CN117036418A - Image processing method, device and equipment - Google Patents

Abstract

The application provides an image processing method, an image processing device and image processing equipment, and relates to an image processing technology, wherein the method comprises the following steps: identifying the multi-frame images to obtain a parameter list of each image; the parameter list comprises parameter information of an object to be identified in the image. For two adjacent frame images in the multi-frame images, determining a target to be identified in the next frame image in the two adjacent frame images according to a parameter list of the previous frame image in the two adjacent frame images and a parameter list of the next frame image in the two adjacent frame images; the target to be identified in the next frame of images in the two adjacent frames is the target to be identified corresponding to the target to be identified in the previous frame of images in the two adjacent frames. Generating an identification code corresponding to the target to be identified in the last frame of image in the multi-frame images, and generating target information corresponding to the identification code. The application solves the technical problem that the behavior information corresponding to the identification code cannot be acquired due to the large tracking difficulty of the identification code of the user.

Description

Image processing methods, devices and equipment

Technical field

The present application relates to image processing technology, and in particular, to an image processing method, device and equipment.

Background technique

Currently, in order to understand students' classroom behavior during class, each student needs to be tracked and identified.

In the existing technology, when tracking and identifying each student, the student's classroom behavior is identified through target detection and action recognition technology, and the number of actions produced by each student in the entire class is calculated based on the target tracking algorithm.

However, in the existing technology, when the number of actions produced by each student in the entire class is calculated based on the target tracking algorithm, students will have disturbing behaviors in the classroom, such as lowering their heads, blocking between students, etc., resulting in the inability to accurately identify The student's identification code is obtained, resulting in the inability to obtain the student's classroom behavior.

Contents of the invention

This application provides an image processing method, device and equipment to solve the technical problem of being unable to obtain behavioral information corresponding to the identification code due to difficulty in tracking the user's identification code.

In the first aspect, this application provides an image processing method, including:

Perform recognition processing on multiple frame images to obtain a parameter list for each image; wherein the parameter list includes parameter information of the target to be recognized in the image;

For two adjacent frames of images in the multi-frame images, the parameter list of the previous frame of the image in the two adjacent frames of images and the parameter list of the subsequent frame of the image in the two adjacent frames of images are determined. The target to be recognized in the latter frame of the two adjacent images is the same as the target in the previous frame of the two adjacent images. The target to be recognized corresponding to the target to be recognized in the image;

Generate an identification code corresponding to the target to be identified in the last frame of the multi-frame image, and generate target information corresponding to the identification code; wherein the identification code represents the identity of the target to be identified, and the target The information represents the number of times the target to be identified has issued a preset action.

Further, for two adjacent frames of images in the multi-frame images, according to the parameter list of the previous frame of the image in the two adjacent frames of images, and the parameter list of the subsequent frame of the image in the two adjacent frames of images. , determining the target to be recognized in the latter frame of the two adjacent images, including:

According to the parameter list of the first frame image in the multi-frame image, determine the Gaussian distribution corresponding to the target to be recognized in the first frame image; wherein the Gaussian distribution represents the position information of the target to be recognized;

For two adjacent frames in the multi-frame images, the parameter list of the latter image in the two adjacent frames corresponds to the target to be recognized in the previous frame of the two adjacent images. The Gaussian distribution is used for probability calculation processing to obtain the probability result information corresponding to the target to be identified in the latter frame of the two adjacent images; wherein the probability result information is used to indicate the latter frame of the two adjacent images. Gaussian distribution corresponding to the target to be identified in ;

According to the probability result information, the target to be recognized in the latter frame of the two adjacent frames of images is determined.

Further, based on the probability result information, determining the target to be recognized in the latter frame of the two adjacent frames of images includes:

If it is determined that the probability result information is to obtain a likelihood probability value, wherein the likelihood probability value represents the parameter information of the target to be identified in the latter frame of two adjacent frames of images, respectively, and the likelihood probability value of the previous two adjacent frames of images. The likelihood probability value between multiple Gaussian distributions in one frame of image is used to determine the maximum likelihood probability value corresponding to the target to be identified; wherein, the maximum likelihood probability value is used to indicate the latter of two adjacent frames of images. Gaussian distribution corresponding to the target to be recognized in the frame image;

According to the maximum likelihood probability value corresponding to the target to be recognized, the target to be recognized in the latter frame of the two adjacent frames of images is determined.

Further, the parameter information corresponding to the target to be recognized includes the face width value; the method further includes:

According to the preset first constraint information, determine the distance between the target to be identified in the latter frame of the two adjacent images and the location of the Gaussian distribution corresponding to the target to be identified; wherein, the first constraint The information represents the location information corresponding to the target to be identified;

If it is determined that the distance is greater than the face width value of the target to be recognized, then the Gaussian distribution corresponding to the target to be recognized in the latter image of the two adjacent frames is replaced with the Gaussian distribution of the target in the previous frame of the two adjacent images. The Gaussian distribution corresponding to the target to be recognized is obtained, and the Gaussian distribution corresponding to the target to be recognized in the latter frame of the updated two adjacent frames of images is obtained;

According to the Gaussian distribution corresponding to the target to be recognized in the latter image of the two adjacent frames after updating, the target to be recognized in the latter image of the two adjacent frames is determined.

Further, based on the probability result information, determining the target to be recognized in the latter frame of the two adjacent frames of images includes:

If it is determined that the probability result information is that the likelihood probability value has not been obtained, wherein the likelihood probability value represents the parameter information of the target to be identified in the latter frame of the two adjacent frames of images, respectively, and the likelihood probability value of the two adjacent frames of images. Based on the likelihood probability values between multiple Gaussian distributions in the previous frame image, the target to be identified is determined to be a new target;

According to the parameter information corresponding to the new target, determine the Gaussian distribution corresponding to the new target in the latter frame of the two adjacent frames;

According to the Gaussian distribution corresponding to the new target in the latter frame of the two adjacent frames of images, the target to be recognized in the latter frame of the two adjacent images is determined.

Further, based on the probability result information, determining the target to be recognized in the latter frame of the two adjacent frames of images includes:

If it is determined that the probability result information is that parameter information corresponding to the Gaussian distribution has not been obtained, then it is determined that the target to be identified corresponding to the Gaussian distribution is in a disappearing state;

Determine a preset number of targets to be identified adjacent to the location of the Gaussian distribution, and parameter information corresponding to the target to be identified; wherein the Gaussian distribution includes the first face area, and the parameter information corresponding to the target to be identified Including the second face area;

According to the preset second constraint information, among the preset number of second face areas, determine the target face area that is equal to the first face area; wherein the second constraint information represents the to-be-recognized The pixel ratio of the target’s face area;

According to the parameter information corresponding to the target face area, determine the Gaussian distribution corresponding to the target to be recognized in the latter image of the two adjacent frames;

According to the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent frames of images, the target to be recognized in the latter frame of the two adjacent images is determined.

Further, perform recognition processing on multiple frames of images to obtain a parameter list for each image, including:

According to the preset depth target detection network, multiple frame images are recognized and processed respectively to obtain a parameter list of each image; wherein, the depth target detection network is used to indicate target identification information, and the parameter list includes images Parameter information of the target to be identified in .

Further, the parameter information corresponding to the target to be recognized includes face information and human body information, wherein the face information includes the abscissa coordinate of the face center, the ordinate of the face center, the face width value, the face height value, and the face height value. Face area, the human body information includes the human body center abscissa, the human body center ordinate, the human body width value, the human body height value, and the human body area.

