CN114567635A - Edge data processing method and device and computer readable storage medium - Google Patents

Abstract

The invention discloses a method and a device for processing edge data and a computer readable storage medium, wherein the method for processing the edge data comprises the following steps: when the redundant computing power of the current edge equipment is smaller than the computing power required for processing the detection data received by the current edge equipment, dividing the detection data into a plurality of first data units, and sending the first data units to a sharing platform; when the redundant computing power of the current edge device is larger than the computing power required for processing a first data unit or a second data unit in the shared platform, receiving and processing the first data unit or the second data unit, wherein the second data unit is data sent to the shared platform by other edge devices. According to the invention, the current edge device and the other edge devices can share computing power through the sharing platform, so that transmission resources occupied by transmission of the detection data are reduced.

Description

Edge data processing method, device and computer-readable storage medium

technical field

The present invention relates to the technical field of edge devices, and in particular, to a method, device and computer-readable storage medium for processing edge data.

Background technique

The rapid development of electronic technology has brought more and more convenient living conditions to people. The construction of smart cities, smart districts and counties, smart communities, smart communities, smart homes, etc. requires various electronic devices. The deployment locations, business functions, and processing performance of these electronic devices are not the same, which leads to various The emergence of edge devices, such as cameras that collect images, directional sound collectors that collect voice signals, network transmission modules that perform wireless transmission, and temperature and humidity collection modules that collect temperature and humidity, etc.

In the same application scenario, the requirements for each edge device are different. Therefore, when the amount of data collected by the edge device is huge, the computing power of the edge device may not be enough to process the collected data. Transfer to the server for processing, however this will take up a lot of transfer resources.

The above content is only used to assist the understanding of the technical solutions of the present invention, and does not mean that the above content is the prior art.

SUMMARY OF THE INVENTION

The main purpose of the present invention is to provide a processing method, device and computer-readable storage medium for edge data, aiming at solving the technical problem of occupying a large amount of transmission resources when processing edge data.

To achieve the above object, the present invention provides a method for processing edge data, the method for processing edge data is applied to edge devices, and the method for processing edge data includes the following steps:

When the redundant computing power of the current edge device is less than the computing power required for processing the detection data received by the current edge device, the detection data is divided into several first data units, and the first data units are sent to bdfgh;

When the redundant computing power of the current edge device is greater than the computing power required to process the first data unit or the second data unit in the shared platform, receive and process the first data unit or the second data unit, Wherein, the second data unit is data sent by other edge devices to the sharing platform.

Optionally, the step of dividing the detection data into several first data units includes:

determining the segmentation rule of the detection data;

The detection data is segmented according to the segmentation rule.

Optionally, the division rule includes division according to time period and division according to basic unit composed of data.

Optionally, after the step of dividing the detection data according to the dividing rule, the step further includes:

When the computing power required for processing the first data unit is greater than the preset computing power, the number of the first data unit obtained after dividing the detection data is increased.

Optionally, the method for processing edge data further includes:

obtaining a mapping relationship between the computing power required to process the first data unit or the second data unit and the CPU utilization of the edge device;

The redundant computing power of the current edge device is determined according to the mapping relationship.

Optionally, the steps of the edge data processing method further include:

determining the arrangement order of the first data unit in the detection time;

The processing results of the first data unit in the shared platform are correspondingly merged according to the arrangement order.

Optionally, the step of correspondingly merging the processing results of the first data unit in the sharing platform according to the arrangement order includes:

receiving the processing result of the first data unit in the sharing platform;

The processing result of the current edge device on the first data unit and the received processing result of the first data unit in the shared platform are combined according to the arrangement order.

Optionally, the steps of the edge data processing method further include:

acquiring a sub-model for processing the detection data, wherein the sub-model acquired by the current edge device and the sub-model loaded on the other edge devices constitute a deep neural network model;

The detection data is input into the sub-model, and the processing result of the detection data by the sub-model is sent to the sub-models on the other edge devices.

In addition, in order to achieve the above object, the present invention also provides an edge device, the edge device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program When executed by the processor, the steps of the method for processing edge data as described above are implemented.

In addition, in order to achieve the above object, the present invention also provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a processing program for edge data, and the processing program for edge data is executed by a processor When implementing the steps of the edge data processing method as described above.

