Unmanned aerial vehicle-based mining violation identification method and system

Abstract

The invention discloses a mining violation identification method and system based on an unmanned aerial vehicle, the unmanned aerial vehicle is used for identifying a target area, the target area corresponding to the unmanned aerial vehicle sequentially comprises a first area, a second area and a third area from far to near, and the method comprises the following steps: the unmanned aerial vehicle obtains video data of the target area, extracting image data from the video data; determining position information of a target object in the target area according to the image data; according to the position information of the target object, when it is determined that the target object reaches a second area of the target area, a head image is intercepted from the image data; and inputting the image data into a violation recognition model for violation recognition, determining whether the target object wears a safety helmet or not, if the target object does not wear the safety helmet, uploading a head image to a server, and performing face recognition by the server according to the head image to determine the identity of the violation person. According to the scheme, the violation behavior can be identified more conveniently.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and system for identifying mining violations based on an unmanned aerial vehicle. Background Art

[0002] In mining and other scenarios, before workers enter the construction area, it is necessary to check whether the workers are wearing safety helmets and whether they comply with construction specifications. The existing solution is to use manual recognition, but this solution is inconvenient to identify violations. Summary of the invention

[0003] The present invention provides a mining violation identification method and system based on unmanned aerial vehicles, which can utilize unmanned aerial vehicles to identify workers to determine whether the workers are wearing safety helmets, thereby more conveniently identifying violation behaviors.

[0004] In order to solve the above-mentioned technical problems, the present invention is achieved as follows: In the first aspect, the present application provides a method for identifying mining violations based on drones, wherein the drone is used to identify a target area, and the target area corresponding to the drone includes a first area, a second area, and a third area from far to near, respectively. The method includes: the drone obtains video data of the target area and extracts image data from the video data; based on the image data, determines the position information of the target object in the target area; based on the position information of the target object, determines when the target object reaches the second area of ​​the target area, and captures a head image from the image data; inputs the image data into a violation recognition model for violation recognition, determines whether the target object is wearing a safety helmet, and if the target object is not wearing a safety helmet, uploads the head image to the server, and the server performs face recognition based on the head image to determine the identity of the violator.

[0005] Preferably, the drone uploads video data to a server, and the method further includes: determining, based on the location information of the target object, a first video segment, a second video segment and a third video segment corresponding to the video data of the target object, the first video segment corresponding to the first area, the second video segment corresponding to the second area, and the third video segment corresponding to the third area; the first video segment and the third video segment capture a first basic image at a first time interval, and compress the image behind the first basic image based on the first basic image to form first compressed data, so as to upload the first basic image and the first compressed data to the server, and the server performs video restoration based on the first basic image and the first compressed data; the second video segment captures a second basic image at a second time interval, and compresses the image behind the second basic image based on the second basic image to form second compressed data, so as to upload the second basic image and the second compressed data to the server, and the server performs video restoration based on the second basic image and the second compressed data; wherein the second time interval is shorter than the first time interval.

[0006] Preferably, the compressing the image behind the first basic image based on the first basic image to form the first compressed data includes: obtaining multiple images between the first basic image and the next first basic image to determine the first motion trajectory of the target object in the first area or the third area as the first compressed data, and uploading the first motion trajectory to the server, and the server performs video restoration based on the first basic image, the next first basic image and the first motion trajectory; the compressing the image behind the second basic image based on the second basic image to form the second compressed data includes: obtaining multiple images between the second basic image and the next second basic image to determine the second motion trajectory of the target object in the second area as the second compressed data, and uploading the second motion trajectory to the server, and the server performs video restoration based on the second basic image, the next second basic image and the second motion trajectory.

[0007] Preferably, the method also includes: the drone obtains environmental video data related to the target area at multiple acquisition positions in the target area, and uploads the video data and the acquisition positions to the server, the server performs modeling based on the environmental video data and the acquisition positions to obtain an environmental model of the target area, the environmental model includes a fixed object in the target area, the fixed object includes a plurality of fixed pixel points, the environmental video data includes a plurality of videos at different times and in different weather conditions, so as to generate multiple types of environmental models, and the multiple types of environmental models correspond to different times and different weather conditions; the step of uploading the first basic image or the second basic image includes: obtaining the target image to be uploaded, and dividing the target image into a first area including the target object and a second area not including the target object, the target image being the first basic image or the second basic image; based on the image of the second area, extracting pixel information corresponding to the fixed pixel points of the fixed object in the target area; uploading the image of the first area and the pixel information of the second area to the server, the server analyzes the type information corresponding to the second area based on the pixel information of the second area, and generates an image corresponding to the second area based on the type information and the environmental model, so as to restore the target image in combination with the image of the first area.