In a second aspect, this application provides an image processing device, including:

A recognition unit, configured to perform recognition processing on multiple frame images to obtain a parameter list for each image; wherein the parameter list includes parameter information of the target to be recognized in the image;

The first determination unit is configured to determine, for two adjacent frames of the multi-frame images, according to the parameter list of the previous frame of the image in the two adjacent frames of images, and the subsequent frame of the two adjacent frames of images. The parameter list of the image determines the target to be recognized in the latter frame of the two adjacent images; wherein the target to be recognized in the latter frame of the two adjacent images is the target that is the same as the adjacent one. The target to be recognized corresponding to the target to be recognized in the previous frame of the two images;

A first generation unit configured to generate an identification code corresponding to the target to be identified in the last frame of the multi-frame image;

The second generation unit is configured to generate target information corresponding to the identification code; wherein the identification code represents the identity of the target to be identified, and the target information represents the number of times the target to be identified has issued a preset action.

Further, the first determining unit includes:

The first determination module is configured to determine the Gaussian distribution corresponding to the target to be identified in the first frame image according to the parameter list of the first frame image in the multi-frame image; wherein the Gaussian distribution represents the position information of the target to be identified;

A calculation module configured to, for two adjacent frames of images in the multi-frame images, compare the parameter list of the latter frame of the image in the two adjacent frames of images with the parameter list of the previous frame of the image in the two adjacent frames of images. The Gaussian distribution corresponding to the target to be identified is subjected to probability calculation processing to obtain the probability result information corresponding to the target to be identified in the latter image of the two adjacent frames; wherein the probability result information is used to indicate the two adjacent frames of images. Gaussian distribution corresponding to the target to be recognized in the next frame of image;

The second determination module is configured to determine the target to be recognized in the latter frame of the two adjacent frames of images based on the probability result information.

Further, the second determination module includes:

The first determination sub-module is used to obtain a likelihood probability value if it is determined that the probability result information is, where the likelihood probability value represents the parameter information of the target to be identified in the latter frame of the two adjacent frames of images, respectively. and the likelihood probability value between the multiple Gaussian distributions in the previous frame image in the two adjacent frames of images, then the maximum likelihood probability value corresponding to the target to be identified is determined; wherein, the maximum likelihood probability value is used to indicate Gaussian distribution corresponding to the target to be recognized in the latter image of two adjacent frames;

The second determination sub-module is used to determine the target to be recognized in the latter frame of the two adjacent frames of images based on the maximum likelihood probability value corresponding to the target to be recognized.

Further, the parameter information corresponding to the target to be recognized includes the face width value; the method further includes:

The second determination unit is used to determine the distance between the target to be identified in the latter frame of the two adjacent images and the position of the Gaussian distribution corresponding to the target to be identified based on the preset first constraint information; wherein , the first constraint information represents the position information corresponding to the target to be identified;

An update unit, configured to replace the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent frames of images with the Gaussian distribution of the target to be identified in the latter frame of the two adjacent frames of images if it is determined that the distance is greater than the face width value of the target to be identified. The Gaussian distribution corresponding to the target to be recognized in the previous frame of the image is obtained, and the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent frames of images is obtained;

The third determination unit is configured to determine the target to be recognized in the latter image of the two adjacent frames according to the updated Gaussian distribution corresponding to the target to be recognized in the latter image of the two adjacent frames.

Further, the second determination module includes:

The third determination sub-module is used to determine that the probability result information is that the likelihood probability value has not been obtained, wherein the likelihood probability value represents the parameter information of the target to be identified in the latter frame of the two adjacent frames of images, If the likelihood probability values between the multiple Gaussian distributions in the previous frame of the two adjacent images are determined, the target to be identified is determined to be a new target;

The fourth determination sub-module is used to determine the Gaussian distribution corresponding to the new target in the latter frame of the two adjacent frames of images based on the parameter information corresponding to the new target;

The fifth determination sub-module is used to determine the target to be recognized in the latter frame of the two adjacent frames of images according to the Gaussian distribution corresponding to the new target in the latter frame of the two adjacent frames of images.

Further, the second determination module includes:

The sixth determination sub-module is used to determine that the target to be identified corresponding to the Gaussian distribution is in a disappearing state if it is determined that the probability result information is not obtained.

The seventh determination sub-module is used to determine a preset number of targets to be identified adjacent to the location of the Gaussian distribution, and parameter information corresponding to the targets to be identified; wherein the Gaussian distribution includes the first face area, so The parameter information corresponding to the target to be recognized includes the second face area;

The eighth determination sub-module is used to determine a target face area equal to the first face area among a preset number of second face areas according to the preset second constraint information; wherein, the The second constraint information represents the pixel ratio of the face area of the target to be recognized;

The ninth determination sub-module is used to determine the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent frames of images based on the parameter information corresponding to the target face area;

The tenth determination sub-module is used to determine the target to be recognized in the latter frame of the two adjacent frames of images according to the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent images.

Further, the identification unit is specifically used for:

According to the preset depth target detection network, multiple frame images are recognized and processed respectively to obtain a parameter list of each image; wherein, the depth target detection network is used to indicate target identification information, and the parameter list includes images Parameter information of the target to be identified in .

Further, the parameter information corresponding to the target to be recognized includes face information and human body information, wherein the face information includes the abscissa coordinate of the face center, the ordinate of the face center, the face width value, the face height value, and the face height value. Face area, the human body information includes the human body center abscissa, the human body center ordinate, the human body width value, the human body height value, and the human body area.

In a third aspect, the present application provides an electronic device, including a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, the first aspect is implemented. the method described.

In a fourth aspect, the present application provides a computer-readable storage medium in which computer-executable instructions are stored, and when executed by a processor, the computer-executable instructions are used to implement the method described in the first aspect.

In a fifth aspect, the present application provides a computer program product, including a computer program that implements the method described in the first aspect when executed by a processor.

This application provides an image processing method, device and equipment that performs recognition processing on multiple frames of images to obtain a parameter list for each image; wherein the parameter list includes parameter information of the target to be recognized in the image. For two adjacent frames of images in a multi-frame image, determine the latter of the two adjacent frames of images based on the parameter list of the previous frame of the image and the parameter list of the latter frame of the two adjacent images. The target to be recognized in one frame of image; wherein, the target to be recognized in the latter frame of two adjacent frames of images is the target to be recognized corresponding to the target to be recognized in the previous frame of the two adjacent images. Generate an identification code corresponding to the target to be identified in the last frame of the multi-frame image, and generate target information corresponding to the identification code; wherein, the identification code represents the identity of the target to be identified, and the target information represents the preset issued by the target to be identified. The number of actions. In this solution, the image includes multiple targets to be recognized, and the multi-frame images are recognized and processed to obtain a parameter list of each image. The parameter list includes parameter information of the target to be recognized in the image. Then, based on the two adjacent frames in the multi-frame images, it is possible to determine the target to be identified in the latter frame of the two adjacent images and the target to be identified in the previous frame of the two adjacent images. The target is the same target to be identified. Then the determined target to be recognized in the latter frame of the two adjacent frames is used as the target to be recognized in the previous frame of the next set of two adjacent frames, and combined with the next set of two adjacent frames. The parameter list in the next frame of images is used to determine the target to be recognized in the next frame of the next two adjacent frames, and so on, until the target to be recognized in the last frame of the multi-frame image is determined. Finally, an identification code corresponding to the target to be identified in the last frame of the multi-frame image is generated, and target information corresponding to the identification code is generated, and then the number of times the target to be identified has issued a preset action is displayed based on the target information. Therefore, the same target to be identified in multiple frames of images can be determined, and the identification code corresponding to the target to be identified can be generated, which greatly improves the stability of tracking the target to be identified and solves the problem of difficulty in tracking the user's identification code. Technical problem of not being able to obtain the behavioral information corresponding to the identification code.