In an edge data processing method proposed by an embodiment of the present invention, when the redundant computing power of the current edge device is less than the computing power required to process the detection data received by the current edge device, the detection data is divided into several a data unit, and send the first data unit to the shared platform; when the redundant computing power of the current edge device is greater than the computing power required to process the first data unit or the second data unit in the shared platform, Receive and process the first data unit or the second data unit, so that the current edge device and the other edge devices can share computing power through the shared platform, reducing the transmission occupied by the transmission of the detection data resource.

Description of drawings

1 is a schematic diagram of a terminal structure of a hardware operating environment involved in an embodiment of the present invention;

2 is a schematic flowchart of a first embodiment of a method for processing edge data according to the present invention;

FIG. 3 is a schematic diagram of the refinement process of step S10 in FIG. 2;

4 is a schematic flowchart of a second embodiment of a method for processing edge data according to the present invention;

FIG. 5 is a schematic diagram of the refinement flow of step S40 in FIG. 4 .

The realization, functional characteristics and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments.

Detailed ways

It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

The main solution of the embodiment of the present invention is: when the redundant computing power of the current edge device is less than the computing power required for processing the detection data received by the current edge device, the detection data is divided into several first data units, Send the first data unit to the shared platform; when the redundant computing power of the current edge device is greater than the computing power required to process the first data unit or the second data unit in the shared platform, receive and process all the The first data unit or the second data unit, wherein the second data unit is data sent by other edge devices to the sharing platform.

In an example solution, the edge device receives the detection data detected by the sensor, and then processes the detection data through the edge device, but when the computing power of the edge device is not enough to process the detection data, it needs to The detection data is transmitted to the server through the edge device, and the detection data sent by the edge device is processed through the server. In the process of transmitting the detection data to the server by the edge device, a large amount of transmission resources will be occupied , resulting in a waste of resources.

The present invention provides a solution, when the redundant computing power of the current edge device is less than the computing power required for processing the detection data received by the current edge device, the detection data is divided into several first data units, and the detection data is divided into several first data units, and the The first data unit is sent to the shared platform; when the redundant computing power of the current edge device is greater than the computing power required to process the first data unit or the second data unit in the shared platform, receive and process the first data unit. A data unit or the second data unit, so that the current edge device and the other edge devices can share computing power through the shared platform, thereby reducing the transmission resources occupied by the transmission of the detection data.

As shown in FIG. 1 , FIG. 1 is a schematic structural diagram of a terminal of a hardware operating environment involved in an embodiment of the present invention.

The terminal in the embodiment of the present invention may be an edge device, a PC, or a mobile terminal device with a display function, such as a smart phone, a tablet computer, and a portable computer.

As shown in FIG. 1 , the terminal may include: a processor 1001 , such as a CPU, a network interface 1004 , a user interface 1003 , a memory 1005 , and a communication bus 1002 . Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. Optionally, the network interface 1004 may include a standard wired interface and a wireless interface (eg, a WI-FI interface). The memory 1005 may be high-speed RAM memory, or may be non-volatile memory, such as disk memory. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001 .

Optionally, the terminal may further include a camera, an RF (Radio Frequency, radio frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like. Among them, sensors such as light sensors, motion sensors and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display screen according to the brightness of the ambient light, and the proximity sensor may turn off the display screen and/or turn off the display screen when the mobile terminal is moved to the ear. Backlight. As a kind of motion sensor, the gravitational acceleration sensor can detect the magnitude of acceleration in various directions (generally three axes), and can detect the magnitude and direction of gravity when stationary, and can be used for applications that recognize the posture of mobile terminals (such as horizontal and vertical screen switching, related games, magnetometer attitude calibration), vibration recognition related functions (such as pedometer, tapping), etc.; of course, the mobile terminal can also be equipped with other sensors such as gyroscope, barometer, hygrometer, thermometer, infrared sensor, etc. No longer.

Those skilled in the art can understand that the terminal structure shown in FIG. 1 does not constitute a limitation on the terminal, and may include more or less components than the one shown, or combine some components, or arrange different components.

As shown in FIG. 1 , the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a processing program for edge data.

