of the narrowband wireless channel, in order to ensure that the remote control device can reliably receive the second feature information, the mobile platform uses the narrowband wireless channel to send the second feature information to the remote control device, thereby improving the ability of the remote control device to receive the second feature information. reliability of information.

In one embodiment, the movable platform sends the second feature information to the remote control device at preset time intervals; the remote control device obtains the second feature information sent by the movable platform at preset time intervals. The preset time may be set based on an actual situation, which is not specifically limited in this embodiment of the present application. Since the wireless communication link between the movable platform and the remote control device may be blocked or interrupted, in order to ensure that the remote control device can reliably receive the second feature information, the movable platform periodically sends the second feature information , thereby improving the reliability of the remote control device receiving the second feature information.

Step S103, if the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the encoding end, where the error feedback information is used to instruct the encoding end to transmit an error-tolerant frame .

After acquiring the first feature information and the second feature information, determine the deviation between the first feature information and the second feature information, and if the deviation between the first feature information and the second feature information is greater than a preset threshold, send the message to the mobile platform Error feedback information, the error feedback information is used to instruct the mobile platform to transmit error-tolerant frames. The error-tolerant frame is an image frame obtained by encoding based on inter-frame encoding and intra-frame encoding, or the error-tolerant frame is an image frame obtained by encoding based on intra-frame encoding. By comparing the first feature information with the second feature information, instead of comparing the entire target image frame with the entire reconstructed frame, the computational efficiency can be greatly improved. In addition, only the feature information of the reconstructed frame is transmitted, instead of Transmitting the entire reconstructed frame can reduce the amount of transmitted data and ensure the reliability of the wireless communication link between the encoding end and the decoding end.

In one embodiment, the deviation between the first feature information and the second feature information may include the difference between the grayscale feature information of the target image frame and the grayscale feature information of the reconstructed frame, and the pixel information of the local area of the target image frame. The difference with the pixel information of the local area of the reconstructed frame, the Hamming distance between the hash string of the target image frame and the hash string of the reconstructed frame, the image structure information of the target image frame and the image of the reconstructed frame Similarity between structural information, etc.

In the image transmission method provided by the above embodiment, the target image frame is obtained by decoding the encoded image frame sent by the encoding end, and the first feature information of the target image frame is determined, and then the encoded image frame sent by the encoding end is obtained. The second feature information of the corresponding reconstructed frame, if the deviation between the first feature information and the second feature information is greater than the preset threshold, send error feedback information to the encoding end to instruct the encoding end to transmit the error-tolerant frame, so that accurate and timely discovery can be made Errors in the image transmission process, and at the same time instruct the encoder to transmit error-tolerant frames, and the decoder can recover the errors based on the error-tolerant frames retransmitted by the encoder, which improves the reliability of wireless image transmission.

Please refer to FIG. 3 , which is a schematic flowchart of steps of another image transmission method provided by an embodiment of the present application. The image transmission method can be applied to a decoding end, the decoding end is connected to the encoding end in communication, the encoding end can include a movable platform, and the decoding end can include a remote control device.

As shown in FIG. 3 , the image transmission method includes steps S201 to S205.

Step S201: Determine whether there is an error in the encoded image frame sent by the encoding end.

Exemplarily, the movable platform encodes the initial image frame, and sends the encoded image frame to the remote control device, and the remote control device performs cyclic redundancy check on the encoded image frame sent by the movable platform; If the encoded image frame sent by the mobile platform fails the cyclic redundancy check, it is determined that the encoded image frame sent by the movable platform has an error; if the encoded image frame sent by the movable platform passes the cyclic redundancy check, it is determined that it can be The encoded image frame sent by the mobile platform is error free. The encoded image frames sent by the movable platform include P frames.

Step S202: If an error occurs in the encoded image frame sent by the encoder, send error feedback information to the encoder, where the error feedback information is used to instruct the encoder to transmit an error-tolerant frame.

If there is an error in the encoded image frame sent by the movable platform, the remote control device sends error feedback information to the movable platform; after obtaining the error feedback information sent by the remote control device, the movable platform encodes the initial image frame to obtain fault tolerance. frame, and decode the fault-tolerant frame to obtain the reconstructed frame corresponding to the fault-tolerant frame; the movable platform determines the second characteristic information of the reconstructed frame, and sends the fault-tolerant frame and the second characteristic information to the remote control device. By comparing the first feature information with the second feature information only when it is determined that an error occurs in the encoded image frame sent by the encoding end, unnecessary computation can be reduced, and computation efficiency and image transmission efficiency can be improved.

Wherein, the second characteristic information is determined by the movable platform when it receives the error feedback information sent by the remote control device, and the error-tolerant frame and the encoded image frame are obtained by encoding the same initial image frame by the movable platform, or, The fault-tolerant frame is obtained by encoding the initial image frame currently collected by the image collection device by the movable platform. In one embodiment, if there is no error in the encoded image frame sent by the movable platform, the encoded image frame sent by the movable platform is decoded to obtain the target image frame, and the target image frame is displayed.

Step S203: Decode the error-tolerant frame sent by the encoding end to obtain a target image frame, and determine the first feature information of the target image frame.

Exemplarily, the first feature information of the target image frame includes at least one of the following: grayscale feature information of the target image frame, pixel information of a local area of the target image frame, hash string of the target image frame, Brightness information and contrast information, image structure information of the target image frame, feature information extracted from the target image frame based on a preset feature extraction network.

Step S204: Obtain second feature information of the reconstructed frame corresponding to the error-tolerant frame sent by the encoder.

Exemplarily, the movable platform inserts the second characteristic information into the SEI field of the fault-tolerant frame, so that the fault-tolerant frame carries the second characteristic information, and sends the fault-tolerant frame carrying the second characteristic information to the remote control device; The second feature information of the reconstructed frame is obtained from the SEI field of the fault-tolerant frame sent by the platform. By inserting the second feature information into the SEI field of the fault-tolerant frame, the second feature information can be transmitted to the remote control device together with the code stream, so that the remote control device can obtain the second feature information in time.

In one embodiment, the movable platform sends the fault-tolerant frame to the remote control device through the broadband wireless channel, and sends the second characteristic information of the reconstructed frame to the remote control device through the narrowband wireless channel; the remote control device obtains the first information sent by the movable platform through the narrowband wireless channel. Two characteristic information. Due to the high reliability of the narrowband wireless channel, in order to ensure that the remote control device can reliably receive the second feature information, the mobile platform uses the narrowband wireless channel to send the second feature information to the remote control device, thereby improving the ability of the remote control device to receive the second feature information. reliability of information.

Step S205, if the deviation between the first feature information and the second feature information is less than or equal to the preset threshold, send correct feedback information to the encoding end, where the correct feedback information is used to indicate the encoding The endpoint stops transmitting error-tolerant frames.

After acquiring the first feature information and the second feature information, determine the deviation between the first feature information and the second feature information, and if the deviation between the first feature information and the second feature information is less than or equal to the preset threshold, the remote control device sends The movable platform sends correct feedback information, and the correct feedback information is used to instruct the movable platform to stop transmitting fault-tolerant frames. After receiving the correct feedback information, the movable platform stops sending fault-tolerant frames to the remote control device, then performs inter-frame encoding on the initial image frame to obtain the inter-frame image frame, and sends the inter-frame image frame to the remote control device.

In one embodiment, if the deviation between the first feature information and the second feature information is greater than a preset threshold, the mobile platform continues to send error feedback information to instruct the mobile platform to continue to transmit error-tolerant frames. When the movable platform receives the error feedback information, it encodes the initial image frame to obtain the error-tolerant frame, and decodes the error-tolerant frame to obtain the reconstructed frame corresponding to the error-tolerant frame; the movable platform determines the second feature of the reconstructed frame information, and continue to send the error-tolerant frame and the second feature information to the remote control device until the correct feedback information sent by the remote control device is received.

The image transmission method provided by the above embodiment, when an error occurs in the encoded image frame sent by the encoding end, instructs the encoding end to transmit the error-tolerant frame, and then decodes the error-tolerant frame sent by the encoding end to obtain the target image frame, and determine the target image frame. The first feature information of the image frame, and finally the second feature information of the reconstructed frame corresponding to the error-tolerant frame sent by the encoder is obtained. If the deviation between the first feature information and the second feature information is less than or equal to a preset threshold, the encoder is instructed to stop Send an error-tolerant frame. If the deviation between the first feature information and the second feature is greater than a preset threshold, continue to instruct the encoder to transmit the error-tolerant frame until the deviation between the first feature information and the second feature information is less than or equal to the preset threshold, so that the Timely detection of errors in the image transmission process can also avoid false detection caused by judging whether the image is transmitted correctly based on the correctness of the video stream, which greatly improves the reliability of wireless image transmission.