Description of the drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure.

Figure 1 is a schematic flowchart of an image processing method provided by an embodiment of the present application;

Figure 2 is a schematic flow chart of another image processing method provided by an embodiment of the present application;

Figure 3 is a schematic structural diagram of an image processing device provided by an embodiment of the present application;

Figure 4 is a schematic structural diagram of another image processing device provided by an embodiment of the present application;

Figure 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application;

Figure 6 is a block diagram of an electronic device provided by an embodiment of the present application.

Specific embodiments of the present disclosure have been shown through the above-mentioned drawings and will be described in more detail below. These drawings and written description are not intended to limit the scope of the disclosed concepts in any way, but rather to illustrate the concepts of the present disclosure to those skilled in the art with reference to the specific embodiments.

Detailed ways

Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present disclosure.

In one example, in order to understand students' classroom behavior during class, each student needs to be tracked and identified. In the existing technology, when tracking and identifying each student, the student's classroom behavior is identified through target detection and action recognition technology, and the number of actions produced by each student in the entire class is calculated based on the target tracking algorithm. However, in the existing technology, when the number of actions produced by each student in the entire class is calculated based on the target tracking algorithm, students will have disturbing behaviors in the classroom, such as lowering their heads, blocking between students, etc., resulting in the inability to accurately identify The student's identification code is obtained, resulting in the inability to obtain the student's classroom behavior.

This application provides an image processing method, device and equipment, aiming to solve the above technical problems of the prior art.

The technical solution of the present application and how the technical solution of the present application solves the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

Figure 1 is a schematic flowchart of an image processing method provided by an embodiment of the present application. As shown in Figure 1, the method includes:

101. Perform recognition processing on multiple frame images to obtain a parameter list of each image; where the parameter list includes parameter information of the target to be recognized in the image.

For example, the execution subject of this embodiment may be an electronic device, or a terminal device, or an image processing device or device, or other device or device that can execute this embodiment, without limitation. In this embodiment, the execution subject is an electronic device.

First, multiple frames of images need to be obtained, and then the multi-frame images need to be recognized and processed. Images can be captured, or images can be obtained from memory; or images transmitted by other devices can be received. Then the multi-frame images can be recognized and processed to obtain a parameter list of each image. The parameter list includes parameter information of the target to be recognized in the image. For example, the recognition process includes processing through a preset depth target monitoring network; this application The technical solution is a multi-target tracking algorithm. In addition to the technical solution of this application, the multi-target tracking algorithm also includes the SORT algorithm and the DeepSORT algorithm.

For example, based on the preset deep target monitoring network, the electronic device performs recognition processing on multiple frames of images, detects the human face in each frame of image, and obtains the bounding box (bbox) coordinates and Width and height, each frame outputs a parameter list bbox. Each parameter list has 10 dimensional features, which are: face center abscissa x, face center ordinate y, face bbox width value, face bbox height value, face bbox area, human body center abscissa x, human body center ordinate y, human bbox width value, human bbox height value, and human bbox area.

102. For two adjacent frame images in the multi-frame image, determine the two adjacent frame images according to the parameter list of the previous frame image in the two adjacent frame images and the parameter list of the subsequent frame image in the two adjacent frame images. The target to be recognized in the latter frame of the image; where the target to be recognized in the latter frame of the two adjacent images is the target to be recognized corresponding to the target to be recognized in the previous frame of the two adjacent images. Target.

Exemplarily, first, according to _the parameter list X ₁ of the first frame image in the multi-frame image, the length of the parameter list (10-dimensional) Gaussian distribution G, which is the mean and variance of each target to be identified. For example, the Gaussian distribution G includes: G ₁ , G ₂ , G ₃ ..., where G ₁ represents the Gaussian distribution of the first person, G ₂ represents the Gaussian distribution of the second person, and G ₃ represents the Gaussian distribution of the third person.

Then, for two adjacent frames in the multi-frame images, based on the preset formula, the parameter list of the latter frame of the two adjacent images is compared with the parameter list to be recognized in the previous frame of the two adjacent images. The Gaussian distribution corresponding to the target is subjected to probability calculation processing to obtain the probability result information corresponding to the target to be identified in the latter frame of the two adjacent images. The probability result information is used to indicate the probability result information in the latter frame of the two adjacent images. The preset formula for the Gaussian distribution corresponding to the target to be identified is as follows:

Among them, P _i,j represents the likelihood probability between the i-th target to be recognized in the latter frame of the two adjacent images and the j-th Gaussian distribution in the previous frame of the two adjacent images. _; ^_ _{_}

Finally, based on the probability result information, the target to be recognized in the latter frame of the two adjacent frames of images is determined, where the target to be recognized in the latter frame of the two adjacent frames of images is the same as the target in the latter frame of the two adjacent images. The target to be recognized corresponding to the target to be recognized in the previous frame of the image is to replace the target to be recognized in the latter frame of the two adjacent frames of images with the target to be recognized in the previous frame of the image in the two adjacent frames of images. the corresponding Gaussian distribution.

103. Generate an identification code corresponding to the target to be identified in the last frame of the multi-frame image, and generate target information corresponding to the identification code; wherein, the identification code represents the identity of the target to be identified, and the target information represents the target sent by the target to be identified. The number of preset actions.

For example, the identification code represents the identity of the target to be identified, and the preset action refers to an action pre-stored by the electronic device. For example, the preset action includes raising a hand. The electronic device can generate an identification code corresponding to the target to be identified in the last frame of the multi-frame image based on the pre-stored total number of people and the Gaussian distribution corresponding to the target to be identified in the last frame of the multi-frame image, and generate Target information corresponding to each identification code, the target information represents the number of times the target to be identified has issued a preset action.

For example, if the total number of people is 30, the electronic device can generate the target in the last frame of the multi-frame image based on the Gaussian distributions corresponding to the 30 people and the 30 targets to be recognized in the last frame of the multi-frame image. The identification code corresponding to the target to be identified, for example, the identification code (i.e., ID) of the first target to be identified is 1, the identification code (i.e., ID) of the second target to be identified is 2... and so on, and generates the corresponding identification code for each target. Target information corresponding to an identification code.

