Target matching method, target matching device, and computer storage medium

Abstract

This application discloses a target matching method, a target matching device, and a storage medium. The target matching method includes: acquiring a first image and a second image of the same monitored area captured by different cameras; obtaining a first target set from the first image and a second target set from the second image; calculating the spatial similarity distance between each target pair in the first target set and the second target set, wherein each target pair includes one target in the first target set and one target in the second target set; and matching the two targets in corresponding target pairs whose spatial similarity distance is less than a first preset distance. Through the above method, this application can match targets in different images by calculating the spatial similarity distance between two targets and comparing it with a preset distance, thereby achieving the fusion of the same target and improving the accuracy of target fusion.

Description

Technical Field

[0001] This application relates to the field of image processing, and in particular to a target matching method, a target matching device, and a computer storage medium. Background Technology

[0002] With the continuous development of computer technology, artificial intelligence (AI) technology, as a branch of computer technology, is also constantly progressing and gradually permeating people's daily lives. AI technology aims to transform machines that were originally simply controlled by programs into intelligent machines capable of working, operating, and reacting like humans. In the field of autonomous driving, various obstacles and other vehicles in the environment can interfere with the vehicle's operation, thus requiring accurate identification of these targets.

[0003] In one application scenario, when matching targets across multiple cameras, the main approach is to use projection to match the image from the current camera with images captured by other cameras, thereby achieving target recognition and fusion. However, this method relies on a projection matrix, and errors in the parameters of the projection matrix can lead to larger projection errors. When processing cases where the same target is projected onto the same image but there is no overlap, it may be identified as a new target, resulting in false positives or false negatives. Summary of the Invention

[0004] The main technical problem addressed by this application is how to improve the accuracy of target matching and fusion across multiple cameras. To this end, this application provides a target matching method, a target matching device, and a computer storage medium.

[0005] To address the aforementioned technical problems, this application provides a target matching method, comprising: acquiring a first image and a second image captured by different cameras of the same monitoring area; obtaining a first target set from the first image and a second target set from the second image; calculating the spatial similarity distance between each target pair in the first target set and the second target set, wherein each target pair includes a target in the first target set and a target in the second target set; and matching two targets in corresponding target pairs whose spatial similarity distance is less than a first preset distance.

[0006] The calculation of the spatial similarity distance between each target pair in the first target set and the second target set includes: obtaining the first range target set of the first target and the second range target set of the second target in each target pair; obtaining the minimum distance from each first range target in the first range target set to the second range target set; and obtaining the spatial similarity distance between the target pairs based on the maximum value of the minimum distances from all first range targets to the second range target set.

[0007] Specifically, obtaining the spatial similarity distance of target pairs based on the maximum value of the minimum distances from all targets in the first range to the second range target set includes: obtaining the spatial similarity distance of target pairs based on the maximum value of the minimum distances from all targets in the first range to the second range target set and the spatial distance between target pairs.

[0008] The process includes, after traversing the distance matrix and matching two targets whose spatial similarity distance is less than a first preset distance, marking the two matched targets as the same target.

[0009] To solve the above-mentioned technical problems, another technical solution adopted in this application is: to provide a target matching method, which includes: acquiring the shooting areas of several cameras and the images captured by the several cameras at the same time; filtering the images captured by each camera using the shooting areas; matching the filtered images of each camera with the images captured by other cameras using a preset target matching method; and matching the images captured by all cameras to obtain the final matching result; wherein, the preset target matching method is the target matching method described above.

[0010] The process of filtering images captured by each camera using the area under its responsibility includes: obtaining targets within the area under its responsibility from the images captured by each camera as the filtering results.

[0011] The process of matching the filtered images with images captured by other cameras using a preset target matching method includes: matching the filtered images with images captured by other cameras one by one until all images captured by all other cameras are matched, and this is taken as the target matching result.

[0012] The process of matching images captured by all cameras to obtain the final matching result includes: filtering images captured by each camera separately; matching the filtered results with images captured by other cameras to obtain the target matching result; and merging all target matching results to obtain the final matching result.

[0013] To solve the above-mentioned technical problems, another technical solution adopted in this application is: to provide a target matching device, which includes a processor and a memory, the memory being coupled to the processor, the memory storing program data, and the processor executing the program data to implement the target matching method as described above.

[0014] To solve the above-mentioned technical problems, another technical solution adopted in this application is to provide a computer-readable storage medium that stores program data, which, when executed, is used to implement the above-mentioned target matching method.