Please refer to FIG. 4 , which is a schematic flowchart of steps of another image transmission method provided by an embodiment of the present application. The image transmission method can be applied to a decoding end, the decoding end is connected to the encoding end in communication, the encoding end can include a movable platform, and the decoding end can include a remote control device.

As shown in FIG. 4 , the image transmission method includes steps S301 to S305.

Step S301: Obtain historical feedback information sent by the decoding end to the encoding end.

The historical feedback information is the feedback information sent by the remote control device to the movable platform regarding the transmission of the last encoded image frame sent by the movable platform, and the historical feedback information is error feedback information or correct feedback information.

Step S302: If the historical feedback information is error feedback information, determine whether the encoded image frame currently sent by the encoding end is an error-tolerant frame.

If the historical feedback information is error feedback information, that is, the feedback information sent by the remote control device to the movable platform regarding the transmission of the last encoded image frame sent by the movable platform is error feedback information, it is determined that the movable platform currently sends Whether the encoded image frame is an error-tolerant frame, if the encoded image frame currently sent by the movable platform is an error-tolerant frame, a feature comparison needs to be performed to determine whether the remote control device can recover the error based on the error-tolerant frame. If the movable platform The encoded image frame currently sent is not an error-tolerant frame, so it is not necessary to perform feature comparison to determine whether the remote control device can recover errors based on the error-tolerant frame, thereby reducing the amount of computation.

Step S303: If the encoded image frame currently sent by the encoding end is an error-tolerant frame, decode the error-tolerant frame to obtain a target image frame, and determine the first feature information of the target image frame.

If the encoded image frame currently sent by the movable platform is an error-tolerant frame, the error-tolerant frame is decoded to obtain a target image frame, and the first feature information of the target image frame is determined. The first feature information includes at least one of the following: grayscale feature information of the target image frame, pixel information of a local area of the target image frame, hash string of the target image frame, brightness information and contrast information of the target image frame, The image structure information of the target image frame, and the feature information extracted from the target image frame based on the preset feature extraction network.

Step S304: Obtain second feature information of the reconstructed frame corresponding to the error-tolerant frame sent by the encoder.

Exemplarily, the movable platform inserts the second characteristic information into the SEI field of the fault-tolerant frame, so that the fault-tolerant frame carries the second characteristic information, and sends the fault-tolerant frame carrying the second characteristic information to the remote control device; The second feature information of the reconstructed frame is obtained from the SEI field of the fault-tolerant frame sent by the platform. By inserting the second feature information into the SEI field of the fault-tolerant frame, the second feature information can be transmitted to the remote control device together with the code stream, so that the remote control device can obtain the second feature information in time.

Step S305, if the deviation between the first feature information and the second feature information is less than or equal to the preset threshold, send correct feedback information to the encoding end, where the correct feedback information is used to indicate the encoding The endpoint stops transmitting error-tolerant frames.

After acquiring the first feature information and the second feature information, determine the deviation between the first feature information and the second feature information, and if the deviation between the first feature information and the second feature information is less than or equal to the preset threshold, the remote control device sends The movable platform sends correct feedback information to instruct the movable platform to stop transmitting fault-tolerant frames. After receiving the correct feedback information, the movable platform stops sending fault-tolerant frames to the remote control device, then performs inter-frame encoding on the initial image frame to obtain the inter-frame image frame, and sends the inter-frame image frame to the remote control device.

In one embodiment, if the deviation between the first feature information and the second feature information is greater than a preset threshold, the mobile platform continues to send error feedback information to instruct the mobile platform to continue to transmit error-tolerant frames. When the movable platform receives the error feedback information, it encodes the initial image frame to obtain the error-tolerant frame, and decodes the error-tolerant frame to obtain the reconstructed frame corresponding to the error-tolerant frame; the movable platform determines the second feature of the reconstructed frame information, and continue to send the error-tolerant frame and the second feature information to the remote control device until the correct feedback information sent by the remote control device is received.

In one embodiment, when the remote control device is powered on, that is, when the remote control device is powered on, the historical feedback information is set as error feedback information, and the movable platform is powered on, that is, when the mobile platform is powered on, the movable platform sends the information to the remote control. The device sends an error-tolerant frame, therefore, it can be determined that the historical feedback information is error feedback information, and the encoded image frame currently sent by the movable platform is an error-tolerant frame, then the remote control device decodes the received error-tolerant frame to obtain the target image frame, and determine the first feature information of the target image frame, and then obtain the second feature information of the reconstructed frame sent by the movable platform, if the deviation between the first feature information and the second feature information is greater than a preset threshold, the remote control device instructs the movable platform Continue to send error-tolerant frames until the deviation between the first feature information and the second feature information is less than or equal to a preset threshold.

If the subsequent remote control device finds that there is a problem with the image frame transmitted by the movable platform to the remote control device based on the cyclic check code, in this case, the historical feedback information is correct feedback information, and the encoded image frame currently sent by the movable platform is not correct. For error-tolerant frames, the remote control device will not perform feature information comparison, but decode the encoded image frame to obtain the target image frame, and display the target image frame, which may cause the image to be blurred. After a period of time, when the remote The device requests a fault-tolerant frame from the movable platform, and after acquiring the fault-tolerant frame sent by the movable platform, the remote control device decodes the received fault-tolerant frame, obtains the target image frame, determines the first feature information of the target image frame, and then obtains the The second feature information of the reconstructed frame sent by the movable platform, if the deviation between the first feature information and the second feature information is greater than the preset threshold, the remote control device instructs the movable platform to continue to send the fault-tolerant frame, until the first feature information and the second feature information. The deviation of the feature information is less than or equal to a preset threshold.

The image transmission method provided by the above-mentioned embodiment determines the first feature information of the target image frame and obtains the second feature information of the reconstructed frame corresponding to the error-tolerant frame sent by the encoding end. If the deviation of the first feature information and the second feature information is less than or equal to the preset threshold, the encoding end is instructed to stop sending error-tolerant frames, and if the deviation between the first feature information and the second feature is greater than the preset threshold, the encoding end is instructed to transmit error-tolerant frames until the first feature information and the second feature The deviation of the information is less than or equal to the preset threshold, so that the errors in the image transmission process can be found in time, and the false detection caused by judging whether the image is transmitted correctly based on the correctness of the video stream can be avoided, which greatly improves the wireless Reliability of image transmission.

Please refer to FIG. 5. FIG. 5 is a schematic flowchart of steps of another image transmission method provided by an embodiment of the present application. The image transmission method is applied to an encoding end, the encoding end is connected to the decoding end in communication, the encoding end may include a movable platform, and the decoding end may include a remote control device.

As shown in FIG. 5 , the image transmission method includes steps S401 to S402.

Step S401: Obtain second feature information of the reconstructed frame corresponding to the encoded image frame.

Exemplarily, the movable platform encodes the initial image frame to obtain the encoded image frame, and decodes the encoded image frame to obtain the reconstructed frame corresponding to the encoded image frame; and determines the corresponding encoded image frame. The second feature information of the reconstructed frame.

Wherein, the second feature information includes grayscale feature information of the reconstructed frame, pixel information of a local area of the reconstructed frame, hash string of the reconstructed frame, brightness information and contrast information of the reconstructed frame, image structure information of the reconstructed frame, pre- Suppose the feature extraction network extracts the feature information from the reconstructed frame, the preset feature extraction network is a pre-trained neural network, and the local area of the reconstructed frame can include the central area, the upper left area, the lower left area, and the lower right corner of the reconstructed frame. area, upper right area.

In one embodiment, the method of determining the second feature information of the reconstructed frame may be: performing hash processing on the reconstructed frame based on a preset hash algorithm to obtain a hash string of the reconstructed frame, or performing discrete cosine transform on the reconstructed frame. processing, and hashing the reconstructed frame after the discrete cosine transform processing based on a preset hash algorithm to obtain a hash string of the reconstructed frame, wherein the preset hash algorithm includes an average hash algorithm, a difference value hash Either algorithm or perceptual hashing algorithm.