[0008] Preferably, the server determines multiple fixed shooting positions based on the environmental model, and the method also includes: when it is determined based on the image data that multiple target objects produce occlusion, determining the best shooting position of the drone in the target area from the fixed shooting positions, and controlling the drone to run to the best shooting position; uploading the position information of the best shooting position to the server to update the position of the drone in the server, so that the server performs video restoration based on the updated position of the drone; updating the first area, the second area and the third area based on the best shooting position, and obtaining video data of the target area through the drone for analysis and data upload.

[0009] Preferably, when the server determines multiple fixed shooting positions, it generates a reference image corresponding to the drone at the fixed shooting position, the reference image includes a reference map of fixed objects in the target area, and the server sends the fixed shooting position and the reference image to the drone. The method also includes: when the drone determines the best shooting position and runs to the best shooting position, the drone extracts the corresponding reference image and obtains the captured image through the drone; matching is performed based on the captured image and the reference image, if they are consistent, determining that the drone runs to the best shooting position, if they are inconsistent, determining an operation plan, and controlling the operation of the drone according to the operation plan.

[0010] Preferably, after establishing the environment model, the server determines the fixed position information of the fixed object and sends it to the drone. The determining the position information of the target object in the target area based on the image data includes: determining the first size information of the target object in the target area and the second size information of the fixed object based on the image data; determining the size ratio information based on the first size information and the second size information, and determining the position information of the target object based on the fixed position information of the fixed object.

[0011] In the second aspect, the present application provides a mining violation identification system based on a drone, characterized in that the drone is used to identify a target area, and the target area corresponding to the drone includes a first area, a second area and a third area from far to near, and the system includes: an image data acquisition module, which is used for the drone to acquire video data of the target area and extract image data from the video data; a position information acquisition module, which is used to determine the position information of the target object in the target area based on the image data; a head image acquisition module, which is used to determine, based on the position information of the target object, when the target object reaches the second area of ​​the target area, and to capture a head image from the image data; a head image recognition module, which is used to input the image data into a violation recognition model for violation recognition, determine whether the target object is wearing a safety helmet, and if the target object is not wearing a safety helmet, upload the head image to the server, and the server performs face recognition based on the head image to determine the identity of the violator.

[0012] In a third aspect, the present application provides an electronic device comprising: a memory and at least one processor; the memory is used to store computer-executable instructions; the at least one processor is used to execute the computer-executable instructions stored in the memory, so that the at least one processor executes the method described in the first aspect.

[0013] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0014] This application can be applied in scenarios where workers are identified to see if they have violated regulations, such as in mining scenarios to identify whether workers are wearing helmets. This solution can use drones to collect videos of areas that need to be monitored, and analyze them based on video images to extract the corresponding head images of workers, and identify whether workers are wearing helmets to analyze workers' violations. This solution can use drones to collect images, and identify workers' violations based on image analysis, and determine and record the identities of workers who violate regulations, which can more conveniently identify and record violations. Specifically, this solution can use drones to shoot video data of target objects (workers) in the target area, where if there is occlusion between workers, the drone's flight capability can be used to adjust the acquisition position to obtain the corresponding image. The target area photographed by the drone can be divided into the first area, the second area, and the third area from far to near. The second area is the area where the drone focuses, and a clear image can be obtained. After the drone collects video data, it can determine the location information of the target object in the target area based on the image data in the video; based on the location information of the target object, determine when the target object reaches the second area of ​​the target area, and capture the head image from the image data; input the image data into the violation recognition model for violation recognition, and determine whether the target object is wearing a helmet. If the target object is not wearing a helmet, upload the head image to the server, and the server performs face recognition based on the head image to determine the identity of the violator, and mark or issue a reminder based on the identity information to avoid violations. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 It is a flowchart of a method for identifying mining violations based on drones according to an embodiment of the present application; Figure 2 It is a structural diagram of a mining violation identification system based on drones according to an embodiment of the present application. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] This application can be applied in the scenario of identifying whether a worker has violated the rules, such as identifying whether a worker is wearing a helmet. This solution can use a drone to collect videos of the area that needs to be monitored, and analyze the video image to extract the corresponding head image of the worker, and identify whether the worker is wearing a helmet to analyze the worker's violation. This solution can use a drone to collect images, and identify the worker's violation based on image analysis, and determine and record the identity of the violating worker, which can more conveniently identify and record violations. Specifically, this solution can use a drone to shoot video data containing a target object (worker) in the target area. If there is occlusion between the workers, the drone's flight capability can be used to adjust the acquisition position to obtain the corresponding image. The target area photographed by the drone can be divided into a first area, a second area, and a third area from far to near. The second area is the area where the drone focuses, and a clear image can be obtained. After the drone collects video data, it can determine the location information of the target object in the target area based on the image data in the video; based on the location information of the target object, determine when the target object reaches the second area of ​​the target area, and capture the head image from the image data; input the image data into the violation recognition model for violation recognition, and determine whether the target object is wearing a helmet. If the target object is not wearing a helmet, upload the head image to the server, and the server performs face recognition based on the head image to determine the identity of the violator, and mark or issue a reminder based on the identity information to avoid violations.

