A group target tracking method and device, computer equipment and storage medium

Abstract

Embodiments of the present application disclose a kind of group target tracking method, device, computer equipment and storage medium, involve visual imaging technical field.The method comprises: obtaining the target image information of tracking target by distributed imaging equipment;Target image information is input into the deep learning model after training, to obtain corresponding target detection image, and the measured position information of each tracking target is obtained according to target detection image;According to the measured position information and the internal parameter matrix of distributed imaging equipment, the two-dimensional position information of each tracking target is calculated;According to two-dimensional position information and the external parameter matrix of distributed imaging equipment, the three-dimensional position information of each tracking target is calculated;According to two-dimensional position information and three-dimensional position information, the individual position coordinates of each tracking target are determined.Thereby better solve the problem of mutual occlusion between individuals, the requirement of less computing power, also avoid the problems such as high cost, high energy consumption, poor portability and narrow application range.

Description

Technical Field

[0001] The present invention relates to the field of visual imaging technology, and in particular to a method, apparatus, computer device and storage medium for tracking a group of targets. Background Technology

[0002] Current visual drone swarm monitoring systems mostly use a single video stream, which is difficult to solve the occlusion problem to some extent, is greatly affected by the environment, and requires complex deep learning networks, thus consuming a lot of computing resources and time. On the other hand, current non-visual systems require the use of radar, radio frequency and radio equipment, and also have drawbacks such as narrow monitoring range, limited monitoring content, high signal source power, poor portability, high energy consumption and high cost, and cannot effectively solve the occlusion problem. Summary of the Invention

[0003] This invention provides a method, apparatus, computer device, and storage medium for tracking groups of targets, in order to solve the problems of occlusion, severe environmental influence, and high computing power requirements of vision systems, as well as the problems of high cost, high energy consumption, poor portability, and narrow applicability of non-vision systems.

[0004] In a first aspect, embodiments of the present invention provide a method for tracking a group of targets, the method comprising:

[0005] Acquire target image information of the tracked target using distributed imaging devices;

[0006] The target image information is input into the trained deep learning model to obtain the corresponding target detection image, and the measured position information of each tracked target is obtained based on the target detection image.

[0007] The two-dimensional position information of each of the tracked targets is calculated based on the measured position information and the intrinsic parameter matrix of the distributed imaging device;

[0008] The three-dimensional position information of each of the tracked targets is calculated based on the two-dimensional position information and the extrinsic parameter matrix of the distributed imaging device;

[0009] The individual position coordinates of each of the tracking targets are determined based on the two-dimensional position information and the three-dimensional position information.

[0010] Optionally, after determining the individual position coordinates of each tracking target based on the two-dimensional position information and the three-dimensional position information, the method further includes:

[0011] Based on the individual location coordinates, three-dimensional video and / or three-dimensional images are reconstructed for each of the tracked targets.

[0012] Optionally, before calculating the three-dimensional position information of each of the tracking targets based on the two-dimensional position information and the extrinsic matrix of the distributed imaging device, the method further includes:

[0013] The viewing angle of the distributed imaging device is adjusted according to the two-dimensional position information so that the tracking target is located in the center of the field of view of the distributed imaging device.

[0014] Optionally, adjusting the viewing angle of the distributed imaging device based on the two-dimensional position information includes:

[0015] The viewing angle of the distributed imaging device is adjusted using a preset gimbal control algorithm developed based on PID control.

[0016] Optionally, obtaining the measured position information of each of the tracked targets based on the target detection image includes:

[0017] The measured location information is obtained based on the Kalman filter algorithm.

[0018] Optionally, before inputting the target image information into the trained deep learning model, the method further includes:

[0019] The deep learning model is trained using a convolutional neural network algorithm.

[0020] Optionally, the distributed imaging device includes a self-zoom camera gimbal.

[0021] Secondly, embodiments of the present invention also provide a group target tracking device, the device comprising:

[0022] The target image information acquisition module is used to acquire target image information of the tracked target through a distributed imaging device.