In the terminal shown in FIG. 1 , the network interface 1004 is mainly used to connect to the background server and perform data communication with the background server; the user interface 1003 is mainly used to connect to the client (client) and perform data communication with the client; and the processor 1001 can be used to invoke a handler for edge data stored in memory 1005 and perform the following operations:

When the redundant computing power of the current edge device is less than the computing power required for processing the detection data received by the current edge device, the detection data is divided into several first data units, and the first data units are sent to bdfgh;

When the redundant computing power of the current edge device is greater than the computing power required for processing the first data unit or the second data unit in the shared platform, the first data unit or the second data unit is received and processed.

Further, the processor 1001 can call the processing program of the edge data stored in the memory 1005, and also perform the following operations:

determining the segmentation rule of the detection data;

The detection data is segmented according to the segmentation rule.

Further, the processor 1001 can call the processing program of the edge data stored in the memory 1005, and also perform the following operations:

When the computing power required for processing the first data unit is greater than the preset computing power, the number of the first data unit obtained after dividing the detection data is increased.

Further, the processor 1001 can call the processing program of the edge data stored in the memory 1005, and also perform the following operations:

obtaining a mapping relationship between the computing power required to process the first data unit or the second data unit and the CPU utilization of the edge device;

The redundant computing power of the current edge device is determined according to the mapping relationship.

Further, the processor 1001 can call the processing program of the edge data stored in the memory 1005, and also perform the following operations:

determining the arrangement order of the first data unit in the detection time;

The processing results of the first data unit in the shared platform are correspondingly merged according to the arrangement order.

Further, the processor 1001 can call the processing program of the edge data stored in the memory 1005, and also perform the following operations:

receiving the processing result of the first data unit in the sharing platform;

The processing result of the current edge device on the first data unit and the received processing result of the first data unit in the shared platform are combined according to the arrangement order.

Further, the processor 1001 can call the processing program of the edge data stored in the memory 1005, and also perform the following operations:

acquiring a sub-model for processing the detection data, wherein the sub-model acquired by the current edge device and the sub-model loaded on the other edge devices constitute a deep neural network model;

The detection data is input into the sub-model, and the processing result of the detection data by the sub-model is sent to the sub-models on the other edge devices.

An embodiment of the present invention provides a method for processing edge data, and the method for processing edge data is applied to an edge device.

As shown in FIG. 2 , the first embodiment of the present invention provides a method for processing edge data, and the method for processing edge data includes:

Step S10, when the redundant computing power of the current edge device is less than the computing power required for processing the detection data received by the current edge device, the detection data is divided into several first data units, and the first data is divided into several first data units. Units are sent to a shared platform.

When building smart cities, smart districts, smart communities, smart communities and smart homes, multiple edge devices need to be used, and sensors with different functions are set on the edge devices, so that the edge devices have different services Function. For example, a camera for capturing images is set on the edge device, so that the edge device has an image capturing function. Optionally, the camera may be a standard-definition camera, a high-definition camera, or an AI analysis camera capable of image detection cameras and image semantic segmentation cameras; a sound sensor is set on the edge device, so that the edge device has a sound collection function, optionally, the sound sensor can be a directional sound collector, or a full A wireless transceiver is set on the edge device, so that the edge device has a wireless transmission function; a temperature sensor and a humidity sensor are set on the edge device, so that the edge device has the function of collecting temperature and humidity .

In practical applications, when the edge device receives the image data detected by the sensor, since the computing power required to process the image data is large, the edge device is required to have a large computing power; When the device receives the sound data, temperature data or humidity data detected by the sensor, since the computing power required to process the sound data, temperature data or humidity data is small, the computing power of the edge device is not required much. The edge devices based on different functions require different computing power to process the corresponding detection data received. Therefore, when designing the edge device, the computing power of the edge device can be appropriately reduced, so that the computing power of the edge device can be reduced. The power is smaller than the maximum computing power required for processing the detection data, and is greater than the minimum computing power required for processing the detection data. When the computing power of the edge device to process the detection data is insufficient, the detection data may be divided, and then calculated by other edge devices with redundant computing power, and finally the calculation results are merged.

The edge device may be a standardized device, so that the interface, computing capability and storage capability of the edge device are all the same. Since the edge device is a standardized device, the edge device can be mass-produced, so as to reduce the production cost of the edge device and facilitate the maintenance of the edge device. When designing the computing power of the edge device, the computing power of the edge device can be determined according to the total computing power required to process the detection data and the number of the edge devices, and in order to make the computing power of the edge device To meet the needs to cope with unexpected situations, the computing power of the edge device can be appropriately increased.