[0018] Specifically, the embodiment of the present application provides a method for identifying mining violations based on a drone, wherein the drone is used to identify a target area, and the target area corresponding to the drone includes a first area, a second area, and a third area in order from far to near, such as Figure 1 As shown, the method includes: Step 102: The drone acquires video data of the target area and extracts image data from the video data.

[0019] Step 104: Determine the position information of the target object in the target area based on the image data.

[0020] Step 106: Determine, based on the position information of the target object, to capture a head image from the image data when the target object reaches a second area of ​​the target area.

[0021] Step 108: input the image data into the violation recognition model to perform violation recognition and determine whether the target object is wearing a helmet. If the target object is not wearing a helmet, upload the head image to the server, and the server performs face recognition based on the head image to determine the identity of the violator.

[0022] This application can be applied in the scenario of identifying whether a worker has violated the rules, such as identifying whether a worker is wearing a helmet. This solution can use a drone to collect videos of the area that needs to be monitored, and analyze the video image to extract the corresponding head image of the worker, and identify whether the worker is wearing a helmet to analyze the worker's violation. This solution can use a drone to collect images, and identify the worker's violation based on image analysis, and determine and record the identity of the violating worker, which can more conveniently identify and record violations. Specifically, this solution can use a drone to shoot video data containing a target object (worker) in the target area. If there is occlusion between the workers, the drone's flight capability can be used to adjust the acquisition position to obtain the corresponding image. The target area photographed by the drone can be divided into a first area, a second area, and a third area from far to near. The second area is the area where the drone focuses, and a clear image can be obtained. After the drone collects video data, it can determine the location information of the target object in the target area based on the image data in the video; based on the location information of the target object, determine when the target object reaches the second area of ​​the target area, and capture the head image from the image data; input the image data into the violation recognition model for violation recognition, and determine whether the target object is wearing a helmet. If the target object is not wearing a helmet, upload the head image to the server, and the server performs face recognition based on the head image to determine the identity of the violator, and mark or issue a reminder based on the identity information to avoid violations.

[0023] The drone in this solution can transmit the collected video data to the server for storage. Different data can be compressed in different ways before uploading. For low-definition data, less data can be uploaded, and for high-definition data, more data can be uploaded to be stored on the server. Specifically, as an optional embodiment, the drone uploads video data to the server, and the method further includes: determining the first video segment, the second video segment and the third video segment corresponding to the video data of the target object according to the location information of the target object, the first video segment corresponding to the first area, the second video segment corresponding to the second area, and the third video segment corresponding to the third area; the first video segment and the third video segment collect the first basic image at a first time interval, and compress the image after the first basic image based on the first basic image to form a first compressed data, so as to upload the first basic image and the first compressed data to the server, and the server performs video restoration based on the first basic image and the first compressed data; the second video segment collects the second basic image at a second time interval, and compresses the image after the second basic image based on the second basic image to form a second compressed data, so as to upload the second basic image and the second compressed data to the server, and the server performs video restoration based on the second basic image and the second compressed data; wherein the second time interval is shorter than the first time interval. The video segments can be divided into two categories, the clearer video segments collect the basic image and upload the basic image and the compressed image at a shorter time interval, and the blurred video segments collect the basic image and upload it at a longer time interval.