[0015] The beneficial effects of this application are as follows: Unlike existing technologies, the target matching method provided by this invention is applied to a target matching device. The target matching device acquires a first image and a second image of the same monitored area captured by different cameras; obtains a first target set from the first image and a second target set from the second image; calculates the spatial similarity distance between each target pair in the first target set and the second target set, wherein each target pair includes one target in the first target set and one target in the second target set; and matches the two targets in the corresponding target pairs whose spatial similarity distance is less than a first preset distance. Through the above method, compared with conventional target matching methods, the method adopted in this application, which acquires different images of the same area captured by different cameras in the target matching device and matches and associates the targets in the images based on a spatial geometric similarity matching algorithm to fuse the same targets in multiple cameras, can accurately calculate the spatial similarity distance between two targets based on the target's own information, thereby improving the accuracy of target matching and fusion. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] in:

[0018] Figure 1 This is a flowchart illustrating the first embodiment of the target matching method provided in this application;

[0019] Figure 2 This is a schematic diagram of an embodiment of the target matching method provided in this application for determining the range of target sets for each target;

[0020] Figure 3 This is a flowchart illustrating the target matching process using a distance matrix in the target matching method provided in this application.

[0021] Figure 4 This is a flowchart illustrating the second embodiment of the target matching method provided in this application;

[0022] Figure 5 This is a schematic diagram of the process of applying the target matching method provided in this application to the target matching device;

[0023] Figure 6 This is a schematic diagram of the region after division in the target matching method provided in this application;

[0024] Figure 7 This is a schematic diagram of the target before fusion in one embodiment of the target matching method provided in this application;

[0025] Figure 8 This is a schematic diagram of the fused target in one embodiment of the target matching method provided in this application;

[0026] Figure 9 This is a schematic diagram of the structure of the first embodiment of the target matching device provided in this application;

[0027] Figure 10 This is a schematic diagram of the structure of the second embodiment of the target matching device provided in this application;

[0028] Figure 11 This is a schematic diagram of an embodiment of the computer-readable storage medium provided in this application. Detailed Implementation

[0029] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0030] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Furthermore, "many" in this document means two or more. Moreover, the term "at least one" in this document means any combination of at least two of any one or more of a plurality of objects. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0031] The target matching method provided in this application is mainly applied to a target matching device, wherein the target matching device can be a server or a system in which the server and terminal devices cooperate with each other. Accordingly, the various parts of the target matching device, such as various units, sub-units, modules, and sub-modules, can all be set in the server, or they can be set in the server and the terminal devices respectively.

[0032] Furthermore, the aforementioned server can be either hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When the server is software, it can be implemented as multiple software programs or software modules, such as software or software modules used to provide distributed servers, or as a single software program or software module; no specific limitation is made here. In some possible implementations, the target matching method of this application embodiment can be implemented by a processor calling computer-readable instructions stored in memory.

[0033] The target matching method provided in this application is mainly applied to target fusion algorithms using multiple cameras and multiple perspectives. In the field of autonomous driving using artificial intelligence, single-vehicle intelligence and vehicle-to-infrastructure (V2I) cooperation are the two major development directions of autonomous driving. Single-vehicle intelligence refers to the vehicle perceiving its surrounding environment through hardware devices such as LiDAR, cameras, and ultrasonic radar, and then transmitting this information to the software system for analysis and decision-making to control the vehicle's driving state. V2I cooperation systems are used to build dynamic real-time interactions between vehicles, between people and vehicles, and between vehicles and people, achieving effective coordination between people, vehicles, and roads, ensuring traffic safety, and improving traffic efficiency—a highly efficient, convenient, and safe road traffic system. Therefore, to build a better V2I cooperation system, it is necessary to match and fuse targets detected by multiple cameras to obtain more accurate vehicle information.

[0034] See Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the target matching method provided in this application.

[0035] Step 11: Acquire the first and second images of the same monitored area captured by different cameras.

[0036] Specifically, the first and second images here can be taken by mobile monitoring equipment or by monitoring equipment in a relatively fixed location. Mobile monitoring equipment can be mobile robots, drones, mobile phones, vehicles, or other devices with video and / or photographic capabilities that are capable of movement. Relatively fixed monitoring equipment can be devices permanently installed in various corners, such as cameras installed on school playgrounds or roads. The monitored area can be a specific area on a road or an area in front of a building; there are no specific limitations.

[0037] Specifically, the target matching device acquires images taken by different cameras at the same time on the same monitored area. The number of images can be two or more, and the number of images can be acquired according to the actual needs of the user, which is not limited here.

[0038] Step 12: Obtain the first target set from the first image and the second target set from the second image.

[0039] Specifically, the target matching device acquires all targets in the first image and the second image, and constructs a first target set and a second target set according to the targets.