[0023] The measured location information acquisition module is used to input the target image information into the trained deep learning model to obtain the corresponding target detection image, and to obtain the measured location information of each of the tracked targets based on the target detection image;

[0024] A two-dimensional position information calculation module is used to calculate the two-dimensional position information of each of the tracking targets based on the measured position information and the intrinsic parameter matrix of the distributed imaging device;

[0025] A three-dimensional position information calculation module is used to calculate the three-dimensional position information of each of the tracking targets based on the two-dimensional position information and the external parameter matrix of the distributed imaging device;

[0026] The individual position coordinate determination module is used to determine the individual position coordinates of each of the tracking targets based on the two-dimensional position information and the three-dimensional position information.

[0027] Thirdly, embodiments of the present invention also provide a computer device, the computer device comprising:

[0028] One or more processors;

[0029] Memory, used to store one or more programs;

[0030] When the one or more programs are executed by the one or more processors, the one or more processors implement the group target tracking method provided in any embodiment of the present invention.

[0031] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the group target tracking method provided in any embodiment of the present invention.

[0032] This invention provides a method for tracking a group of targets. First, target image information of the targets is acquired using a distributed imaging device. Then, this target image information is input into a trained deep learning model to obtain corresponding target detection images. Based on these detection images, the measured position information of each tracked target is obtained. Next, the two-dimensional and three-dimensional position information of each tracked target is calculated using the intrinsic and extrinsic parameter matrices of the distributed imaging device. Finally, the individual position coordinates of each tracked target are determined based on the obtained two-dimensional and three-dimensional position information. The group target tracking method provided by this invention allows for observation from multiple perspectives through a distributed layout. When imaging devices at different locations are aligned with the same group of targets, the positions of each tracked target in the group can be calculated by combining the images obtained from the poses of each imaging device. This better solves the problem of mutual occlusion between individuals. Compared to existing vision systems, it requires less computing power, and compared to existing non-vision systems, it avoids problems such as high cost, high energy consumption, poor portability, and narrow applicability. It also broadens the scope of application for monitoring, enabling it to be used primarily for monitoring drone swarms, while providing a wider range of scenarios for monitoring swarm targets, thus offering a more economical and applicable swarm target monitoring solution for more users. Attached Figure Description

[0033] Figure 1 A flowchart of a group target tracking method provided in Embodiment 1 of the present invention;

[0034] Figure 2 This is a schematic diagram of the structure of the group target tracking device provided in Embodiment 2 of the present invention;

[0035] Figure 3 This is a schematic diagram of the structure of a computer device provided in Embodiment 3 of the present invention. Detailed Implementation

[0036] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0037] Before discussing the exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the steps as sequential processes, many of these steps can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the steps can be rearranged. The process can be terminated when its operation is complete, but may also have additional steps not included in the figures. The process can correspond to a method, function, procedure, subroutine, subroutine, etc.

[0038] Example 1

[0039] Figure 1 This is a flowchart of a group target tracking method provided in Embodiment 1 of the present invention. This embodiment is applicable to tracking and monitoring various group targets (such as drones). The method can be executed by the group target tracking device provided in this embodiment, which can be implemented in hardware and / or software, and is generally integrated into a computer device, specifically a server. Figure 1 As shown, the specific steps include the following:

[0040] S11. Acquire target image information of the tracked target through distributed imaging equipment.

[0041] S12. Input the target image information into the trained deep learning model to obtain the corresponding target detection image, and obtain the measured position information of each of the tracked targets based on the target detection image.

[0042] S13. Calculate the two-dimensional position information of each of the tracking targets based on the measured position information and the intrinsic parameter matrix of the distributed imaging device.

[0043] S14. Calculate the three-dimensional position information of each of the tracking targets based on the two-dimensional position information and the extrinsic parameter matrix of the distributed imaging device.