In the embodiment of the present application, multiple frame images are subjected to recognition processing to obtain a parameter list of each image; wherein the parameter list includes parameter information of the target to be recognized in the image. For two adjacent frames of images in a multi-frame image, determine the latter of the two adjacent frames of images based on the parameter list of the previous frame of the image and the parameter list of the latter frame of the two adjacent images. The target to be recognized in one frame of image; wherein, the target to be recognized in the latter frame of two adjacent frames of images is the target to be recognized corresponding to the target to be recognized in the previous frame of the two adjacent images. Generate an identification code corresponding to the target to be identified in the last frame of the multi-frame image, and generate target information corresponding to the identification code; wherein, the identification code represents the identity of the target to be identified, and the target information represents the preset issued by the target to be identified. The number of actions. In this solution, the image includes multiple targets to be recognized, and the multi-frame images are recognized and processed to obtain a parameter list of each image. The parameter list includes parameter information of the target to be recognized in the image. Then, based on the two adjacent frames in the multi-frame images, it is possible to determine the target to be identified in the latter frame of the two adjacent images and the target to be identified in the previous frame of the two adjacent images. The target is the same target to be identified. Then the determined target to be recognized in the latter frame of the two adjacent frames is used as the target to be recognized in the previous frame of the next set of two adjacent frames, and combined with the next set of two adjacent frames. The parameter list in the next frame of images is used to determine the target to be recognized in the next frame of the next two adjacent frames, and so on, until the target to be recognized in the last frame of the multi-frame image is determined. Finally, an identification code corresponding to the target to be identified in the last frame of the multi-frame image is generated, and target information corresponding to the identification code is generated, and then the number of times the target to be identified has issued a preset action is displayed based on the target information. Therefore, the same target to be identified in multiple frames of images can be determined, and the identification code corresponding to the target to be identified can be generated, which greatly improves the stability of tracking the target to be identified and solves the problem of difficulty in tracking the user's identification code. Technical problem of not being able to obtain the behavioral information corresponding to the identification code.

Figure 2 is a schematic flow chart of another image processing method provided by an embodiment of the present application. As shown in Figure 2, the method includes:

201. According to the preset depth target detection network, perform recognition processing on multiple frames of images respectively to obtain a parameter list for each image; among them, the depth target detection network is used to indicate target identification information, and the parameter list includes the objects to be recognized in the image. Target parameter information.

In one example, the parameter information corresponding to the target to be recognized includes face information and human body information, where the face information includes the abscissa of the face center, the ordinate of the face center, the face width value, the face height value, and the face Area, human body information includes the horizontal coordinate of the human body center, the vertical coordinate of the human body center, the human body width value, the human body height value, and the human body area.

For example, this step can refer to step 101 in Figure 1 and will not be described again.

202. According to the parameter list of the first frame image in the multi-frame image, determine the Gaussian distribution corresponding to the target to be recognized in the first frame image; wherein the Gaussian distribution represents the position information of the target to be recognized.

Exemplarily, the electronic device initializes N (N=n+10, 10 is a reserved placeholder) parameters according to the parameter list X ₁ of the first frame image in the multi-frame image. The length of the parameter list X ₁ is n. Multidimensional (10-dimensional) Gaussian distribution G, which is the mean and variance of each target to be identified. For example, the Gaussian distribution G includes: G ₁ , G ₂ , G ₃ ..., where G ₁ represents the Gaussian distribution of the first person, G ₂ represents the Gaussian distribution of the second person, and G ₃ represents the Gaussian distribution of the third person.

203. For two adjacent frames in the multi-frame image, compare the parameter list of the latter frame of the two adjacent images with the Gaussian distribution corresponding to the target to be recognized in the previous frame of the two adjacent images. Probability calculation processing is performed to obtain the probability result information corresponding to the target to be identified in the latter frame of the two adjacent images; wherein the probability result information is used to indicate the correspondence of the target to be identified in the latter frame of the two adjacent images. Gaussian distribution.

204. Based on the probability result information, determine the target to be recognized in the latter frame of the two adjacent frames of images.

Step 204 includes three implementation methods:

The first implementation method of step 204: if the probability result information is determined to obtain a likelihood probability value, where the likelihood probability value represents the parameter information of the target to be identified in the latter frame of two adjacent frames of images, respectively with the adjacent The likelihood probability value between the multiple Gaussian distributions in the previous frame image in the two frames of images is used to determine the maximum likelihood probability value corresponding to the target to be identified; where the maximum likelihood probability value is used to indicate two adjacent frames of images. The Gaussian distribution corresponding to the target to be recognized in the latter frame of the image is determined; according to the maximum likelihood probability value corresponding to the target to be recognized, the target to be recognized in the latter frame of the image in the two adjacent frames is determined.

In one example, the parameter information corresponding to the target to be recognized includes a face width value; the first implementation of step 204 also includes: determining, according to the preset first constraint information, the width of the face in the latter frame of the two adjacent frames. The distance between the target to be recognized and the position of the Gaussian distribution corresponding to the target to be recognized; where the first constraint information represents the position information corresponding to the target to be recognized; if it is determined that the distance is greater than the face width value of the target to be recognized, then Replace the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent images with the Gaussian distribution corresponding to the target to be recognized in the previous frame of the two adjacent images to obtain the updated adjacent two frames. The Gaussian distribution corresponding to the target to be recognized in the latter frame of the image; according to the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent frames of images after updating, determine the latter of the two adjacent frames of images. The target to be recognized in the frame image.

The second implementation method of step 204: If it is determined that the probability result information is that the likelihood probability value is not obtained, where the likelihood probability value represents the parameter information of the target to be identified in the latter frame of the two adjacent frames of images, respectively with the corresponding According to the likelihood probability values between multiple Gaussian distributions in the previous frame of the image in the two adjacent frames, the target to be identified is determined to be a new target; according to the parameter information corresponding to the new target, the target in the two adjacent frames of images is determined. The Gaussian distribution corresponding to the new target in one frame of image; according to the Gaussian distribution corresponding to the new target in the latter frame of two adjacent images, determine the target to be recognized in the latter frame of the two adjacent images. .

The third implementation method of step 204: If it is determined that the probability result information is that parameter information corresponding to the Gaussian distribution has not been obtained, then it is determined that the target to be identified corresponding to the Gaussian distribution is in a disappearing state; it is determined that the target is adjacent to the location of the Gaussian distribution and has a preset number The target to be recognized, and the parameter information corresponding to the target to be recognized; wherein, the Gaussian distribution includes the first face area, and the parameter information corresponding to the target to be recognized includes the second face area; according to the preset second constraint information, in Among the preset number of second face areas, determine the target face area that is equal to the first face area; wherein, the second constraint information represents the pixel ratio of the face area of the target to be recognized; according to the target face area According to the parameter information corresponding to the area, determine the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent images; determine the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent images. The target to be recognized in the latter image of the two adjacent frames.

For example, for two adjacent frames in a multi-frame image, based on a preset formula, the electronic device compares the parameter list of the latter frame of the two adjacent images with the previous frame of the two adjacent images. The Gaussian distribution corresponding to the target to be identified in the image is subjected to probability calculation processing to obtain the probability result information corresponding to the target to be identified in the latter image of the two adjacent frames. The probability result information is used to indicate the latter of the two adjacent frames of images. Gaussian distribution corresponding to the target to be recognized in the frame image. According to the probability result information, the target to be recognized in the latter frame of the two adjacent images is determined, where the target to be recognized in the latter frame of the two adjacent images is the same as the target in the previous frame of the two adjacent images. The target to be recognized corresponding to the target to be recognized in the frame image.