In one embodiment, the feature information extracted from the reconstructed frame based on the preset feature extraction network may include texture features of the reconstructed frame, and the training method of the neural network may be: acquiring a plurality of sample data, the sample data including image frames, The texture feature corresponding to the image frame; the neural network is iteratively trained according to a plurality of sample data, until the neural network converges or the number of training times reaches a preset number of training times, and a preset feature extraction network is obtained. The neural network may include a convolutional neural network, a recurrent neural network, a deep neural network, and a recurrent convolutional neural network.

In one embodiment, the method of determining the first feature information of the reconstructed frame may be: according to the gray value of each pixel in the reconstructed frame, determine the target of each image row, image column or image block in the reconstructed frame. Grayscale value; according to the target grayscale value of each image row, image column or image block in the reconstructed frame, determine the grayscale feature information of the reconstructed frame. Wherein, the target gray value includes an average gray value or a total gray value determined based on the gray value of each pixel point in the image row, image column or image block.

Exemplarily, the target grayscale values of each image row, image column or image block in the reconstructed frame are arranged to obtain a grayscale feature vector whose dimension is the image height, image width or the number of image blocks of the reconstructed frame; The grayscale feature vector is normalized, and the normalized grayscale feature vector is rounded to obtain the grayscale feature information of the reconstructed frame.

For example, if the reconstructed frame includes _N2 image lines, the target grayscale values of the _N2 image lines are arranged to obtain an _N2 -dimensional grayscale feature vector. For another example, if the target image frame includes M ₂ image columns, then by arranging the target grayscale values of the M ₂ image columns, an M ₁ -dimensional grayscale feature vector can be obtained. For another example, if the target image frame includes P ₂ image blocks, the target gray values of the P ₂ image blocks are arranged to obtain a P ₂ -dimensional gray feature vector, where N ₂ , M ₂ and P ₂ are all positive Integer.

Step S402: Send the encoded image frame and the second feature information to the decoding end, so that the decoding end decodes the encoded image frame to obtain a target image frame, and determines the target image frame. The first feature information of the image frame, if the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the encoding end, and the error feedback information is used to indicate The encoding end transmits error-tolerant frames.

In one embodiment, the movable platform inserts the second feature information into the Supplemental Enhancement Information (SEI) field of the encoded image frame, so that the encoded image frame carries the second feature information, and sends the information to the remote control. The device sends the encoded image frame carrying the second feature information; the remote control device obtains the encoded image frame sent by the movable platform, and obtains the first image frame of the reconstructed frame from the SEI field of the encoded image frame sent by the movable platform. Two characteristic information. By inserting the second feature information into the SEI field of the encoded image frame, the second feature information can be transmitted to the remote control device together with the code stream, so that the remote control device can obtain the second feature information in time.

In one embodiment, the movable platform sends the encoded image frames to the remote control device through a broadband wireless channel between the movable platform and the remote control device, and sends the encoded image frames to the remote control device through a narrowband wireless channel between the movable platform and the remote control device. Reconstructing the second feature information of the frame; the remote control device obtains the encoded image frame sent by the movable platform through the broadband wireless channel, and obtains the second feature information sent by the movable platform through the narrowband wireless channel. Due to the high reliability of the narrowband wireless channel, in order to ensure that the remote control device can reliably receive the second feature information, the mobile platform uses the narrowband wireless channel to send the second feature information to the remote control device, thereby improving the ability of the remote control device to receive the second feature information. reliability of information.

In one embodiment, the movable platform sends the encoded image frame to the remote control device, and sends the second characteristic information to the remote control device at preset time intervals; the remote control device obtains the encoded image frame from the movable platform and sends the encoded image frame and the movable platform Second feature information sent at preset time intervals. The preset time may be set based on an actual situation, which is not specifically limited in this embodiment of the present application. Since the wireless communication link between the movable platform and the remote control device may be blocked or interrupted, in order to ensure that the remote control device can reliably receive the second feature information, the movable platform periodically sends the second feature information , thereby improving the reliability of the remote control device receiving the second feature information.

The image transmission method provided by the above-mentioned embodiment obtains the second characteristic information of the reconstructed frame corresponding to the encoded image frame, and sends the encoded image frame and the second characteristic information to the decoding end, so that the decoding end can compare the encoded image. The frame is decoded to obtain the target image frame, and the first characteristic information of the target image frame is determined. If the deviation between the first characteristic information and the second characteristic information is greater than the preset threshold, the error feedback information is sent to the encoding end to Instruct the encoder to transmit error-tolerant frames, so that errors in the image transmission process can be accurately and timely detected, and at the same time instruct the encoder to transmit error-tolerant frames. reliability.

Please refer to FIG. 6. FIG. 6 is a schematic flowchart of steps of another image transmission method provided by an embodiment of the present application. The image transmission method is applied to an encoding end, the encoding end is connected to the decoding end in communication, the encoding end may include a movable platform, and the decoding end may include a remote control device.

As shown in FIG. 6 , the image transmission method may include steps S501 to S503.

Step S501: Acquire feedback information sent by the decoding end.

Wherein, the feedback information includes error feedback information or correct feedback information, the error feedback information is used to instruct the movable platform to transmit the error-tolerant frame to the remote control device, and the correct feedback information is used to instruct the movable platform to stop transmitting the error-tolerant frame to the remote control device.

Step S502: If the feedback information is error feedback information, acquire second feature information of the reconstructed frame corresponding to the error-tolerant frame.

If the feedback information sent by the remote control device is error feedback information, the movable platform encodes the initial image frame to obtain the error-tolerant frame, and decodes the error-tolerant frame to obtain the reconstructed frame corresponding to the error-tolerant frame; the movable platform determines the reconstruction frame. The second characteristic information of the frame; if the feedback information is correct feedback information, stop sending the error-tolerant frame and the second characteristic information of the reconstructed frame to the remote control device.

Wherein, the second characteristic information is determined by the movable platform when it receives the error feedback information sent by the remote control device, and the error-tolerant frame and the encoded image frame are obtained by encoding the same initial image frame by the movable platform, or, The fault-tolerant frame is obtained by encoding the initial image frame currently collected by the image collection device on the movable platform.

Step S503, sending the error-tolerant frame and the second feature information to the decoding end, so that the decoding end decodes the error-tolerant frame to obtain a target image frame, and determines the first feature of the target image frame information, and if the deviation between the first feature information and the second feature information is greater than a preset threshold, the error feedback information is sent to the encoding end.

Exemplarily, the movable platform sends an error-tolerant frame to the remote control device through a broadband wireless channel, and sends the second feature information of the reconstructed frame to the remote control device through a narrowband wireless channel; the remote control device acquires the second feature sent by the movable platform through the narrowband wireless channel. information. Due to the high reliability of the narrowband wireless channel, in order to ensure that the remote control device can reliably receive the second feature information, the mobile platform uses the narrowband wireless channel to send the second feature information to the remote control device, thereby improving the ability of the remote control device to receive the second feature information. reliability of information.

Exemplarily, the movable platform inserts the second characteristic information into the SEI field of the fault-tolerant frame, so that the fault-tolerant frame carries the second characteristic information, and sends the fault-tolerant frame carrying the second characteristic information to the remote control device. By inserting the second feature information into the SEI field of the fault-tolerant frame, the second feature information can be transmitted to the remote control device together with the code stream, so that the remote control device can obtain the second feature information in time.

After the remote control device obtains the first feature information and the second feature information, the deviation between the first feature information and the second feature information is determined, and if the deviation between the first feature information and the second feature information is less than or equal to the preset threshold, the remote control The device sends correct feedback information to the movable platform, where the correct feedback information is used to instruct the movable platform to stop transmitting the fault-tolerant frame. After receiving the correct feedback information, the movable platform stops sending fault-tolerant frames to the remote control device, then performs inter-frame encoding on the initial image frame to obtain the inter-frame image frame, and sends the inter-frame image frame to the remote control device.

In one embodiment, when the mobile platform is powered on, that is, when the mobile platform is powered on, the mobile platform sends an error-tolerant frame to the remote control device by default, and the feedback information sent by the default remote control device is error feedback information. Therefore, the mobile platform Encode the initial image frame to obtain a fault-tolerant frame, and decode the fault-tolerant frame to obtain a reconstructed frame corresponding to the fault-tolerant frame; determine the second feature information of the reconstructed frame, and send the fault-tolerant frame and the second feature information to the remote control device , the remote control device decodes the received error-tolerant frame, obtains the target image frame, determines the first feature information of the target image frame, and then obtains the second feature information of the reconstructed frame sent by the movable platform. If the deviation of the second feature information is greater than the preset threshold, the remote control device instructs the movable platform to continue sending fault-tolerant frames until the deviation between the first feature information and the second feature information is less than or equal to the preset threshold, if the first feature information and the second feature If the deviation of the information is less than or equal to the preset threshold, the movable platform stops sending error-tolerant frames to the remote control device, and sends inter-frame coded frames instead.