[0044] S15. Determine the individual position coordinates of each of the tracking targets based on the two-dimensional position information and the three-dimensional position information.

[0045] Specifically, a certain number of imaging devices can be pre-distributed, and optionally, the distributed imaging devices include self-focusing camera gimbals. When a tracking target appears in the field of view of the distributed imaging devices, target image information of the tracking target can be obtained by capturing images of the target. The number of tracking targets can be multiple. The obtained target image information can then be input into a trained deep learning model, which can process and obtain target detection images corresponding to each target image for subsequent processing. Optionally, before inputting the target image information into the trained deep learning model, the model is further trained using a convolutional neural network algorithm to obtain the desired trained deep learning model. After obtaining the target detection images, the measured position information of each tracking target can be determined in the target detection images. Optionally, obtaining the measured position information of each tracking target based on the target detection images includes obtaining the measured position information based on a Kalman filter algorithm. Specifically, the target detection images can be combined with an algorithm based on Kalman filtering to track the tracking targets, thereby obtaining the measured position information of each tracking target at the current moment. After obtaining the measured position information, the measured position information of each tracking target at the current moment can be combined with the intrinsic parameter matrix of the distributed imaging device to calculate the two-dimensional position information of each tracking target. Then, the two-dimensional position information can be solved in conjunction with the extrinsic parameter matrix of the distributed imaging device to obtain the three-dimensional position information of each tracking target. Subsequently, the individual position coordinates of each tracking target can be calculated through the target matching algorithm, specifically obtaining the two-dimensional and three-dimensional coordinates of each tracking target.

[0046] Based on the above technical solution, optionally, after determining the individual position coordinates of each tracking target according to the two-dimensional position information and the three-dimensional position information, the method further includes: reconstructing three-dimensional video and / or three-dimensional images of each tracking target based on the individual position coordinates, thereby realizing the visualization of each tracking target and more intuitively and three-dimensionally reflecting the position and movement information of the group of targets. Simultaneously, the location data information of each tracking target can also be output for user use, such as obtaining more accurate information like the number of groups based on this location data information.

[0047] Based on the above technical solution, optionally, before calculating the three-dimensional position information of each tracking target according to the two-dimensional position information and the extrinsic parameter matrix of the distributed imaging device, the method further includes: adjusting the viewing angle of the distributed imaging device according to the two-dimensional position information so that the tracking target is located in the center of the field of view of the distributed imaging device. Optionally, adjusting the viewing angle of the distributed imaging device according to the two-dimensional position information includes: adjusting the viewing angle of the distributed imaging device using a preset gimbal control algorithm developed based on PID control.

[0048] Specifically, traditional solutions use fixed-focal-length, fixed-extrinsic-parameter cameras to acquire video information, and cannot achieve extrinsic-parameter control for visual cameras. The tracking window is short, and it cannot consistently track and detect multiple targets. Taking a zoom camera gimbal as an example, this method combines two-dimensional position information with a preset gimbal control algorithm developed based on PID control to adjust the gimbal's rotation, ensuring the center of the tracked target is in the center of the camera's field of view, and increasing the adjustability of distributed visual extrinsic parameters. For zoomable camera gimbals capable of acquiring real-time attitude, by developing recognition and tracking algorithms, suspicious groups in the image can be automatically identified, and their relative positions within the image can be determined. Based on this, the focal length and gimbal angle can be adjusted, making the group of targets magnified and appear in the center of the field of view, improving the resolution of suspicious groups in the image and significantly increasing the accuracy of target identification. Furthermore, the collaboration of various gimbals ensures that cameras at different locations can be aimed at the same target, achieving matching of information obtained from various points in the distributed gimbal camera network. By forming a network of multiple distributed gimbals, multi-angle observation of the same group of targets can be achieved. When a tracked target appears in the field of view of a camera at a certain point, other cameras may not necessarily see the target. However, based on the information obtained from the points that have already detected the target, the approximate location can be estimated, and other cameras can be adjusted to rotate towards that approximate location, thereby achieving coordinated detection by various pan-tilt units. At the same time, when multiple points observe the tracked target, it can be analyzed whether they are the same target, thus achieving collaborative monitoring of a group of targets in a wide area.