[0024] In this solution, since the area monitored by the drone is fixed, only the motion trajectory of the target object can be uploaded for the compressed data, and there is no need to collect other data for processing. Specifically, as an optional embodiment, the image after the first basic image is compressed based on the first basic image to form the first compressed data, including: obtaining multiple images between the first basic image and the next first basic image to determine the first motion trajectory of the target object in the first area or the third area as the first compressed data, and uploading the first motion trajectory to the server, and the server performs video restoration based on the first basic image, the next first basic image and the first motion trajectory; the image after the second basic image is compressed based on the second basic image to form the second compressed data, including: obtaining multiple images between the second basic image and the next second basic image to determine the second motion trajectory of the target object in the second area as the second compressed data, and uploading the second motion trajectory to the server, and the server performs video restoration based on the second basic image, the next second basic image and the second motion trajectory.

[0025] The drone can collect the surrounding environment video in advance and upload it to the server for modeling. The server restores the image based on the motion trajectory and the pre-modeled environment model. In addition, since the environment of the target area is fixed, this solution can only obtain the pixel values ​​of the fixed pixels of the fixed objects in the target area and upload them to analyze the environment type corresponding to the environment model to restore the image. Specifically, as an optional embodiment, the method also includes: the drone obtains environmental video data related to the target area at multiple acquisition positions in the target area, and uploads the video data and the acquisition positions to the server, the server performs modeling based on the environmental video data and the acquisition positions to obtain an environmental model of the target area, the environmental model includes a fixed object in the target area, the fixed object includes a plurality of fixed pixel points, the environmental video data includes a plurality of videos at different times and in different weather conditions, so as to generate multiple types of environmental models, and the multiple types of environmental models correspond to different times and different weather conditions; the step of uploading the first basic image or the second basic image includes: obtaining the target image to be uploaded, and dividing the target image into a first area including the target object and a second area not including the target object, the target image being the first basic image or the second basic image; based on the image of the second area, extracting pixel information corresponding to the fixed pixel points of the fixed object in the target area; uploading the image of the first area and the pixel information of the second area to the server, the server analyzes the type information corresponding to the second area based on the pixel information of the second area, and generates an image corresponding to the second area based on the type information and the environmental model, so as to restore the target image in combination with the image of the first area.

[0026] If there is occlusion between workers, there may be recognition errors. The flying ability of the drone can be used to adjust the position and collect images. In addition, this solution can set several fixed positions for the drone to analyze and locate. Specifically, as an optional embodiment, the server determines multiple fixed shooting positions based on the environment model. The method also includes: when it is determined that multiple target objects are blocked based on the image data, the best shooting position of the drone in the target area is determined from the fixed shooting positions, and the drone is controlled to run to the best shooting position; uploading the position information of the best shooting position to the server to update the position of the drone in the server, so that the server performs video restoration based on the updated position of the drone; updating the first area, the second area and the third area based on the best shooting position, and obtaining the video data of the target area through the drone for analysis and data upload. The drone may have deviations during operation, and the surrounding environment of the drone is fixed. Therefore, the server can determine a reference image containing fixed objects based on the environment model. If the image collected by the drone is consistent with the reference image, the drone has reached the specified position. Specifically, as an optional embodiment, when the server determines multiple fixed shooting positions, it generates a reference image corresponding to the drone at the fixed shooting position, the reference image includes a reference map of the fixed objects in the target area, and the server sends the fixed shooting position and the reference image to the drone. The method also includes: when the drone determines the best shooting position and runs to the best shooting position, the drone extracts the corresponding reference image and obtains the captured image through the drone; matches the captured image and the reference image, if they are consistent, it is determined that the drone runs to the best shooting position, if they are inconsistent, an operation plan is determined, and the operation of the drone is controlled according to the operation plan.

[0027] The drone can determine the position of the target object through depth analysis. The depth analysis can analyze the distance between the target object and the drone by analyzing the image. The server can analyze the depth and size of each fixed object through multiple images, and then send it to the drone for processing. Specifically, as an optional embodiment, the server determines the fixed position information of the fixed object after establishing the environment model and sends it to the drone. The position information of the target object in the target area is determined based on the image data, including: determining the first size information of the target object in the target area and the second size information of the fixed object based on the image data; determining the size ratio information based on the first size information and the second size information, and determining the position information of the target object based on the fixed position information of the fixed object.