After the sensor collects the detection data, it sends the detection data to the current edge device, so that the detection data is processed by the current edge device. When the redundant computing power of the current edge device is less than the computing power required to process the detection data received by the current edge device, it means that the computing power of the current edge device is insufficient to process the received detection data, At this time, the current edge device divides the received detection data to divide the detection data into several first data units; since the first data unit is obtained by dividing the detection data, therefore The computing power required for processing the first data unit is less than the computing power required for processing the detection data, and then the first data unit is processed through the current edge device and other edge devices. The redundant computing power of the current edge device refers to the computing power currently available to the current edge device.

Optionally, after the current edge device divides the detection data into the first data units, it may send all the divided first data units to the shared platform, and other edge devices receive the shared platform. the first data unit sent by the shared platform, and process the first data unit; or, the current edge device and the other edge devices receive the first data unit sent by the sharing platform, and process the first data unit first data unit. After the detection data is divided into the first data unit, the computing power required for processing the first data unit is reduced, so the computing power of the edge device can meet the requirements for processing the first data unit.

Optionally, after the current edge device divides the detection data into the first data units, the divided first data units may be sent to the sharing platform, and other edge devices may receive the sharing platform. The first data unit sent and processed, while the first data unit not sent to the sharing platform is processed by the current edge device.

The shared platform may be a shared platform built by a DDS (Data Distribution Service, data distribution service) protocol, which is used to solve the data publishing/subscription problem in the distributed real-time system. The DDS protocol standardizes the interfaces and behaviors of data publishing, transmission and reception in distributed real-time systems, defines a data-centric publish-subscribe mechanism, and provides a shared platform. The DDS protocol defines the data transmitted in the distributed network as a topic, and defines the data generating and receiving objects as publishers and subscribers respectively, thus constituting the publish/subscribe transmission model of data.

Step S20, when the redundant computing power of the current edge device is greater than the computing power required to process the first data unit or the second data unit in the shared platform, receive and process the first data unit or the second data unit. data unit.

When the computing power of other edge devices is insufficient to process the detection data received by them, the detection data will also be divided to divide the detection data received by the other edge devices into second data units, and the detection data received by the other edge devices will be divided into second data units. The second data unit is sent to the sharing platform, and the second data unit is allocated to an edge device with redundant computing power through the sharing platform to process the second data unit.

The shared platform can store the first data unit or the second data unit sent by the edge device, and the redundant computing power of the current edge device is greater than that required for processing the first data unit or the second data unit in the shared platform When the computing power is calculated, the first data unit or the second data unit in the shared platform is received to process it.

In the edge data processing method provided by the embodiment of the present invention, when the redundant computing power of the current edge device is less than the computing power required for processing the detection data received by the current edge device, the detection data is divided into a plurality of first data unit, and send the first data unit to the shared platform; when the redundant computing power of the current edge device is greater than the computing power required to process the first data unit or the second data unit in the shared platform, receive and The first data unit or the second data unit is processed, so that the current edge device and the other edge devices can share computing power through the shared platform, thereby reducing the transmission resources occupied by the transmission of the detection data.

Optionally, when the redundant computing power of the current edge device is greater than the computing power required to process the detection data received by the current edge device, the detection data can also be divided into several first data through the current edge device. unit.

Optionally, after the received detection data is divided into several first data units or second data units, the computing power required to process each of the first data units or the second data units may be different, or can be the same.

Optionally, as shown in FIG. 3 , the step of dividing the detection data into several first data units includes:

Step S11, determining the segmentation rule of the detection data;

Step S12, dividing the detection data according to the dividing rule.

When the detection data is divided into the first data unit or the second data unit, there are different segmentation rules, and an appropriate segmentation rule can be selected according to the actual situation, and then the detection data is divided according to the segmentation rules. Usually, it is necessary to determine the segmentation rule of the detection data according to the type of the detection data.

Optionally, the division rule includes division according to time period and division according to basic unit composed of data. Segmentation by time period refers to the division according to the length of the time period in which the sensor collects the detection data. For example, it is divided according to the time period of 2 seconds. The divided first data unit or second data unit contains the data collected by the sensor within 2 seconds. Data; segmentation according to the basic unit composed of data refers to dividing the detection data into the smallest unit. For example, when the detection data is video data, the smallest unit of the video data is an image frame, and the image frame is the smallest unit. segmentation.