For example, in the first implementation of step 204, if the probability result information is determined to obtain a likelihood probability value, where the likelihood probability value represents each target to be identified in the latter frame of the two adjacent images. parameter information, respectively, and the likelihood probability values between the multiple Gaussian distributions in the previous frame image in the two adjacent frames of images, then among the multiple likelihood probability values, the maximum likelihood corresponding to the target to be identified is determined Probability value, where the maximum likelihood probability value is used to indicate the Gaussian distribution corresponding to the target to be recognized in the latter image of two adjacent frames. Finally, according to the maximum likelihood probability value corresponding to the target to be recognized, the target to be recognized in the latter frame of the two adjacent frames of images is determined, where the target to be recognized in the latter frame of the two adjacent frames of images is The target to be recognized is equal to the Gaussian distribution of the target to be recognized in the previous frame of the two adjacent images.

After determining the target to be recognized in the latter frame of the two adjacent images, the Gaussian distribution of the target to be recognized in the latter frame of the two adjacent images can be further verified. Specifically, the parameter information corresponding to the target to be recognized includes the face width value. The electronic device determines the target to be recognized in the latter frame of the two adjacent frames of images according to the preset first constraint information, corresponding to the target to be recognized. The distance between the positions of the Gaussian distribution, and the first constraint information represents the position information corresponding to the target to be identified. Then compare the distance of the target to be recognized with the face width value of the target to be recognized. If it is determined that the distance of the target to be recognized is greater than the face width value of the target to be recognized, it means that the target to be recognized is the last in the two adjacent frames of images. The displacement in the frame image is too large, and the target to be recognized is not the same as the target to be recognized in the previous frame of the two adjacent frames. Therefore, the target to be recognized in the latter frame of the two adjacent frames is corresponding to The Gaussian distribution is replaced with the Gaussian distribution corresponding to the target to be recognized in the previous frame of the two adjacent images, and the Gaussian distribution corresponding to the target to be recognized in the latter image of the two adjacent frames is obtained. Finally, according to the updated Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent frames of images, the target to be recognized in the latter frame of the two adjacent frames of images is determined.

For example, if the probability result information is determined to obtain a likelihood probability value, the likelihood probability value represents the parameter information of the first target to be recognized in the latter image of two adjacent frames, respectively, and the parameter information of the first target to be identified in the latter image of the two adjacent frames. The likelihood probability value between multiple Gaussian distributions in the previous frame image. The likelihood probability value includes: the parameter information of the first target to be recognized and the Gaussian distribution G ₁ in the previous frame image in the two adjacent frames of images. The likelihood probability value P _1,1 between the parameter information of the first target to be identified and the Gaussian distribution G ₂ in the previous frame of the two adjacent images, P _1,2 , The likelihood probability value P _1,3 between the parameter information of the first target to be recognized and the Gaussian distribution G ₃ in the previous frame of the two adjacent images. Then among multiple likelihood probability values, the maximum likelihood probability value P _1,3 corresponding to the target to be identified is determined. Then, according to the maximum likelihood probability value P _1,3 corresponding to the target to be identified, two adjacent ones can be determined. The Gaussian distribution of the first target to be recognized in the next frame of the image is G ₃ .

For example, in the second implementation of step 204, if it is determined that the probability result information is that the likelihood probability value is not obtained, where the likelihood probability value represents the target to be identified in the latter frame of the two adjacent images. parameter information, respectively, and the likelihood probability values between multiple Gaussian distributions in the previous frame of the two adjacent images, then it is determined that the target to be identified in the latter frame of the two adjacent frames is the new target. , and then calculate the Gaussian distribution corresponding to the new target in the latter frame of the two adjacent frames of images based on the parameter information corresponding to the new target. Finally, according to the Gaussian distribution corresponding to the new target in the latter image of the two adjacent frames, the target to be recognized in the latter image of the two adjacent frames is determined.

For example, in the third implementation of step 204, if it is determined that the probability result information is that parameter information corresponding to the Gaussian distribution has not been obtained, it is determined that the target to be identified corresponding to the Gaussian distribution is in a disappearing state, where the disappearing state includes the to-be-identified target. Recognize that the target is blocked or lowered, etc., determine a preset number of targets to be identified adjacent to the location of the Gaussian distribution, and parameter information corresponding to the preset number of targets to be identified, where the Gaussian distribution includes the first face area, The parameter information corresponding to the target to be recognized includes the second face area. The preset number of targets to be recognized can be the three targets to be recognized that are closest to the location of the Gaussian distribution. There is no limit to the preset number.

Then, according to the preset second constraint condition information, among the preset number of second face areas, the preset number of second face areas are compared with the first face area in sequence, and the difference between the preset number of second face areas and the first person's face area is determined. The target face area is equal to the face area, and the second constraint information represents the pixel ratio of the face area of the target to be recognized. Among them, because the seating arrangement in the classroom is usually relatively fixed, with a matrix layout of multiple rows in the front and back, when occlusion occurs, it is usually the students in the front row who block the students in the back row, and the students in the front row occupy a larger proportion of the pixels in the video screen. Therefore, the pixel ratio of the face area of the target to be recognized can be stored in advance. Finally, according to the parameter information corresponding to the target face area, the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent frames of images is determined, and then based on the target to be recognized in the latter frame of the two adjacent frames of images, the Gaussian distribution is determined. The Gaussian distribution corresponding to the target determines the target to be recognized in the latter image of the two adjacent frames.

205. Generate an identification code corresponding to the target to be identified in the last frame of the multi-frame image, and generate target information corresponding to the identification code; wherein, the identification code represents the identity of the target to be identified, and the target information represents the target sent by the target to be identified. The number of preset actions.

For example, this step can refer to step 103 in Figure 1 and will not be described again.

In the embodiment of the present application, multiple frame images are recognized and processed respectively according to the preset depth target detection network to obtain a parameter list of each image; wherein, the depth target detection network is used to indicate target identification information, and the parameter list includes images Parameter information of the target to be identified in . According to the parameter list of the first frame image in the multi-frame image, the Gaussian distribution corresponding to the target to be recognized in the first frame image is determined; wherein the Gaussian distribution represents the position information of the target to be recognized. For two adjacent frames in a multi-frame image, calculate the probability of the parameter list of the latter frame of the two adjacent images and the Gaussian distribution corresponding to the target to be recognized in the previous frame of the two adjacent images. Process to obtain the probability result information corresponding to the target to be identified in the latter frame of the two adjacent images; where the probability result information is used to indicate the Gaussian corresponding to the target to be identified in the latter frame of the two adjacent images. distributed. According to the probability result information, the target to be recognized in the latter frame of the two adjacent frames of images is determined. Generate an identification code corresponding to the target to be identified in the last frame of the multi-frame image, and generate target information corresponding to the identification code; wherein, the identification code represents the identity of the target to be identified, and the target information represents the preset issued by the target to be identified. The number of actions. Therefore, by combining multiple constraint information, the same target to be identified in multiple frames of images can be determined, and an identification code corresponding to the target to be identified can be generated, which greatly improves the stability of tracking the target to be identified and solves the user's problem. It is difficult to track the identification code, resulting in technical problems such as the inability to obtain the behavioral information corresponding to the identification code.

Figure 3 is a schematic structural diagram of an image processing device provided by an embodiment of the present application. As shown in Figure 3, the device includes:

The recognition unit 31 is used to perform recognition processing on multiple frame images to obtain a parameter list of each image; wherein the parameter list includes parameter information of the target to be recognized in the image.