In the image transmission method provided by the above-mentioned embodiments, only when the feedback information sent by the decoding end is error feedback information, the encoded image frame and the second characteristic information of the reconstructed frame corresponding to the encoded image frame are sent to the decoding end, so that The decoding end decodes the encoded image frame to obtain the target image frame, and determines the first feature information of the target image frame. If the deviation between the first feature information and the second feature information is greater than the preset threshold, the encoding The end sends error feedback information to instruct the encoder to transmit error-tolerant frames, which can reduce the amount of computation, and can accurately and timely detect errors in the image transmission process, so that the decoder can recover errors based on the error-tolerant frames retransmitted by the encoder. The reliability of wireless image transmission.

Please refer to FIG. 7. FIG. 7 is a schematic flowchart of steps of another image transmission method provided by an embodiment of the present application. The image transmission method is applied to an encoding end, the encoding end is connected to the decoding end in communication, the encoding end may include a movable platform, and the decoding end may include a remote control device.

As shown in FIG. 7 , the image transmission method may include steps S601 to S603.

Step 601: Send the encoded image frame to the decoding end, so that the decoding end can decode the encoded image frame to obtain a target image frame and determine the first feature information of the target image frame.

The movable platform encodes the initial image frame, obtains the encoded image frame, and sends the encoded image frame to the remote control device. After decoding, the target image frame is obtained, and then the remote control device determines the first characteristic information of the target image frame, and sends the first characteristic information to the movable platform.

Step 602: Obtain the second feature information of the reconstructed frame corresponding to the encoded image frame, and obtain the first feature information sent by the decoding end;

The movable platform obtains the first feature information sent by the remote control device, and obtains the second feature information of the reconstructed frame corresponding to the encoded image frame. The first feature information includes at least one of the following: grayscale feature information of the target image frame, pixel information of a local area of the target image frame, hash string of the target image frame, brightness information and contrast information of the target image frame, The image structure information of the target image frame, the feature information extracted from the target image frame based on the preset feature extraction network, the second feature information includes the grayscale feature information of the reconstructed frame, the pixel information of the local area of the reconstructed frame, the The hash string, the brightness information and contrast information of the reconstructed frame, the image structure information of the reconstructed frame, and the feature information extracted from the reconstructed frame based on the preset feature extraction network.

Step 603: If the deviation between the first feature information and the second feature information is greater than a preset threshold, send an error-tolerant frame to the decoding end for the decoding end to decode the error-tolerant frame.

After the movable platform obtains the first characteristic information and the second characteristic information, it determines the deviation between the first characteristic information and the second characteristic information, and if the deviation between the first characteristic information and the second characteristic information is greater than the preset threshold An error-tolerant frame is sent, and the error-tolerant frame is decoded by the remote control device; if the deviation between the first feature information and the second feature information is less than or equal to a preset threshold, an inter-frame coded frame is sent to the remote control device.

In one embodiment, the feedback information sent by the decoding end is obtained; if the feedback information is error feedback information, the error-tolerant frame is sent to the decoding end, so that the decoding end decodes the error-tolerant frame to obtain the target image frame and determines the size of the target image frame. the first feature information; obtain the second feature information of the reconstructed frame corresponding to the fault-tolerant frame; if the deviation between the first feature information and the second feature information is greater than the preset threshold, continue to send the fault-tolerant frame to the decoding end; If the deviation of the second feature information is less than or equal to the preset threshold, stop sending the error-tolerant frame to the decoding end; perform inter-frame encoding on the initial image frame to obtain the inter-frame image frame, and send the inter-frame image frame to the decoding end. By comparing the first feature information with the second feature information only when the feedback information is erroneous feedback information, the amount of computation can be reduced.

In the image transmission method provided by the above embodiment, the encoded image frame is sent to the decoding end, then the second characteristic information of the reconstructed frame corresponding to the encoded image frame is obtained, and the first characteristic information sent by the decoding end is obtained. If the deviation between the first feature information and the second feature information is greater than the preset threshold, an error-tolerant frame is sent to the decoding end, so that the decoding end can decode the error-tolerant frame, so that errors in the image transmission process can be accurately and timely discovered, and at the same time the encoding end The error-tolerant frame is sent to the decoding end, and the decoding end can recover the error based on the error-tolerant frame retransmitted by the coding end, thereby improving the reliability of wireless image transmission.

Please refer to FIG. 8 . FIG. 8 is a schematic structural block diagram of a remote control device provided by an embodiment of the present application.

As shown in FIG. 8 , the remote control device 700 includes a processor 710 and a memory 720, and the processor 710 and the memory 720 are connected through a bus 730, such as an I2C (Inter-integrated Circuit) bus.

Specifically, the processor 710 may be a micro-controller unit (Micro-controller Unit, MCU), a central processing unit (Central Processing Unit, CPU) or a digital signal processor (Digital Signal Processor, DSP) or the like.

Specifically, the memory 720 may be a Flash chip, a read-only memory (ROM, Read-Only Memory) magnetic disk, an optical disk, a U disk, or a removable hard disk, and the like.

Wherein, the processor 710 is configured to run the computer program stored in the memory 720, and implement the following steps when executing the computer program:

Decoding the encoded image frame sent by the movable platform to obtain a target image frame, and determining the first feature information of the target image frame;

acquiring second feature information of the reconstructed frame sent by the movable platform, wherein the reconstructed frame is obtained by decoding the encoded image frame by the movable platform;

If the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the movable platform, where the error feedback information is used to instruct the movable platform to transmit an error-tolerant frame.

In one embodiment, the processor is further configured to implement the following steps:

If the deviation between the first feature information and the second feature information is less than or equal to the preset threshold, send correct feedback information to the movable platform, where the correct feedback information is used to indicate the movable platform Stop transmitting error-tolerant frames.

In an embodiment, the first feature information includes at least one of the following:

Grayscale feature information of the target image frame;

pixel information of the local area of the target image frame;

the hash string of the target image frame;

brightness information and contrast information of the target image frame;

image structure information of the target image frame;

Feature information extracted from the target image frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network.

In one embodiment, when the processor determines the first feature information of the target image frame, the processor is configured to:

According to the gray value of each pixel in the target image frame, determine the target gray value of each image row, image column or image block in the target image frame;

The grayscale feature information of the target image frame is determined according to the target grayscale value of each image row, image column or image block in the target image frame.

In one embodiment, the target gray value includes an average gray value or a total gray value determined based on the gray value of each pixel in the image row, image column or image block.

In one embodiment, when the processor determines the grayscale feature information of the target image frame according to the target grayscale value of each image row, image column or image block in the target image frame, use To achieve:

Arrange the target grayscale values of each of the image rows, image columns or image blocks to obtain a grayscale feature vector whose dimension is the image height, image width or the number of image blocks of the target image frame;

The grayscale feature vector is normalized, and the normalized grayscale feature vector is rounded to obtain grayscale feature information of the target image frame.

In an embodiment, the second feature information includes at least one of the following:

grayscale feature information of the reconstructed frame;

pixel information of the local area of the reconstructed frame;

the hash string of the reconstructed frame;

Brightness information and contrast information of the reconstructed frame;

image structure information of the reconstructed frame;

Feature information extracted from the reconstructed frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network.

In one embodiment, when the processor acquires the second feature information of the reconstructed frame sent by the movable platform, the processor is configured to:

The second feature information of the reconstructed frame is acquired from the supplementary enhancement information SEI field of the encoded image frame sent by the movable platform.

In one embodiment, when the processor acquires the second feature information of the reconstructed frame sent by the movable platform, the processor is configured to:

Acquire second feature information of the reconstructed frame sent by the movable platform through the narrowband wireless channel.

In one embodiment, the error-tolerant frame is an image frame obtained by coding based on inter-frame coding and intra-frame coding, or the error-tolerant frame is an image frame obtained by coding based on intra-frame coding.

In one embodiment, the second characteristic information is determined by the movable platform upon receiving the error feedback information sent by the remote control device.

In one embodiment, when the processor acquires the second feature information of the reconstructed frame sent by the movable platform, the processor is configured to:

Acquire second feature information of reconstructed frames sent by the movable platform at preset time intervals.