Optionally, after the step of dividing the detection data according to the dividing rule, the step further includes:

When the computing power required for processing the first data unit is greater than the preset computing power, the number of the first data unit obtained after dividing the detection data is increased.

Due to the limited computing power of the edge device, when the computing power required to process the first data unit or the second data unit is greater than the preset computing power, it means that the computing power required to process the first data unit or the second data unit is too high. is large, so that the computing power of the edge device is not enough to process the first data unit or the second data unit, it is necessary to increase the number of the first data unit or the second data unit obtained after dividing the detected data to reduce the processing computing power required by the first data unit or the second data unit, so that the edge device can receive the first data unit or the second data unit and process the first data unit or the second data unit, even if The current edge device can receive and process the first data unit or the second data unit. Wherein, the preset computing power is determined according to the computing power of the edge device.

Optionally, the method for processing edge data further includes:

obtaining a mapping relationship between the computing power required to process the first data unit or the second data unit and the CPU utilization of the edge device;

The redundant computing power of the current edge device is determined according to the mapping relationship.

The ability of the CPU of the edge device to process the first data unit or the second data unit is certain. Therefore, a relationship between the computing power required to process the first data unit or the second data unit and the CPU utilization of the edge device can be established. The mapping relationship, and the CPU utilization of the current edge device can be obtained through statistics. Therefore, when the CPU utilization of the current edge device is known, the redundancy of the current edge device can be obtained according to the mapping relationship. computing power.

As shown in FIG. 4 , based on the first embodiment, the edge data processing method provided by the second embodiment of the present invention further includes:

Step S30, determining the arrangement order of the first data unit in the detection time;

Step S40, correspondingly merge the processing results of the first data unit in the sharing platform according to the arrangement order.

After the edge device receives the detection data detected by the sensor, the detection data is divided into the first data unit or the second data unit, and then processed by other edge devices respectively. Therefore, the corresponding processing result obtained is for the first data unit or the second data unit. The processing result of the data unit or the second data unit needs to be combined with the processing results of the first data unit or the second data unit, and the first data unit or The processing result of the second data unit, thereby forming the processing result of the detection data, to achieve the purpose of processing the detection data.

After other edge devices process the first data unit or the second data unit and obtain a corresponding processing result, the processing result of the first data unit or the second data unit needs to be sent to the sharing platform. After the detection data is divided into the first data unit or the second data unit according to the collection time, the arrangement of the first data unit or the second data unit in the detection time forms an arrangement sequence (that is, The sequence of collecting each of the first data units or the second data units), so as to combine the corresponding processing results in the shared platform according to the arrangement sequence.

Optionally, after processing the first data unit or the second data unit, the current edge device and other edge devices will send the corresponding processing result to the sharing platform, so that the sharing platform can send all the data to the sharing platform. The corresponding processing results are merged to form a processing result for the detection data, and the merged processing result is sent to the edge device that receives the detection data.

Optionally, as shown in FIG. 5 , the step of correspondingly merging the processing results of the first data unit in the sharing platform according to the arrangement order includes:

Step S41, receiving the processing result of the first data unit in the sharing platform;

Step S42 , consolidate the processing result of the current edge device on the first data unit and the received processing result of the first data unit in the shared platform according to the arrangement order.

When the current edge device receives the detection data, the current edge device does not need to send the processing result of the first data unit to the sharing platform, but receives the processing result of the first data unit in the sharing platform, and passes the processing result of the first data unit to the sharing platform. The current edge device combines the processing result of the first data unit by the current edge device and the processing result of the first data unit received from the shared platform to form a detection data. process result.

Based on the first embodiment or the second embodiment, the edge data processing method provided by the third embodiment of the present invention further includes:

obtaining a submodel for processing the detection data;

The detection data is input into the sub-model, and the processing result of the detection data by the sub-model is sent to the sub-models on the other edge devices.

The deep neural network model has high requirements on the computing power of the edge device. When the computing power of the edge device is low, the deep neural network model may not be able to run on the edge device. The neural network model is segmented to segment the deep neural network model into multiple sub-models, each sub-model being an independent module. Load each of the divided sub-models into each edge device, and after the current edge device obtains the detection data, input the detection data into the sub-model, and after the sub-model processes the detection data, The current edge device sends the processing result of the detection data by the sub-model to other edge devices, and then processes the processing result of the detection data by the current edge device through the sub-model on the other edge device, Thus, the processing of the detection data by the deep neural network model is realized.