[0049] The technical solution provided by this invention first acquires target image information of the tracked target through a distributed imaging device. Then, this target image information is input into a trained deep learning model to obtain corresponding target detection images. Based on these target detection images, the measured position information of each tracked target is obtained. Next, the two-dimensional and three-dimensional position information of each tracked target is calculated by combining the intrinsic and extrinsic parameter matrices of the distributed imaging device. Finally, the individual position coordinates of each tracked target can be determined based on the obtained two-dimensional and three-dimensional position information. The distributed layout enables observation from multiple perspectives. When imaging devices at different locations are aligned with the same group of targets, the position of each tracked target in the group can be calculated by combining the images obtained from the poses of each imaging device. This better solves the problem of mutual occlusion between individuals. Compared to existing vision systems, it requires less computing power. Compared to existing non-vision systems, it avoids problems such as high cost, high energy consumption, poor portability, and narrow applicability. It also broadens the scope of monitoring applications, providing a wider range of group target monitoring scenarios while primarily using it for monitoring drone swarms, offering a more economical and applicable group target monitoring solution for more users.

[0050] Example 2

[0051] Figure 2 This is a schematic diagram of the structure of a group target tracking device provided in Embodiment 2 of the present invention. This device can be implemented in hardware and / or software, and is generally integrated into a computer device to execute the group target tracking method provided in any embodiment of the present invention. Figure 2 As shown, the device includes:

[0052] Target image information acquisition module 21 is used to acquire target image information of the tracked target through a distributed imaging device;

[0053] The measured location information acquisition module 22 is used to input the target image information into the trained deep learning model to obtain the corresponding target detection image, and to obtain the measured location information of each of the tracked targets based on the target detection image;

[0054] The two-dimensional position information calculation module 23 is used to calculate the two-dimensional position information of each of the tracking targets based on the measured position information and the intrinsic parameter matrix of the distributed imaging device.

[0055] The three-dimensional position information calculation module 24 is used to calculate the three-dimensional position information of each of the tracking targets based on the two-dimensional position information and the external parameter matrix of the distributed imaging device.

[0056] The individual position coordinate determination module 25 is used to determine the individual position coordinates of each of the tracking targets based on the two-dimensional position information and the three-dimensional position information.

[0057] The technical solution provided by this invention first acquires target image information of the tracked target through a distributed imaging device. Then, this target image information is input into a trained deep learning model to obtain corresponding target detection images. Based on these target detection images, the measured position information of each tracked target is obtained. Next, the two-dimensional and three-dimensional position information of each tracked target is calculated by combining the intrinsic and extrinsic parameter matrices of the distributed imaging device. Finally, the individual position coordinates of each tracked target can be determined based on the obtained two-dimensional and three-dimensional position information. The distributed layout enables observation from multiple perspectives. When imaging devices at different locations are aligned with the same group of targets, the position of each tracked target in the group can be calculated by combining the images obtained from the poses of each imaging device. This better solves the problem of mutual occlusion between individuals. Compared to existing vision systems, it requires less computing power. Compared to existing non-vision systems, it avoids problems such as high cost, high energy consumption, poor portability, and narrow applicability. It also broadens the scope of monitoring applications, providing a wider range of group target monitoring scenarios while primarily using it for monitoring drone swarms, offering a more economical and applicable group target monitoring solution for more users.

[0058] Based on the above technical solution, optionally, the tracking device for the group of targets also includes:

[0059] The three-dimensional image reconstruction module is used to reconstruct three-dimensional video and / or three-dimensional images of each of the tracked targets based on the individual position coordinates of each of the tracked targets according to the individual position coordinates after the individual position coordinates of each of the tracked targets are determined according to the two-dimensional position information and the three-dimensional position information.