The first determining unit 32 is configured to, for two adjacent frame images in the multi-frame images, based on the parameter list of the previous frame image in the two adjacent frame images and the parameter list of the subsequent frame image in the two adjacent frame images, Determine the target to be recognized in the latter frame of the two adjacent images; where the target to be recognized in the latter frame of the two adjacent frames is the same as the target to be recognized in the previous frame of the two adjacent images. The target to be recognized corresponding to the recognized target.

The first generation unit 33 is used to generate an identification code corresponding to the target to be identified in the last frame of the multi-frame image.

The second generation unit 34 is used to generate target information corresponding to the identification code; wherein the identification code represents the identity of the target to be identified, and the target information represents the number of times the target to be identified has issued a preset action.

The device of this embodiment can execute the technical solution in the above method. Its specific implementation process and technical principles are the same and will not be described again here.

Figure 4 is a schematic structural diagram of another image processing device provided by an embodiment of the present application. Based on the embodiment shown in Figure 3, as shown in Figure 4, the first determination unit 32 includes:

The first determination module 321 is used to determine the Gaussian distribution corresponding to the target to be recognized in the first frame image according to the parameter list of the first frame image in the multi-frame image; wherein the Gaussian distribution represents the position information of the target to be recognized.

The calculation module 322 is used to, for two adjacent frames of images in the multi-frame images, correspond the parameter list of the latter frame of the image in the two adjacent frames of images to the target to be recognized in the previous frame of the image of the two adjacent frames of images. Probability calculation processing is performed on the Gaussian distribution to obtain the probability result information corresponding to the target to be identified in the latter frame of the two adjacent images; where the probability result information is used to indicate the probability result information in the latter frame of the two adjacent images. Gaussian distribution corresponding to the target to be recognized.

The second determination module 323 is used to determine the target to be recognized in the latter frame of two adjacent frames of images based on the probability result information.

In an example, the second determination module 323 includes:

The first determination sub-module 3231 is used to obtain the likelihood probability value if the probability result information is determined, where the likelihood probability value represents the parameter information of the target to be identified in the latter frame of the two adjacent frames of images, respectively with the adjacent The likelihood probability value between the multiple Gaussian distributions in the previous frame image in the two frames of images is used to determine the maximum likelihood probability value corresponding to the target to be identified; where the maximum likelihood probability value is used to indicate two adjacent frames of images. Gaussian distribution corresponding to the target to be recognized in the next frame of image.

The second determination sub-module 3232 is used to determine the target to be recognized in the latter frame of the two adjacent frames of images based on the maximum likelihood probability value corresponding to the target to be recognized.

In one example, the parameter information corresponding to the target to be recognized includes a face width value; the device also includes:

The second determination unit 41 is configured to determine the distance between the target to be recognized in the latter frame of the two adjacent frames of images and the location of the Gaussian distribution corresponding to the target to be recognized based on the preset first constraint information; Among them, the first constraint information represents the position information corresponding to the target to be identified.

The update unit 42 is configured to, if it is determined that the distance is greater than the face width value of the target to be recognized, replace the Gaussian distribution corresponding to the target to be recognized in the latter image of the two adjacent frames with the Gaussian distribution of the target to be recognized in the previous two adjacent frames of images. The Gaussian distribution corresponding to the target to be recognized in one frame of image is obtained, and the Gaussian distribution corresponding to the target to be recognized in the latter frame of two adjacent frames of images is obtained.

The third determination unit 43 is configured to determine the target to be recognized in the latter image of the two adjacent frames according to the updated Gaussian distribution corresponding to the target to be recognized in the latter image of the two adjacent frames.

In an example, the second determination module 323 includes:

The third determination sub-module 3233 is used to determine that the probability result information is that the likelihood probability value has not been obtained, where the likelihood probability value represents the parameter information of the target to be identified in the latter frame of the two adjacent frames of images, respectively with the corresponding If the likelihood probability value between the multiple Gaussian distributions in the previous frame image in the two adjacent frames is determined, the target to be identified is determined to be a new target.

The fourth determination sub-module 3234 is used to determine the Gaussian distribution corresponding to the new target in the latter frame of the two adjacent frames of images based on the parameter information corresponding to the new target.

The fifth determination sub-module 3235 is used to determine the target to be recognized in the latter frame of the two adjacent frames of images based on the Gaussian distribution corresponding to the newly added target in the latter frame of the two adjacent images.

In an example, the second determination module 323 includes:

The sixth determination sub-module 3236 is used to determine that the target to be identified corresponding to the Gaussian distribution is in a disappeared state if the probability result information is determined to be that parameter information corresponding to the Gaussian distribution has not been obtained.

The seventh determination sub-module 3237 is used to determine a preset number of targets to be identified adjacent to the location of the Gaussian distribution, and parameter information corresponding to the targets to be identified; wherein the Gaussian distribution includes the first face area, and the target to be identified corresponds to The parameter information includes the second face area.

The eighth determination sub-module 3238 is used to determine a target face area equal to the first face area among a preset number of second face areas according to the preset second constraint information; wherein, the second constraint The condition information represents the pixel ratio of the face area of the target to be recognized.

The ninth determination sub-module 3239 is used to determine the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent frames of images based on the parameter information corresponding to the target face area.

The tenth determination sub-module 32310 is used to determine the target to be recognized in the latter frame of the two adjacent frames of images according to the Gaussian distribution corresponding to the target to be recognized in the latter frame of the two adjacent images.

In an example, the identification unit 31 is specifically used for:

According to the preset depth target detection network, multiple frame images are recognized and processed respectively to obtain a parameter list for each image; among them, the depth target detection network is used to indicate the target identification information, and the parameter list includes the parameters of the target to be recognized in the image. Parameter information.

In one example, the parameter information corresponding to the target to be recognized includes face information and human body information, where the face information includes the abscissa of the face center, the ordinate of the face center, the face width value, the face height value, and the face Area, human body information includes the horizontal coordinate of the human body center, the vertical coordinate of the human body center, the human body width value, the human body height value, and the human body area.

The device of this embodiment can execute the technical solution in the above method. Its specific implementation process and technical principles are the same and will not be described again here.

FIG. 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As shown in FIG. 5 , the electronic device includes: a memory 51 and a processor 52 .

The memory 51 stores computer programs that can be run on the processor 52 .

The processor 52 is configured to perform the method as provided in the above embodiments.

The electronic device also includes a receiver 53 and a transmitter 54 . The receiver 53 is used to receive instructions and data sent by the external device, and the transmitter 54 is used to send instructions and data to the external device.

Figure 6 is a block diagram of an electronic device provided by an embodiment of the present application. The electronic device may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, or a personal digital assistant. wait.

Device 600 may include one or more of the following components: processing component 602, memory 604, power component 606, multimedia component 608, audio component 610, input/output (I/O) interface 612, sensor component 614, and communications component 616.

Processing component 602 generally controls the overall operations of device 600, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 602 may include one or more processors 620 to execute instructions to complete all or part of the steps of the above method. Additionally, processing component 602 may include one or more modules that facilitate interaction between processing component 602 and other components. For example, processing component 602 may include a multimedia module to facilitate interaction between multimedia component 608 and processing component 602.

Memory 604 is configured to store various types of data to support operations at device 600 . Examples of such data include instructions for any application or method operating on device 600, contact data, phonebook data, messages, pictures, videos, etc. Memory 604 may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EEPROM), Programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk.

Power supply component 606 provides power to the various components of device 600. Power supply components 606 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to device 600 .