In one embodiment, before the processor decodes the encoded image frame sent by the movable platform to obtain the target image frame, the processor is further configured to:

determining whether there is an error in the encoded image frame sent by the movable platform;

If there is an error in the encoded image frame sent by the movable platform, send error feedback information to the movable platform, where the error feedback information is used to instruct the movable platform to transmit an error-tolerant frame;

When implementing the decoding of the encoded image frame sent by the movable platform to obtain the target image frame, the processor is used to implement:

Decoding the fault-tolerant frame sent by the movable platform to obtain the target image frame;

When obtaining the second feature information of the reconstructed frame sent by the movable platform, the processor is configured to:

Acquire second feature information of the reconstructed frame corresponding to the error-tolerant frame sent by the movable platform.

In one embodiment, the error-tolerant frame and the encoded image frame are obtained by encoding the same initial image frame by the movable platform.

In one embodiment, the error-tolerant frame is obtained by encoding the initial image frame currently collected by the image collection device by the movable platform.

In one embodiment, the processor is further configured to implement the following steps:

If there is no error in the encoded image frame sent by the movable platform, the encoded image frame sent by the movable platform is decoded to obtain a target image frame, and the target image frame is displayed.

In one embodiment, when the processor determines whether an error occurs in the encoded image frame sent by the movable platform, the processor is configured to:

Carrying out a cyclic redundancy check on the encoded image frame sent by the movable platform;

If the encoded image frame sent by the movable platform fails the cyclic redundancy check, it is determined that an error occurs in the encoded image frame sent by the movable platform.

In one embodiment, before the processor decodes the encoded image frame sent by the movable platform to obtain the target image frame, the processor is further configured to:

obtaining historical feedback information sent by the remote control device to the movable platform;

When implementing the decoding of the encoded image frame sent by the movable platform to obtain the target image frame, the processor is used to implement:

If the historical feedback information is error feedback information, then determine whether the encoded image frame currently sent by the movable platform is an error-tolerant frame;

If the encoded image frame currently sent by the movable platform is an error-tolerant frame, decoding the error-tolerant frame to obtain a target image frame;

When obtaining the second feature information of the reconstructed frame sent by the movable platform, the processor is configured to:

Acquire second feature information of the reconstructed frame corresponding to the error-tolerant frame sent by the movable platform.

In one embodiment, the processor is further configured to implement the following steps:

If the historical feedback information is correct feedback information, the encoded image frame sent by the movable platform is decoded to obtain a target image frame, and the target image frame is displayed.

In one embodiment, the processor is further configured to implement the following steps:

When the remote control device is powered on, the historical feedback information is set as error feedback information.

It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of the description, for the specific working process of the remote control device described above, reference may be made to the corresponding process in the foregoing image transmission method embodiment, which is not repeated here. Repeat.

Please refer to FIG. 9. FIG. 9 is a schematic structural block diagram of a movable platform provided by an embodiment of the present application.

As shown in FIG. 9 , the movable platform 800 includes a processor 810 and a memory 820, and the processor 810 and the memory 820 are connected by a bus 830, such as an I2C (Inter-integrated Circuit) bus.

Specifically, the processor 810 may be a micro-controller unit (Micro-controller Unit, MCU), a central processing unit (Central Processing Unit, CPU), or a digital signal processor (Digital Signal Processor, DSP) or the like.

Specifically, the memory 820 may be a Flash chip, a read-only memory (ROM, Read-Only Memory) magnetic disk, an optical disk, a U disk, or a removable hard disk, and the like.

Wherein, the processor 810 is configured to run the computer program stored in the memory 820, and implement the following steps when executing the computer program:

acquiring second feature information of the reconstructed frame corresponding to the encoded image frame;

Send the encoded image frame and the second feature information to the remote control device, so that the remote control device can decode the encoded image frame to obtain a target image frame, and determine the target image frame. First feature information, if the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the movable platform, where the error feedback information is used to indicate the The removable platform transmits fault-tolerant frames.

In an embodiment, the first feature information includes at least one of the following:

Grayscale feature information of the target image frame;

pixel information of the local area of the target image frame;

the hash string of the target image frame;

brightness information and contrast information of the target image frame;

image structure information of the target image frame;

Feature information extracted from the target image frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network.

In an embodiment, the second feature information includes at least one of the following:

grayscale feature information of the reconstructed frame;

pixel information of the local area of the reconstructed frame;

the hash string of the reconstructed frame;

Brightness information and contrast information of the reconstructed frame;

image structure information of the reconstructed frame;

Feature information extracted from the reconstructed frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network.

In one embodiment, when the processor acquires the second feature information of the reconstructed frame corresponding to the encoded image frame, the processor is configured to:

According to the gray value of each pixel in the reconstructed frame, determine the target gray value of each image row, image column or image block in the reconstructed frame;

The grayscale feature of the reconstructed frame is determined according to the target grayscale value of each image row, image column or image block in the reconstructed frame.

In one embodiment, the target gray value includes an average gray value or a total gray value determined based on the gray value of each pixel in the image row, image column or image block.

In one embodiment, when determining the grayscale feature of the reconstructed frame according to the target grayscale value of each image row, image column or image block in the reconstructed frame, the processor is configured to:

Arrange the target gray value of each of the image rows, image columns or image blocks to obtain a grayscale feature vector whose dimension is the image height, image width or the number of image blocks of the reconstructed frame;

The grayscale feature vector is normalized, and the normalized grayscale feature vector is rounded to obtain the grayscale feature of the reconstructed frame.

In one embodiment, when the processor sends the encoded image frame and the second feature information to the remote control device, the processor is configured to:

inserting the second feature information into the supplementary enhancement information SEI field of the encoded image frame, so that the encoded image frame carries the second feature information;

Sending the encoded image frame carrying the second feature information to the remote control device.

In one embodiment, when the processor sends the encoded image frame and the second feature information to the remote control device, the processor is configured to:

sending the encoded image frame to the remote control device through a broadband wireless channel between the movable platform and the remote control device;

The second characteristic information is transmitted over a narrowband wireless channel between the movable platform and the remote control device.

In one embodiment, when the processor sends the encoded image frame and the second feature information to the remote control device, the processor is configured to:

The encoded image frame is sent to the remote control device, and the second feature information is sent to the remote control device at preset time intervals.

In one embodiment, before obtaining the second feature information of the reconstructed frame corresponding to the encoded image frame, the processor is further configured to:

obtain feedback information sent by the remote control device;

When obtaining the second feature information of the reconstructed frame corresponding to the encoded image frame, the processor is used to realize:

If the feedback information is error feedback information, acquiring second feature information of the reconstructed frame corresponding to the fault-tolerant frame;

When implementing sending the encoded image frame and the second feature information to the remote control device, the processor is configured to implement:

Send the error-tolerant frame and the second feature information to the remote control device, so that the remote control device can decode the error-tolerant frame to obtain a target image frame, and determine the first feature information of the target image frame, if If the deviation between the first feature information and the second feature information is greater than a preset threshold, the error feedback information is sent to the movable platform.

In one embodiment, the error-tolerant frame and the encoded image frame are obtained by encoding the same initial image frame by the movable platform.

In one embodiment, the error-tolerant frame is obtained by encoding the initial image frame currently collected by the image collection device by the movable platform.

In another embodiment, the processor 810 is configured to run a computer program stored in the memory 820, and implement the following steps when executing the computer program:

Send the encoded image frame to the remote control device, so that the remote control device decodes the encoded image frame to obtain a target image frame and determine the first feature information of the target image frame;

acquiring the second feature information of the reconstructed frame corresponding to the encoded image frame, and acquiring the first feature information sent by the remote control device;

If the deviation between the first feature information and the second feature information is greater than a preset threshold, an error-tolerant frame is sent to the remote control device for the remote control device to decode the error-tolerant frame.

In one embodiment, before the processor sends the encoded image frame to the remote control device, the processor is further configured to:

obtain feedback information sent by the remote control device;

When the processor implements sending the encoded image frame to the remote control device, the processor is configured to implement:

If the feedback information is error feedback information, sending an error-tolerant frame to the remote control device, so that the remote control device can decode the error-tolerant frame to obtain a target image frame and determine the first feature information of the target image frame;

When obtaining the second feature information of the reconstructed frame corresponding to the encoded image frame, the processor is configured to:

obtaining the second feature information of the reconstructed frame corresponding to the fault-tolerant frame;

When the processor implements sending a fault-tolerant frame to the remote control device, the processor is configured to implement:

If the deviation between the first feature information and the second feature information is greater than a preset threshold, continue to send an error-tolerant frame to the remote control device.