[0060] Based on the above technical solution, optionally, the tracking device for the group of targets also includes:

[0061] An imaging device perspective adjustment module is used to adjust the perspective of the distributed imaging device according to the two-dimensional position information before calculating the three-dimensional position information of each of the tracked targets based on the two-dimensional position information and the extrinsic matrix of the distributed imaging device, so that the tracked target is located in the center of the field of view of the distributed imaging device.

[0062] Based on the above technical solution, optionally, the imaging device viewing angle adjustment module is specifically used for:

[0063] The viewing angle of the distributed imaging device is adjusted using a preset gimbal control algorithm developed based on PID control.

[0064] Based on the above technical solution, optionally, the measured location information acquisition module 22 is specifically used for:

[0065] The measured location information is obtained based on the Kalman filter algorithm.

[0066] Based on the above technical solution, optionally, the tracking device for the group of targets also includes:

[0067] The model training module is used to train the deep learning model using a convolutional neural network algorithm before inputting the target image information into the trained deep learning model.

[0068] Optionally, based on the above technical solution, the distributed imaging device may include a self-zoom camera gimbal.

[0069] The target tracking device provided in the embodiments of the present invention can execute the target tracking method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0070] It is worth noting that in the embodiments of the above-mentioned group target tracking device, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0071] Example 3

[0072] Figure 3 This is a schematic diagram of the structure of a computer device provided in Embodiment 3 of the present invention, showing a block diagram of an exemplary computer device suitable for implementing the embodiments of the present invention. Figure 3 The computer device shown is merely an example and should not be construed as limiting the functionality or scope of the embodiments of the present invention. Figure 3 As shown, the computer device includes a processor 31, a memory 32, an input device 33, and an output device 34; the number of processors 31 in the computer device can be one or more. Figure 3 Taking a processor 31 as an example, the processor 31, memory 32, input device 33, and output device 34 in a computer device can be connected via a bus or other means. Figure 3 Taking the example of a connection between China and Israel via a bus.

[0073] The memory 32, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the group target tracking method in this embodiment of the invention (e.g., the target image information acquisition module 21, the measured position information acquisition module 22, the two-dimensional position information calculation module 23, the three-dimensional position information calculation module 24, and the individual position coordinate determination module 25 in the group target tracking device). The processor 31 executes various functional applications and data processing of the computer device by running the software programs, instructions, and modules stored in the memory 32, thereby realizing the above-mentioned group target tracking method.

[0074] The memory 32 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 32 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, the memory 32 may further include memory remotely located relative to the processor 31, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0075] Input device 33 can be used to acquire target image information of the tracked target, and to generate key signal inputs related to user settings and function control of the computer device. Output device 34 can be used to output the final determined position data, etc.

[0076] Example 4

[0077] Embodiment 4 of the present invention also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a method for tracking a group of targets, the method comprising:

[0078] Acquire target image information of the tracked target using distributed imaging devices;

[0079] The target image information is input into the trained deep learning model to obtain the corresponding target detection image, and the measured position information of each tracked target is obtained based on the target detection image.

[0080] The two-dimensional position information of each of the tracked targets is calculated based on the measured position information and the intrinsic parameter matrix of the distributed imaging device;

[0081] The three-dimensional position information of each of the tracked targets is calculated based on the two-dimensional position information and the extrinsic parameter matrix of the distributed imaging device;

[0082] The individual position coordinates of each of the tracking targets are determined based on the two-dimensional position information and the three-dimensional position information.

[0083] Storage media can be any type of memory device or storage device. The term "storage media" is intended to include: mounting media, such as CD-ROMs, floppy disks, or magnetic tape devices; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (e.g., hard disks or optical storage); registers or other similar types of memory elements. Storage media may also include other types of memory or combinations thereof. Furthermore, storage media may reside in a computer system in which the program is executed, or may reside in a different second computer system connected to the computer system via a network (such as the Internet). The second computer system can provide program instructions to the computer for execution. The term "storage media" can include two or more storage media that may reside in different locations (e.g., in different computer systems connected via a network). Storage media may store program instructions (e.g., specifically implemented as a computer program) that can be executed by one or more processors.