Optionally, the processing result of the detection data by the sub-model on the current edge device may be directly sent to the other edge devices; the detection data may also be processed by the sub-model on the current edge device. The processing result is sent to the sharing platform, and then sent to the other edge devices through the sharing platform.

In addition, an embodiment of the present invention further provides an edge device, the edge device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being The processor implements the steps of any one of the foregoing edge data processing methods when executed.

In addition, an embodiment of the present invention also provides a computer-readable storage medium, where a data processing program of an edge device is stored on the computer-readable storage medium, and when the data processing program of the edge device is executed by a processor, any of the above-mentioned steps are implemented. 1. The steps of the edge data processing method.

It should be noted that, herein, the terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or system comprising a series of elements includes not only those elements, It also includes other elements not expressly listed or inherent to such a process, method, article or system. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, method, article or system that includes the element.

The above-mentioned serial numbers of the embodiments of the present invention are only for description, and do not represent the advantages or disadvantages of the embodiments.

From the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform, and of course hardware can also be used, but in many cases the former is better implementation. Based on this understanding, the technical solutions of the present invention can be embodied in the form of software products in essence or the parts that make contributions to the prior art. The computer software products are stored in a storage medium (such as ROM/RAM) as described above. , magnetic disk, optical disk), including several instructions to make a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) execute the methods described in the various embodiments of the present invention.

The above are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the description and drawings of the present invention, or directly or indirectly applied in other related technical fields , are similarly included in the scope of patent protection of the present invention.

Claims (10)

1. a processing method of edge data, is characterized in that, the processing method of described edge data is applied to edge equipment, and the processing method of described edge data comprises the following steps: When the redundant computing power of the current edge device is less than the computing power required for processing the detection data received by the current edge device, the detection data is divided into several first data units, and the first data units are sent to bdfgh; When the redundant computing power of the current edge device is greater than the computing power required to process the first data unit or the second data unit in the shared platform, receive and process the first data unit or the second data unit, Wherein, the second data unit is data sent by other edge devices to the sharing platform. 2. The method for processing edge data according to claim 1, wherein the step of dividing the detected data into several first data units comprises: determining the segmentation rule of the detection data; The detection data is segmented according to the segmentation rule. 3 . The edge data processing method according to claim 2 , wherein the segmentation rules include segmentation according to time periods and segmentation according to basic units composed of data. 4 . 4. The method for processing edge data according to claim 2, wherein the step of dividing the detection data according to the dividing rule further comprises: When the computing power required for processing the first data unit is greater than the preset computing power, the number of the first data units obtained after dividing the detection data is increased. 5. The method for processing edge data according to claim 1, wherein the method for processing edge data further comprises: Obtain the mapping relationship between the computing power required to process the first data unit or the second data unit and the CPU utilization of the edge device; The redundant computing power of the current edge device is determined according to the mapping relationship. 6. The method for processing edge data according to claim 1, wherein the step of the method for processing edge data further comprises: determining the arrangement order of the first data unit in the detection time; The processing results of the first data unit in the shared platform are correspondingly merged according to the arrangement order. 7. The edge data processing method according to claim 6, wherein the step of correspondingly merging the processing results of the first data unit in the shared platform according to the arrangement sequence comprises: receiving a processing result of the first data unit in the sharing platform; The processing result of the current edge device on the first data unit and the received processing result of the first data unit in the shared platform are combined according to the arrangement order. 8. The method for processing edge data according to claim 1, wherein the step of the method for processing edge data further comprises: acquiring a sub-model for processing the detection data, wherein the sub-model acquired by the current edge device and the sub-model loaded on the other edge devices constitute a deep neural network model; The detection data is input into the sub-model, and the processing result of the detection data by the sub-model is sent to the sub-models on the other edge devices. 9. An edge device, characterized in that the edge device comprises: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being executed by the processor The steps of implementing the method for processing edge data according to any one of claims 1 to 8 when executed. 10. A computer-readable storage medium, characterized in that, a processing program for edge data is stored on the computer-readable storage medium, and the processing program for edge data is implemented as in claims 1 to 8 when executed by a processor. The steps of any one of the processing methods for edge data.

Images