[0028] On the basis of the above embodiments, the embodiments of the present application further provide a mining violation identification system based on drones, wherein the drones are used to identify target areas, and the target areas corresponding to the drones include a first area, a second area, and a third area in order from far to near, such as Figure 2 As shown, the system comprises: The image data acquisition module 202 is used for the drone to acquire video data of the target area and extract image data from the video data.

[0029] The position information acquisition module 204 is used to determine the position information of the target object in the target area according to the image data.

[0030] The head image acquisition module 206 is used to determine, based on the position information of the target object, to capture the head image from the image data when the target object reaches the second area of ​​the target area.

[0031] The head image recognition module 208 is used to input the image data into the violation recognition model for violation recognition and determine whether the target object is wearing a helmet. If the target object is not wearing a helmet, the head image is uploaded to the server, and the server performs face recognition based on the head image to determine the identity of the violator.

[0032] The implementation of the embodiment of the present application is similar to the implementation of the embodiment of the method described above. The specific implementation can refer to the specific implementation of the method described above, which will not be repeated here.

[0033] This application can be applied in the scenario of identifying whether a worker has violated the rules, such as identifying whether a worker is wearing a helmet. This solution can use a drone to collect videos of the area that needs to be monitored, and analyze the video image to extract the corresponding head image of the worker, and identify whether the worker is wearing a helmet to analyze the worker's violation. This solution can use a drone to collect images, and identify the worker's violation based on image analysis, and determine and record the identity of the violating worker, which can more conveniently identify and record violations. Specifically, this solution can use a drone to shoot video data containing a target object (worker) in the target area. If there is occlusion between the workers, the drone's flight capability can be used to adjust the acquisition position to obtain the corresponding image. The target area photographed by the drone can be divided into a first area, a second area, and a third area from far to near. The second area is the area where the drone focuses, and a clear image can be obtained. After the drone collects video data, it can determine the location information of the target object in the target area based on the image data in the video; based on the location information of the target object, determine when the target object reaches the second area of ​​the target area, and capture the head image from the image data; input the image data into the violation recognition model for violation recognition, and determine whether the target object is wearing a helmet. If the target object is not wearing a helmet, upload the head image to the server, and the server performs face recognition based on the head image to determine the identity of the violator, and mark or issue a reminder based on the identity information to avoid violations.

[0034] Based on the above embodiments, the present application also provides an electronic device, comprising: a memory and at least one processor; the memory is used to store computer execution instructions; the at least one processor is used to execute the computer execution instructions stored in the memory, so that the at least one processor executes the method described in the above embodiments.

[0035] The embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, each process of the above-mentioned data processing method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0036] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0037] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0038] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0039] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0040] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0041] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0042] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. According to the definition in this article, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

Claims

1. A method for identifying mining violations based on drones, characterized in that: The drone is used to identify a target area, and the target area corresponding to the drone includes a first area, a second area, and a third area in order from far to near. The method includes: The drone acquires video data of the target area and extracts image data from the video data; Determine the position information of the target object in the target area based on the image data; Determining, based on the position information of the target object, to intercept the head image from the image data when the target object reaches a second area of ​​the target area; The image data is input into the violation recognition model for violation recognition to determine whether the target object is wearing a helmet. If the target object is not wearing a helmet, the head image is uploaded to the server, and the server performs face recognition based on the head image to determine the identity of the violator.

2. The method according to claim 1, characterized in that The drone uploads video data to the server, and the method further includes: Determine, according to the location information of the target object, a first video segment, a second video segment, and a third video segment corresponding to the video data of the target object, wherein the first video segment corresponds to the first area, the second video segment corresponds to the second area, and the third video segment corresponds to the third area; The first video segment and the third video segment capture a first basic image at a first time interval, and compress an image behind the first basic image based on the first basic image to form first compressed data, so as to upload the first basic image and the first compressed data to a server, and the server performs video restoration according to the first basic image and the first compressed data; The second video segment captures a second basic image at a second time interval, and compresses the image behind the second basic image based on the second basic image to form second compressed data, so as to upload the second basic image and the second compressed data to the server, and the server restores the video based on the second basic image and the second compressed data; wherein the second time interval is shorter than the first time interval.