[0084] Of course, the computer-executable instructions provided in the embodiments of the present invention are not limited to the method operations described above, but can also perform related operations in the group target tracking method provided in any embodiment of the present invention.

[0085] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0086] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0087] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0088] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for tracking a group of targets, characterized in that, include: Acquire target image information of the tracked target using distributed imaging devices; The target image information is input into the trained deep learning model to obtain the corresponding target detection image, and the measured position information of each tracked target is obtained based on the target detection image. The two-dimensional position information of each of the tracked targets is calculated based on the measured position information and the intrinsic parameter matrix of the distributed imaging device; The three-dimensional position information of each of the tracked targets is calculated based on the two-dimensional position information and the extrinsic parameter matrix of the distributed imaging device; The individual position coordinates of each of the tracking targets are determined based on the two-dimensional position information and the three-dimensional position information; Before calculating the three-dimensional position information of each tracking target based on the two-dimensional position information and the extrinsic matrix of the distributed imaging device, the method further includes: The viewing angle of the distributed imaging device is adjusted according to the two-dimensional position information so that the tracking target is located in the center of the field of view of the distributed imaging device. When a target appears in the field of view of a camera at a certain point, the position of the target is estimated based on the information obtained from that point, and other points are adjusted to rotate toward the position of the target.

2. The method for tracking a group of targets according to claim 1, characterized in that, After determining the individual position coordinates of each tracking target based on the two-dimensional position information and the three-dimensional position information, the method further includes: Based on the individual location coordinates, three-dimensional video and / or three-dimensional images are reconstructed for each of the tracked targets.

3. The method for tracking a group of targets according to claim 1, characterized in that, The step of adjusting the viewing angle of the distributed imaging device based on the two-dimensional position information includes: The viewing angle of the distributed imaging device is adjusted using a preset gimbal control algorithm developed based on PID control.

4. The method for tracking a group of targets according to claim 1, characterized in that, The step of obtaining the measured position information of each of the tracked targets based on the target detection image includes: The measured location information is obtained based on the Kalman filter algorithm.

5. The method for tracking a group of targets according to claim 1, characterized in that, Before inputting the target image information into the trained deep learning model, the method further includes: The deep learning model is trained using a convolutional neural network algorithm.

6. The method for tracking a group of targets according to claim 1, characterized in that, The distributed imaging device includes a self-zoom camera gimbal.

7. A device for tracking groups of targets, characterized in that, include: The target image information acquisition module is used to acquire target image information of the tracked target through a distributed imaging device. The measured location information acquisition module is used to input the target image information into the trained deep learning model to obtain the corresponding target detection image, and to obtain the measured location information of each of the tracked targets based on the target detection image; A two-dimensional position information calculation module is used to calculate the two-dimensional position information of each of the tracking targets based on the measured position information and the intrinsic parameter matrix of the distributed imaging device; A three-dimensional position information calculation module is used to calculate the three-dimensional position information of each of the tracking targets based on the two-dimensional position information and the external parameter matrix of the distributed imaging device; An individual position coordinate determination module is used to determine the individual position coordinates of each of the tracking targets based on the two-dimensional position information and the three-dimensional position information; The device further includes: The imaging device perspective adjustment module is used to adjust the perspective of the distributed imaging device according to the two-dimensional position information before calculating the three-dimensional position information of each tracking target based on the two-dimensional position information and the extrinsic matrix of the distributed imaging device, so that the tracking target is located in the center of the field of view of the distributed imaging device; when a tracking target appears in the field of view of a certain point camera, the position of the tracking target is estimated based on the information obtained from that point, and other points are adjusted to rotate toward the position of the tracking target.

8. A computer device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the group target tracking method as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the group target tracking method as described in any one of claims 1-6.

Images