In one embodiment, the processor is further configured to implement the following steps:

If the deviation between the first feature information and the second feature information is less than or equal to a preset threshold, stop sending error-tolerant frames to the remote control device;

Perform inter-frame coding on the initial image frame to obtain the inter-frame image frame, and send the inter-frame image frame to the remote control device.

It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of the description, for the specific working process of the movable platform described above, reference may be made to the corresponding process in the foregoing image transmission method embodiment, which is not described here. Repeat.

Please refer to FIG. 10. FIG. 10 is a schematic structural block diagram of a control system provided by an embodiment of the present application. As shown in FIG. 10 , the control system 900 includes a remote control device 910 and a movable platform 920 . The remote control device 910 is connected in communication with the movable platform 920 , and the remote control device 910 is used to control the movable platform 920 .

It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of the description, for the specific working process of the control system described above, reference may be made to the corresponding process in the foregoing image transmission method embodiment, which is not repeated here. Repeat.

Embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, the computer program includes program instructions, and the processor executes the program instructions, so as to realize the provision of the above embodiments. The steps of the image transmission method.

The computer-readable storage medium may be an internal storage unit of the remote control device or the movable platform described in any of the foregoing embodiments, such as a hard disk or memory of the remote control device or the movable platform. The computer-readable storage medium may also be an external storage device of the remote control device or the removable platform, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC) equipped on the remote control device or the removable platform. , Secure Digital (Secure Digital, SD) card, flash memory card (Flash Card) and so on.

It should be understood that the terms used in the specification of the present application herein are for the purpose of describing particular embodiments only and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural unless the context clearly dictates otherwise.

It will also be understood that, as used in this specification and the appended claims, the term "and/or" refers to and including any and all possible combinations of one or more of the associated listed items.

The above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalents within the technical scope disclosed in the present application. Modifications or substitutions shall be covered by the protection scope of this application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (72)