Multimedia component 608 includes a screen that provides an output interface between device 600 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. A touch sensor can not only sense the boundaries of a touch or swipe action, but also detect the duration and pressure associated with the touch or swipe action. In some embodiments, multimedia component 608 includes a front-facing camera and/or a rear-facing camera. When the device 600 is in an operating mode, such as a shooting mode or a video mode, the front camera and/or the rear camera may receive external multimedia data. Each front-facing camera and rear-facing camera can be a fixed optical lens system or have a focal length and optical zoom capabilities.

Audio component 610 is configured to output and/or input audio signals. For example, audio component 610 includes a microphone (MIC) configured to receive external audio signals when device 600 is in operating modes, such as call mode, recording mode, and speech recognition mode. The received audio signals may be further stored in memory 604 or sent via communications component 616 . In some embodiments, audio component 610 also includes a speaker for outputting audio signals.

The I/O interface 612 provides an interface between the processing component 602 and a peripheral interface module, which may be a keyboard, a click wheel, a button, etc. These buttons may include, but are not limited to: Home button, Volume buttons, Start button, and Lock button.

Sensor component 614 includes one or more sensors for providing various aspects of status assessment for device 600 . For example, the sensor component 614 can detect the open/closed state of the device 600, the relative positioning of components, such as the display and keypad of the device 600, the sensor component 614 can also detect the position change of the device 600 or a component of the device 600, the user The presence or absence of contact with device 600, device 600 orientation or acceleration/deceleration and temperature changes of device 600. Sensor assembly 614 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 614 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor component 614 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

Communication component 616 is configured to facilitate wired or wireless communication between apparatus 600 and other devices. Device 600 may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In one exemplary embodiment, the communication component 616 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communications component 616 also includes a near field communications (NFC) module to facilitate short-range communications. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

In an exemplary embodiment, apparatus 600 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable Gate array (FPGA), controller, microcontroller, microprocessor or other electronic components are implemented for executing the above method.

In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions, such as a memory 604 including instructions, which are executable by the processor 620 of the device 600 to complete the above method is also provided. For example, non-transitory computer-readable storage media may be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

Embodiments of the present application also provide a non-transitory computer-readable storage medium. When instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the method provided by the above embodiments.

Embodiments of the present application also provide a computer program product. The computer program product includes: a computer program. The computer program is stored in a readable storage medium. At least one processor of the electronic device can read the computer program from the readable storage medium. At least A processor executes a computer program so that the electronic device executes the solution provided by any of the above embodiments.

Other embodiments of the disclosure will be readily apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure that follow the general principles of the disclosure and include common knowledge or customary technical means in the technical field that are not disclosed in the disclosure. . It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It is to be understood that the present disclosure is not limited to the precise structures described above and illustrated in the accompanying drawings, and various modifications and changes may be made without departing from the scope thereof. The scope of the disclosure is limited only by the appended claims.

Claims (19)

1. An image processing method, comprising:

performing identification processing on multi-frame images to obtain a parameter list of each image; the parameter list comprises parameter information of an object to be identified in the image;

for two adjacent frame images in the multi-frame images, determining a target to be identified in a later frame image in the two adjacent frame images according to a parameter list of a former frame image in the two adjacent frame images and a parameter list of a later frame image in the two adjacent frame images; the target to be identified in the next frame of images in the two adjacent frames is the target to be identified corresponding to the target to be identified in the previous frame of images in the two adjacent frames;

generating an identification code corresponding to a target to be identified in the last frame of image in the multi-frame image, and generating target information corresponding to the identification code; the identification code characterizes the identity of the object to be identified, and the object information characterizes the times of preset actions sent by the object to be identified.

2. The method according to claim 1, wherein determining, for two adjacent frames of the multi-frame images, the object to be identified in a next frame of the two adjacent frames of images based on the parameter list of a previous frame of the two adjacent frames of images and the parameter list of a next frame of the two adjacent frames of images, comprises:

according to a parameter list of a first frame image in the multi-frame image, determining Gaussian distribution corresponding to a target to be identified in the first frame image; wherein the gaussian distribution characterizes position information of the object to be identified;

aiming at two adjacent frame images in the multi-frame images, carrying out probability calculation processing on a parameter list of a next frame image in the two adjacent frame images and Gaussian distribution corresponding to a target to be identified in a previous frame image in the two adjacent frame images to obtain probability result information corresponding to the target to be identified in the next frame image in the two adjacent frame images; the probability result information is used for indicating Gaussian distribution corresponding to a target to be identified in a later frame of images in two adjacent frames of images;

and determining the target to be identified in the next frame of images in the two adjacent frames of images according to the probability result information.

3. The method according to claim 2, wherein determining the target to be identified in the next image of the two adjacent images according to the probability result information comprises:

if the probability result information is determined to be a likelihood probability value, wherein the likelihood probability value represents parameter information of an object to be identified in a next frame image in two adjacent frame images and likelihood probability values between the parameter information and a plurality of Gaussian distributions in a previous frame image in the two adjacent frame images respectively, determining a maximum likelihood probability value corresponding to the object to be identified; the maximum likelihood probability value is used for indicating Gaussian distribution corresponding to a target to be identified in a later frame image in two adjacent frame images;

and determining the target to be identified in the next frame of images in the two adjacent frames according to the maximum likelihood probability value corresponding to the target to be identified.

4. A method according to claim 3, wherein the parameter information corresponding to the object to be identified comprises a face width value; the method further comprises the steps of:

according to preset first constraint condition information, determining a target to be identified in a later frame of images in two adjacent frames of images, and determining the distance between positions of Gaussian distribution corresponding to the target to be identified; the first constraint condition information characterizes position information corresponding to a target to be identified;

If the distance is determined to be larger than the face width value of the target to be identified, the Gaussian distribution corresponding to the target to be identified in the next frame of images in the two adjacent frames is replaced by the Gaussian distribution corresponding to the target to be identified in the previous frame of images in the two adjacent frames, and the Gaussian distribution corresponding to the target to be identified in the next frame of images in the two adjacent frames after updating is obtained;

and determining the target to be identified in the next frame of images in the two adjacent frames according to the Gaussian distribution corresponding to the target to be identified in the next frame of images in the two adjacent frames after updating.

5. The method according to claim 2, wherein determining the target to be identified in the next image of the two adjacent images according to the probability result information comprises:

if the probability result information is determined to be the likelihood probability value which is not obtained, wherein the likelihood probability value represents parameter information of the target to be identified in the next frame image in the two adjacent frame images and likelihood probability values between the parameter information and a plurality of Gaussian distributions in the previous frame image in the two adjacent frame images respectively, the target to be identified is determined to be a newly added target;

according to the parameter information corresponding to the newly-added target, determining Gaussian distribution corresponding to the newly-added target in the next frame of image in the two adjacent frames of images;

And determining the target to be identified in the next frame image in the two adjacent frame images according to the Gaussian distribution corresponding to the new target in the next frame image in the two adjacent frame images.