3. The method according to claim 2, characterized in that The step of compressing the image after the first basic image based on the first basic image to form first compressed data includes: Acquire multiple images between the first basic image and the next first basic image to determine a first motion trajectory of the target object in the first area or the third area as first compressed data, and upload the first motion trajectory to the server, and the server performs video restoration based on the first basic image, the next first basic image and the first motion trajectory; The step of compressing the image after the second basic image based on the second basic image to form second compressed data includes: A plurality of images between the second basic image and the next second basic image are obtained to determine a second motion trajectory of the target object in the second area as second compressed data, and the second motion trajectory is uploaded to the server, which performs video restoration based on the second basic image, the next second basic image and the second motion trajectory.

4. The method according to claim 3, characterized in that: The method further comprises: The drone obtains environmental video data related to the target area at multiple acquisition positions in the target area, and uploads the video data and the acquisition positions to the server. The server performs modeling based on the environmental video data and the acquisition positions to obtain an environmental model of the target area. The environmental model includes fixed objects in the target area, and the fixed objects include multiple fixed pixels. The environmental video data includes multiple videos at different times and in different weather conditions to generate multiple types of environmental models. The multiple types of environmental models correspond to different times and different weather conditions. The step of uploading the first base image or the second base image comprises: Acquire a target image to be uploaded, and divide the target image into a first area containing a target object and a second area not containing the target object, the target image being the first basic image or the second basic image; Based on the image of the second area, pixel information corresponding to fixed pixel points of fixed objects in the target area is extracted; the image of the first area and the pixel information of the second area are uploaded to the server, and the server analyzes the type information corresponding to the second area based on the pixel information of the second area, and generates an image corresponding to the second area based on the type information and the environmental model, so as to restore the target image in combination with the image of the first area.

5. The method according to claim 4, characterized in that The server determines a plurality of fixed shooting positions according to the environment model, and the method further includes: When it is determined based on the image data that multiple target objects are blocked, the best shooting position of the drone in the target area is determined from the fixed shooting positions, and the drone is controlled to move to the best shooting position; Upload the location information of the best shooting position to the server to update the position of the drone in the server, so that the server can restore the video according to the updated position of the drone; The first area, the second area, and the third area are updated according to the best shooting position, and the video data of the target area is obtained through the drone for analysis and data upload.

6. The method according to claim 5, characterized in that When the server determines a plurality of fixed shooting positions, a reference image corresponding to the drone at the fixed shooting position is generated, the reference image including a reference image of a fixed object in the target area, and the server sends the fixed shooting position and the reference image to the drone. The method further includes: In the process of determining the best shooting position and moving to the best shooting position, the drone extracts the corresponding reference image and obtains the collected image through the drone; The collected image and the reference image are matched. If they are consistent, the drone is determined to move to the best shooting position. If they are inconsistent, an operation plan is determined and the drone is controlled according to the operation plan.

7. The method according to claim 6, characterized in that After the server establishes the environment model, it determines the fixed position information of the fixed object and sends it to the drone. The determining the position information of the target object in the target area based on the image data includes: Determine first size information of a target object and second size information of a fixed object in a target area according to the image data; The size ratio information is determined according to the first size information and the second size information, and the position information of the target object is determined according to the fixed position information of the fixed object.

8. A mining violation identification system based on drones, characterized in that: The drone is used to identify the target area. The target area corresponding to the drone includes a first area, a second area and a third area in order from far to near. The system includes: An image data acquisition module is used by the drone to acquire video data of a target area and extract image data from the video data; A position information acquisition module, used to determine the position information of the target object in the target area based on the image data; A head image acquisition module, used to determine, based on the position information of the target object, when the target object reaches the second area of ​​the target area, to intercept the head image from the image data; The head image recognition module is used to input image data into the violation recognition model for violation recognition and determine whether the target object is wearing a helmet. If the target object is not wearing a helmet, the head image is uploaded to the server, and the server performs face recognition based on the head image to determine the identity of the violator.

9. An electronic device, characterized in that: include: memory and at least one processor; The memory is used to store computer-executable instructions; The at least one processor is configured to execute computer-executable instructions stored in the memory, so that the at least one processor performs the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored on a computer-readable storage medium, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images