An image transmission method, characterized in that it is applied to a decoding end, and the decoding end is communicatively connected to an encoding end, and the method includes: Decoding the encoded image frame sent by the encoding end to obtain a target image frame, and determining the first feature information of the target image frame; Obtain the second feature information of the reconstructed frame sent by the encoding end, wherein the reconstructed frame is obtained by the encoding end decoding the encoded image frame; If the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the encoder, where the error feedback information is used to instruct the encoder to transmit an error-tolerant frame. The image transmission method according to claim 1, wherein the method further comprises: If the deviation between the first feature information and the second feature information is less than or equal to the preset threshold, send correct feedback information to the encoder, where the correct feedback information is used to instruct the encoder to stop transmission Error-tolerant frames. The image transmission method according to claim 1, wherein the first feature information includes at least one of the following: Grayscale feature information of the target image frame; pixel information of the local area of the target image frame; the hash string of the target image frame; brightness information and contrast information of the target image frame; image structure information of the target image frame; Feature information extracted from the target image frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network. The image transmission method according to claim 1, wherein the determining the first feature information of the target image frame comprises: According to the gray value of each pixel in the target image frame, determine the target gray value of each image row, image column or image block in the target image frame; The grayscale feature information of the target image frame is determined according to the target grayscale value of each image row, image column or image block in the target image frame. The image transmission method according to claim 4, wherein the target gray value comprises an average gray value or a gray value determined based on the gray value of each pixel in the image row, image column or image block. Total gray value. The image transmission method according to claim 4, wherein the grayscale value of the target image frame is determined according to the target grayscale value of each image row, image column or image block in the target image frame Characteristic information, including: Arrange the target grayscale values of each of the image rows, image columns or image blocks to obtain a grayscale feature vector whose dimension is the image height, image width or the number of image blocks of the target image frame; The grayscale feature vector is normalized, and the normalized grayscale feature vector is rounded to obtain grayscale feature information of the target image frame. The image transmission method according to claim 1, wherein the second characteristic information includes at least one of the following: grayscale feature information of the reconstructed frame; pixel information of the local area of the reconstructed frame; the hash string of the reconstructed frame; Brightness information and contrast information of the reconstructed frame; image structure information of the reconstructed frame; Feature information extracted from the reconstructed frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network. The image transmission method according to claim 1, wherein the acquiring the second feature information of the reconstructed frame sent by the encoding end comprises: The second feature information of the reconstructed frame is acquired from the supplementary enhancement information SEI field of the encoded image frame sent by the encoding end. The image transmission method according to claim 1, wherein the acquiring the second feature information of the reconstructed frame sent by the encoding end comprises: Acquire second feature information of the reconstructed frame sent by the encoding end through the narrowband wireless channel. The image transmission method according to claim 1, wherein the error-tolerant frame is an image frame obtained by coding based on inter-frame coding and intra-frame coding, or the error-tolerant frame is an image obtained by coding based on intra-frame coding frame. The image transmission method according to claim 1, wherein the second characteristic information is determined by the encoding end when receiving error feedback information sent by the decoding end. The image transmission method according to claim 1, wherein the acquiring the second feature information of the reconstructed frame sent by the encoding end comprises: Acquire second feature information of reconstructed frames sent by the encoding end at preset time intervals. The image transmission method according to any one of claims 1-12, wherein before the encoded image frame sent by the encoding end is decoded to obtain the target image frame, it also includes: Determine whether there is an error in the encoded image frame sent by the encoding end; If there is an error in the encoded image frame sent by the encoder, send error feedback information to the encoder, where the error feedback information is used to instruct the encoder to transmit an error-tolerant frame; The decoding of the encoded image frame sent by the encoding end to obtain the target image frame includes: Decoding the error-tolerant frame sent by the encoding end to obtain the target image frame; The obtaining of the second feature information of the reconstructed frame sent by the encoding end includes: Acquire second feature information of the reconstructed frame corresponding to the error-tolerant frame sent by the encoder. The image transmission method according to claim 13, wherein the error-tolerant frame and the encoded image frame are obtained by encoding the same initial image frame by the encoding end. The image transmission method according to claim 13, wherein the error-tolerant frame is obtained by encoding the initial image frame currently collected by the image collection device by the coding end. The image transmission method according to claim 13, wherein the method further comprises: If there is no error in the encoded image frame sent by the encoding end, the encoded image frame sent by the encoding end is decoded to obtain a target image frame, and the target image frame is displayed. The image transmission method according to claim 13, wherein the determining whether an error occurs in the encoded image frame sent by the encoding end comprises: performing a cyclic redundancy check on the encoded image frame sent by the encoding end; If the encoded image frame sent by the encoding end fails the cyclic redundancy check, it is determined that an error occurs in the encoded image frame sent by the encoding end. The image transmission method according to any one of claims 1-12, wherein before the decoding the encoded image frame sent by the encoding end to obtain the target image frame, the method further comprises: obtaining historical feedback information sent by the decoding end to the encoding end; The decoding of the encoded image frame sent by the encoding end to obtain the target image frame includes: If the historical feedback information is error feedback information, then determine whether the encoded image frame currently sent by the encoding end is an error-tolerant frame; If the encoded image frame currently sent by the encoding end is an error-tolerant frame, decoding the error-tolerant frame to obtain a target image frame; The obtaining of the second feature information of the reconstructed frame sent by the encoding end includes: Acquire second feature information of the reconstructed frame corresponding to the error-tolerant frame sent by the encoder. The image transmission method according to claim 18, wherein the method further comprises: If the historical feedback information is correct feedback information, the encoded image frame sent by the encoding end is decoded to obtain a target image frame, and the target image frame is displayed. The image transmission method according to claim 18, wherein the method further comprises: When the decoding end is powered on, the historical feedback information is set as error feedback information. An image transmission method, characterized in that it is applied to an encoding end, and the encoding end is communicatively connected to a decoding end, and the method comprises: acquiring second feature information of the reconstructed frame corresponding to the encoded image frame; Send the encoded image frame and the second feature information to the decoding end, so that the decoding end decodes the encoded image frame to obtain the target image frame, and determines the target image frame. First feature information, if the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the encoding end, where the error feedback information is used to indicate the encoding The terminal transmits error-tolerant frames. The image transmission method according to claim 21, wherein the first characteristic information includes at least one of the following: Grayscale feature information of the target image frame; pixel information of the local area of the target image frame; the hash string of the target image frame; Brightness information and contrast information of the target image frame; image structure information of the target image frame; Feature information extracted from the target image frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network. The image transmission method according to claim 21, wherein the second characteristic information comprises at least one of the following: grayscale feature information of the reconstructed frame; pixel information of the local area of the reconstructed frame; the hash string of the reconstructed frame; Brightness information and contrast information of the reconstructed frame; image structure information of the reconstructed frame; Feature information extracted from the reconstructed frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network. image transmission method according to claim 21, is characterized in that, the second characteristic information of the reconstruction frame corresponding to the described image frame obtained after encoding, comprises: According to the gray value of each pixel in the reconstructed frame, determine the target gray value of each image row, image column or image block in the reconstructed frame; The grayscale feature of the reconstructed frame is determined according to the target grayscale value of each image row, image column or image block in the reconstructed frame. The image transmission method according to claim 24, wherein the target gray value comprises an average gray value determined based on the gray value of each pixel in the image row, image column or image block or Total gray value. The image transmission method according to claim 24, wherein the grayscale feature of the reconstructed frame is determined according to the target grayscale value of each image row, image column or image block in the reconstructed frame, include: arranging the target grayscale values of each of the image rows, image columns or image blocks to obtain a grayscale feature vector whose dimension is the image height, image width or the number of image blocks of the reconstructed frame; The grayscale feature vector is normalized, and the normalized grayscale feature vector is rounded to obtain the grayscale feature of the reconstructed frame. The image transmission method according to claim 21, wherein the sending the encoded image frame and the second feature information to the decoding end comprises: inserting the second feature information into the supplementary enhancement information SEI field of the encoded image frame, so that the encoded image frame carries the second feature information; Send the encoded image frame carrying the second feature information to the decoding end. The image transmission method according to claim 21, wherein the sending the encoded image frame and the second feature information to the decoding end comprises: Send the encoded image frame to the decoding end through the broadband wireless channel between the encoding end and the decoding end; The second feature information is sent through a narrowband wireless channel between the encoding end and the decoding end. The image transmission method according to claim 21, wherein the sending the encoded image frame and the second feature information to the decoding end comprises: The encoded image frame is sent to the decoding end, and the second feature information is sent to the decoding end at preset time intervals. The image transmission method according to any one of claims 21-29, wherein before acquiring the second feature information of the reconstructed frame corresponding to the encoded image frame, the method further comprises: obtaining the feedback information sent by the decoding end; The obtaining of the second feature information of the reconstructed frame corresponding to the encoded image frame includes: If the feedback information is error feedback information, acquiring second feature information of the reconstructed frame corresponding to the fault-tolerant frame; The sending of the encoded image frame and the second feature information to the decoding end includes: Send the error-tolerant frame and the second feature information to the decoding end, so that the decoding end decodes the error-tolerant frame to obtain a target image frame, and determines the first feature information of the target image frame, if If the deviation between the first feature information and the second feature information is greater than a preset threshold, the error feedback information is sent to the encoding end. The image transmission method according to claim 30, wherein the error-tolerant frame and the encoded image frame are obtained by encoding the same initial image frame by the encoding end. The image transmission method according to claim 30, wherein the error-tolerant frame is obtained by the encoding end encoding the initial image frame currently collected by the image collection device. An image transmission method, characterized in that it is applied to an encoding end, and the encoding end is communicatively connected to a decoding end, and the method comprises: Send the encoded image frame to the decoding end, so that the decoding end decodes the encoded image frame to obtain the target image frame and determine the first feature information of the target image frame; obtaining the second feature information of the reconstructed frame corresponding to the encoded image frame, and obtaining the first feature information sent by the decoding end; If the deviation between the first feature information and the second feature information is greater than a preset threshold, an error-tolerant frame is sent to the decoding end for the decoding end to decode the error-tolerant frame. The image transmission method according to claim 33, wherein before sending the encoded image frame to the decoding end, the method further comprises: obtaining the feedback information sent by the decoding end; The sending of the encoded image frame to the decoding end includes: If the feedback information is error feedback information, sending an error-tolerant frame to the decoding end, so that the decoding end decodes the error-tolerant frame to obtain a target image frame and determines the first feature information of the target image frame; The acquiring the second feature information of the reconstructed frame corresponding to the encoded image frame includes: obtaining the second feature information of the reconstructed frame corresponding to the fault-tolerant frame; The sending an error-tolerant frame to the decoding end includes: If the deviation between the first feature information and the second feature information is greater than a preset threshold, continue to send an error-tolerant frame to the decoding end. The image transmission method according to claim 34, wherein the method further comprises: If the deviation between the first feature information and the second feature information is less than or equal to a preset threshold, stop sending error-tolerant frames to the decoding end; Perform inter-frame coding on the initial image frame to obtain an inter-frame image frame, and send the inter-frame image frame to the decoding end. A remote control device, characterized in that the remote control device is connected in communication with a movable platform, and the remote control device includes a memory and a processor; the memory is used to store computer programs; The processor is configured to execute the computer program and implement the following steps when executing the computer program: Decoding the encoded image frame sent by the movable platform to obtain a target image frame, and determining the first feature information of the target image frame; acquiring second feature information of the reconstructed frame sent by the movable platform, wherein the reconstructed frame is obtained by decoding the encoded image frame by the movable platform; If the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the movable platform, where the error feedback information is used to instruct the movable platform to transmit an error-tolerant frame. The remote control device of claim 36, wherein the processor is further configured to implement the following steps: If the deviation between the first feature information and the second feature information is less than or equal to the preset threshold, send correct feedback information to the movable platform, where the correct feedback information is used to indicate the movable platform Stop