6. The method according to claim 2, wherein determining the target to be identified in the next image of the two adjacent images according to the probability result information comprises:

if the probability result information is determined to be that the parameter information corresponding to the Gaussian distribution is not obtained, determining that the target to be identified corresponding to the Gaussian distribution is in a vanishing state;

determining a preset number of targets to be identified adjacent to positions of Gaussian distribution and parameter information corresponding to the targets to be identified; the Gaussian distribution comprises a first face area, and the parameter information corresponding to the target to be identified comprises a second face area;

determining a target face area equal to the first face area in a preset number of second face areas according to preset second constraint condition information; wherein the second constraint condition information characterizes the pixel duty ratio of the face area of the object to be identified;

according to the parameter information corresponding to the target face area, determining Gaussian distribution corresponding to a target to be identified in a later frame of images in two adjacent frames of images;

And determining the target to be identified in the next frame of images in the two adjacent frames according to the Gaussian distribution corresponding to the target to be identified in the next frame of images in the two adjacent frames.

7. The method of claim 1, wherein the identifying of the plurality of images to obtain the parameter list for each of the images comprises:

respectively carrying out identification processing on multiple frames of images according to a preset depth target monitoring network to obtain a parameter list of each image; the depth target detection network is used for indicating target identification information, and the parameter list comprises parameter information of targets to be identified in the image.

8. The method according to any one of claims 1-7, wherein the parameter information corresponding to the object to be identified includes face information and human body information, wherein the face information includes a face center abscissa, a face center ordinate, a face width value, a face height value, and a face area, and the human body information includes a human body center abscissa, a human body center ordinate, a human body width value, a human body height value, and a human body area.

9. An image processing apparatus, comprising:

The identification unit is used for carrying out identification processing on the multi-frame images to obtain a parameter list of each image; the parameter list comprises parameter information of an object to be identified in the image;

the first determining unit is used for determining a target to be identified in a next frame image in the two adjacent frame images according to a parameter list of a previous frame image in the two adjacent frame images and a parameter list of a next frame image in the two adjacent frame images aiming at the two adjacent frame images; the target to be identified in the next frame of images in the two adjacent frames is the target to be identified corresponding to the target to be identified in the previous frame of images in the two adjacent frames;

the first generation unit is used for generating an identification code corresponding to the target to be identified in the last frame of image in the multi-frame images;

a second generation unit, configured to generate target information corresponding to the identification code; the identification code characterizes the identity of the object to be identified, and the object information characterizes the times of preset actions sent by the object to be identified.

10. The apparatus according to claim 9, wherein the first determining unit includes:

The first determining module is used for determining Gaussian distribution corresponding to a target to be identified in a first frame image according to a parameter list of the first frame image in the multi-frame image; wherein the gaussian distribution characterizes position information of the object to be identified;

the computing module is used for carrying out probability computing processing on a parameter list of a next frame image in the two adjacent frame images and Gaussian distribution corresponding to a target to be identified in a previous frame image in the two adjacent frame images aiming at the two adjacent frame images in the multi-frame image to obtain probability result information corresponding to the target to be identified in the next frame image in the two adjacent frame images; the probability result information is used for indicating Gaussian distribution corresponding to a target to be identified in a later frame of images in two adjacent frames of images;

and the second determining module is used for determining the target to be identified in the next frame of image in the two adjacent frames of images according to the probability result information.

11. The apparatus of claim 10, wherein the second determining module comprises:

the first determining submodule is used for determining a maximum likelihood probability value corresponding to the object to be identified if the probability result information is determined to be the likelihood probability value, wherein the likelihood probability value represents parameter information of the object to be identified in a later frame image in two adjacent frame images and likelihood probability values between the parameter information and a plurality of Gaussian distributions in the previous frame image in the two adjacent frame images respectively; the maximum likelihood probability value is used for indicating Gaussian distribution corresponding to a target to be identified in a later frame image in two adjacent frame images;

And the second determining submodule is used for determining the target to be identified in the next frame of images in the two adjacent frames of images according to the maximum likelihood probability value corresponding to the target to be identified.

12. The apparatus according to claim 11, wherein the parameter information corresponding to the object to be identified includes a face width value; the apparatus further comprises:

the second determining unit is used for determining a target to be identified in a next frame of images in two adjacent frames according to preset first constraint condition information, and the distance between positions of Gaussian distribution corresponding to the target to be identified; the first constraint condition information characterizes position information corresponding to a target to be identified;

the updating unit is used for replacing the Gaussian distribution corresponding to the target to be identified in the next frame of images in the next two frames of images with the Gaussian distribution corresponding to the target to be identified in the previous frame of images in the next two frames of images if the distance is determined to be larger than the face width value of the target to be identified, so that the Gaussian distribution corresponding to the target to be identified in the next frame of images in the next two frames of images after updating is obtained;

and the third determining unit is used for determining the target to be identified in the next frame image in the two adjacent frame images according to the Gaussian distribution corresponding to the target to be identified in the next frame image in the two adjacent frame images after updating.

13. The apparatus of claim 10, wherein the second determining module comprises:

a third determining sub-module, configured to determine that the object to be identified is an added object if the probability result information is determined that the likelihood probability value is not obtained, where the likelihood probability value characterizes parameter information of the object to be identified in a next frame of image in two adjacent frames of images, and likelihood probability values between the parameter information and a plurality of gaussian distributions in a previous frame of image in two adjacent frames of images respectively;

a fourth determining submodule, configured to determine gaussian distribution corresponding to a new added target in a next frame of image in two adjacent frames of images according to parameter information corresponding to the new added target;

and a fifth determining submodule, configured to determine a target to be identified in a next frame image in two adjacent frame images according to gaussian distribution corresponding to a new target in the next frame image in the two adjacent frame images.

14. The apparatus of claim 10, wherein the second determining module comprises:

a sixth determining submodule, configured to determine that a target to be identified corresponding to the gaussian distribution is in a vanishing state if it is determined that the probability result information is parameter information corresponding to the gaussian distribution;

A seventh determining submodule, configured to determine a preset number of targets to be identified adjacent to positions where gaussian distributions are located, and parameter information corresponding to the targets to be identified; the Gaussian distribution comprises a first face area, and the parameter information corresponding to the target to be identified comprises a second face area;

an eighth determining submodule, configured to determine a target face area equal to the first face area among a preset number of second face areas according to preset second constraint condition information; wherein the second constraint condition information characterizes the pixel duty ratio of the face area of the object to be identified;

a ninth determining submodule, configured to determine gaussian distribution corresponding to a target to be identified in a later frame of images in two adjacent frames of images according to parameter information corresponding to the target face area;

and the tenth determination submodule is used for determining the target to be identified in the next frame image in the two adjacent frame images according to the Gaussian distribution corresponding to the target to be identified in the next frame image in the two adjacent frame images.

15. The device according to claim 9, characterized in that said identification unit is in particular adapted to:

respectively carrying out identification processing on multiple frames of images according to a preset depth target monitoring network to obtain a parameter list of each image; the depth target detection network is used for indicating target identification information, and the parameter list comprises parameter information of targets to be identified in the image.

16. The apparatus according to any one of claims 9-15, wherein the parameter information corresponding to the object to be identified includes face information and human body information, wherein the face information includes a face center abscissa, a face center ordinate, a face width value, a face height value, and a face area, and the human body information includes a human body center abscissa, a human body center ordinate, a human body width value, a human body height value, and a human body area.

17. An electronic device comprising a memory, a processor, the memory having stored therein a computer program executable on the processor, the processor implementing the method of any of the preceding claims 1-8 when the computer program is executed.

18. A computer readable storage medium having stored therein computer executable instructions which when executed by a processor are adapted to carry out the method of any one of claims 1-8.

19. A computer program product comprising a computer program which, when executed by a processor, implements the method of any of claims 1-8.