transmitting error-tolerant frames. The remote control device according to claim 36, wherein the first characteristic information comprises at least one of the following: Grayscale feature information of the target image frame; pixel information of the local area of the target image frame; the hash string of the target image frame; brightness information and contrast information of the target image frame; image structure information of the target image frame; Feature information extracted from the target image frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network. remote control device according to claim 36, is characterized in that, when described processor realizes determining the first characteristic information of described target image frame, is used for realizing: According to the gray value of each pixel in the target image frame, determine the target gray value of each image row, image column or image block in the target image frame; The grayscale feature information of the target image frame is determined according to the target grayscale value of each image row, image column or image block in the target image frame. The remote control device according to claim 39, wherein the target gray value comprises an average gray value or a total gray value determined based on the gray value of each pixel in the image row, image column or image block. grayscale value. The remote control device according to claim 39, wherein the processor determines the target image frame according to the target gray value of each image row, image column or image block in the target image frame. When the grayscale feature information of , is used to realize: arranging the target grayscale values of each of the image rows, image columns or image blocks to obtain a grayscale feature vector whose dimension is the image height, image width or the number of image blocks of the target image frame; The grayscale feature vector is normalized, and the normalized grayscale feature vector is rounded to obtain grayscale feature information of the target image frame. The remote control device according to claim 36, wherein the second characteristic information includes at least one of the following: grayscale feature information of the reconstructed frame; pixel information of the local area of the reconstructed frame; the hash string of the reconstructed frame; Brightness information and contrast information of the reconstructed frame; image structure information of the reconstructed frame; Feature information extracted from the reconstructed frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network. The remote control device according to claim 36, wherein when the processor acquires the second feature information of the reconstructed frame sent by the movable platform, the processor is configured to: The second feature information of the reconstructed frame is acquired from the supplementary enhancement information SEI field of the encoded image frame sent by the movable platform. The remote control device according to claim 36, wherein when the processor acquires the second feature information of the reconstructed frame sent by the movable platform, the processor is configured to: Acquire second feature information of the reconstructed frame sent by the movable platform through the narrowband wireless channel. The remote control device according to claim 36, wherein the error-tolerant frame is an image frame obtained by coding based on inter-frame coding and intra-frame coding, or the error-tolerant frame is an image frame obtained by coding based on intra-frame coding . The remote control device according to claim 36, wherein the second characteristic information is determined by the movable platform when receiving error feedback information sent by the remote control device. The remote control device according to claim 36, wherein when the processor acquires the second feature information of the reconstructed frame sent by the movable platform, the processor is configured to: Acquire second feature information of reconstructed frames sent by the movable platform at preset time intervals. The remote control device according to any one of claims 36-47, characterized in that before the processor decodes the encoded image frame sent by the movable platform to obtain the target image frame, the processor further uses To achieve: determining whether there is an error in the encoded image frame sent by the movable platform; If there is an error in the encoded image frame sent by the movable platform, send error feedback information to the movable platform, where the error feedback information is used to instruct the movable platform to transmit an error-tolerant frame; When implementing the decoding of the encoded image frame sent by the movable platform to obtain the target image frame, the processor is used to implement: Decoding the fault-tolerant frame sent by the movable platform to obtain the target image frame; When obtaining the second feature information of the reconstructed frame sent by the movable platform, the processor is configured to: Acquire second feature information of the reconstructed frame corresponding to the error-tolerant frame sent by the movable platform. The remote control device according to claim 48, wherein the error-tolerant frame and the encoded image frame are obtained by encoding the same initial image frame by the movable platform. The remote control device according to claim 48, wherein the error-tolerant frame is obtained by encoding the initial image frame currently collected by the image collecting device by the movable platform. The remote control device of claim 48, wherein the processor is further configured to implement the following steps: If there is no error in the encoded image frame sent by the movable platform, the encoded image frame sent by the movable platform is decoded to obtain a target image frame, and the target image frame is displayed. The remote control device according to claim 48, wherein when the processor determines whether an error occurs in the encoded image frame sent by the movable platform, the processor is configured to: Carrying out a cyclic redundancy check on the encoded image frame sent by the movable platform; If the encoded image frame sent by the movable platform fails the cyclic redundancy check, it is determined that an error occurs in the encoded image frame sent by the movable platform. The remote control device according to any one of claims 36-47, characterized in that before the processor decodes the encoded image frame sent by the movable platform to obtain the target image frame, the processor further uses To achieve: obtaining historical feedback information sent by the remote control device to the movable platform; When implementing the decoding of the encoded image frame sent by the movable platform to obtain the target image frame, the processor is used to implement: If the historical feedback information is error feedback information, then determine whether the encoded image frame currently sent by the movable platform is an error-tolerant frame; If the encoded image frame currently sent by the movable platform is an error-tolerant frame, decoding the error-tolerant frame to obtain a target image frame; When obtaining the second feature information of the reconstructed frame sent by the movable platform, the processor is configured to: Acquire second feature information of the reconstructed frame corresponding to the error-tolerant frame sent by the movable platform. The remote control device of claim 53, wherein the processor is further configured to implement the following steps: If the historical feedback information is correct feedback information, the encoded image frame sent by the movable platform is decoded to obtain a target image frame, and the target image frame is displayed. The remote control device of claim 53, wherein the processor is further configured to implement the following steps: When the remote control device is powered on, the historical feedback information is set as error feedback information. A movable platform, characterized in that the movable platform communicates with a remote control device, and the movable platform includes a memory and a processor; the memory is used to store computer programs; The processor is configured to execute the computer program and implement the following steps when executing the computer program: obtaining second feature information of the reconstructed frame corresponding to the encoded image frame; Send the encoded image frame and the second feature information to the remote control device, so that the remote control device can decode the encoded image frame to obtain a target image frame, and determine the target image frame. First feature information, if the deviation between the first feature information and the second feature information is greater than a preset threshold, send error feedback information to the movable platform, where the error feedback information is used to indicate the The removable platform transmits fault-tolerant frames. The movable platform according to claim 56, wherein the first characteristic information includes at least one of the following: Grayscale feature information of the target image frame; pixel information of the local area of the target image frame; the hash string of the target image frame; brightness information and contrast information of the target image frame; image structure information of the target image frame; Feature information extracted from the target image frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network. The movable platform according to claim 56, wherein the second characteristic information includes at least one of the following: grayscale feature information of the reconstructed frame; pixel information of the local area of the reconstructed frame; the hash string of the reconstructed frame; Brightness information and contrast information of the reconstructed frame; image structure information of the reconstructed frame; Feature information extracted from the reconstructed frame based on a preset feature extraction network, where the preset feature extraction network is a pre-trained neural network. The movable platform according to claim 56, wherein when the processor obtains the second feature information of the reconstructed frame corresponding to the encoded image frame, the processor is configured to: According to the gray value of each pixel in the reconstructed frame, determine the target gray value of each image row, image column or image block in the reconstructed frame; The grayscale feature of the reconstructed frame is determined according to the target grayscale value of each image row, image column or image block in the reconstructed frame. The movable platform according to claim 59, wherein the target gray value comprises an average gray value determined based on the gray value of each pixel in the image row, image column or image block or Total gray value. The movable platform according to claim 59, wherein the processor determines the value of the reconstructed frame according to the target gray value of each image row, image column or image block in the reconstructed frame. When grayscale features are used, it is used to achieve: Arrange the target gray value of each of the image rows, image columns or image blocks to obtain a grayscale feature vector whose dimension is the image height, image width or the number of image blocks of the reconstructed frame; The grayscale feature vector is normalized, and the normalized grayscale feature vector is rounded to obtain the grayscale feature of the reconstructed frame. The movable platform according to claim 56, wherein when the processor implements sending the encoded image frame and the second feature information to the remote control device, the processor is configured to implement: inserting the second feature information into the supplementary enhancement information SEI field of the encoded image frame, so that the encoded image frame carries the second feature information; Sending the encoded image frame carrying the second feature information to the remote control device. The movable platform according to claim 56, wherein when the processor implements sending the encoded image frame and the second feature information to the remote control device, the processor is configured to implement: Send the encoded image frame to the remote control device through a broadband wireless channel between the movable platform and the remote control device; The second characteristic information is transmitted over a narrowband wireless channel between the movable platform and the remote control device. The movable platform according to claim 57, wherein when the processor sends the encoded image frame and the second feature information to the remote control device, the processor is configured to: The encoded image frame is sent to the remote control device, and the second feature information is sent to the remote control device at preset time intervals. The movable platform according to any one of claims 56 to 64, wherein, before obtaining the second feature information of the reconstructed frame corresponding to the encoded image frame, the processor is further configured to: obtain feedback information sent by the remote control device; When obtaining the second feature information of the reconstructed frame corresponding to the encoded image frame, the processor is used to achieve: If the feedback information is error feedback information, acquiring second feature information of the reconstructed frame corresponding to the error-tolerant frame; When implementing sending the encoded image frame and the second feature information to the remote control device, the processor is configured to implement: Send the error-tolerant frame and the second feature information to the remote control device, so that the remote control device can decode the error-tolerant frame to obtain a target image frame, and determine the first feature information of the target image frame, if If the deviation between the first feature information and the second feature information is greater than a preset threshold, the error feedback information is sent to the movable platform. The movable platform according to claim 65, wherein the error-tolerant frame and the encoded image frame are obtained by encoding the same initial image frame by the movable platform. The movable platform according to claim 65, wherein the error-tolerant frame is obtained by encoding the initial image frame currently collected by the image collection device by the movable platform. A movable platform, characterized in that the movable platform communicates with a remote control device, and the movable platform includes a memory and a processor; the memory is used to store computer programs; The processor is configured to execute the computer program and implement the following steps when executing the computer program: Send the encoded image frame to the remote control device, so that the remote control device decodes the encoded image frame to obtain a target image frame and determine the first feature information of the target image frame; acquiring the second feature information of the reconstructed frame corresponding to the encoded image frame, and acquiring the first feature information sent by the remote control device; If the deviation between the first feature information and the second feature information is greater than a preset threshold, an error-tolerant frame is sent to the remote control device for the remote control device to decode the error-tolerant frame. The movable platform according to claim 68, wherein before the processor transmits the encoded image frame to the remote control device, the processor is further configured to: obtain feedback information sent by the remote control device; When the processor transmits the encoded image frame to the remote control device, the processor is configured to: If the feedback information is error feedback information, sending an error-tolerant frame to the remote control device, so that the remote control device can decode the error-tolerant frame to obtain a target image frame and determine the first feature information of the target image frame; When obtaining the second feature information of the reconstructed frame corresponding to the encoded image frame, the processor is configured to: obtaining the second feature information of the reconstructed frame corresponding to the fault-tolerant frame; When the processor implements sending a fault-tolerant frame to the remote control device, the processor is configured to implement: If the deviation between the first feature information and the second feature information is greater than a preset threshold, continue to send an error-tolerant frame to the remote control device. The movable platform of claim 69, wherein the processor is further configured to implement the following steps: If the deviation between the first feature information and the second feature information is less than or equal to a preset threshold, stop sending error-tolerant frames to the remote control device; Perform inter-frame coding on the initial image frame to obtain the inter-frame image frame, and send the inter-frame image frame to the remote control device. A control system, characterized in that the control system comprises the remote control device according to any one of claims 36-55 and the movable platform according to any one of claims 56-70, the remote control device and the movable platform according to any one of claims 56-70. The mobile platform is communicatively connected, and the remote control device is used to control the movable platform. A computer-readable storage medium, characterized in that, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor can realize the implementation of any one of claims 1-35. The steps of the image transmission method.

Images