[0040] Step 13: Calculate the spatial similarity distance between each pair of targets in the first target set and the second target set, wherein each target pair includes a target in the first target set and a target in the second target set.

[0041] See Figure 2 , Figure 2 This is a schematic diagram of an embodiment of the target matching method provided in this application for determining the range of target sets for each target. In the diagram, each square represents a target, and the yellow dot represents the center position of the target. Within a radius of thresh1 centered on the center point of a1, the center points of a1, a2, a3, a4, a5, and a6 all fall within this radius. Therefore, a... 1s ={a1,a2,a3,a4,a5,a6}, a 1s That is, the target set within the range of a1.

[0042] A can be obtained in the same way. set ={a 1s ,a 2s ,a 3s ,a 4s …,a ns}, where a 1s ,a 2s ,a 3s ,a 4s …,a ns It is A set The set of points within the range of targets whose L2 distance to the target is less than thresh1 is the target set corresponding to several targets. Similarly, B can be obtained. set ={b 1s ,b 2s ,b 3s ,b 4s …,b ms}, here A set A corresponds one-to-one with B, and the same applies to B. set Too.

[0043] Wherein, the L2 distance can be expressed as In other embodiments of this application, the Euclidean distance, Manhattan distance, Minkowski distance, Chebyshev distance, Minkowski distance, etc., between targets can be selected for measurement, and no limitation is made here.

[0044] Optionally, the spatial similarity distance between targets can be measured not only by the distance between the center points of the targets, but also by the length, width, size, trajectory equation, acceleration, orientation, and motion state estimation equation of the vehicle.

[0045] Specifically, the target matching device acquires a first range target set for the first target and a second range target set for the second target in each target pair; acquires the minimum distance from each first range target in the first range target set to the second range target set; and acquires the spatial similarity distance of the target pair based on the maximum value of the minimum distances from all first range targets to the second range target set.

[0046] For example, the first range of target set a of the first target a1 1s ={a1,a2,a3,a4,a5,a6}, the second range of target set b for the second target b1. 1s ={b1,b2,b3,b4}, the target matching device obtains a respectively 1s From a1 to b 1s Given the distances between b1, b2, b3, and b4, take the minimum value a1b2. The distance from a2 to b... 1s Given the distances between b1, b2, b3, and b4, take the minimum value a2b3. The distance from a3 to b... 1s Find the minimum distances between points b1, b2, b3, and b4, and then take the minimum value a3b1. The distance from a4 to b... 1s Given the distances from b1, b2, b3, and b4, take the minimum value a4b1. Also, consider the distances from a5 to b. 1s Given the distances from b1, b2, b3, and b4, take the minimum value a5b3. Then, from a6 to b... 1s The minimum value, a6b4, is taken from the distances between b1, b2, b3, and b4. Finally, the target matching device takes the maximum value, a2b3, among a1b2, a2b3, a3b1, a4b1, a5b3, and a6b4 as the spatial similarity distance between the first target a1 and the second target b1.

[0047] Optionally, the target matching device can also obtain the spatial similarity distance between target pairs based on the maximum value of the minimum distances from all targets in the first range to the second range target set and the spatial distance between target pairs. The spatial distance can be the L2 distance, and the target matching device uses the sum of the spatial distance and the maximum value of the minimum distances from the first range targets to the second range target set as the spatial similarity distance between the first target and the second target.

[0048] The spatial similarity distance between the first target and the second target can be expressed by the following formula:

[0049]

[0050] Where a1 and b1 are the first and second objectives, respectively, a 1s and b 1s Let be the first range target set and the second range target set, respectively, and d(a1,b1) be the spatial distance between the first target and the second target.

[0051] Step 14: Match the two targets in the target pairs whose spatial similarity distance is less than the first preset distance.

[0052] Optionally, the target matching device can also establish a distance matrix based on the spatial similarity distance of each target pair, such as... Figure 3 As shown, Figure 3 This is a flowchart illustrating the target matching process using a distance matrix in the target matching method provided in this application. The distance matrix D(N,M) represents the spatial similarity distance matrix from each target in the first target set to each target in the second target set, and is N×M, where there are N targets in the first target set and M targets in the second target set.

[0053] Specifically, the target matching device iterates through each column i in the distance matrix D(N,M), finds all columns whose index values ​​have not been recorded, and then iterates through them to find the index j of the column with the smallest spatial similarity distance, and records the index. The target matching device then judges the value D[i][j]. If the value D[i][j] is less than the predetermined threshold match_thresh, then it considers the two targets corresponding to that value to be a match.

[0054] Specifically, the first preset distance can be set by the user according to their needs, and there are no restrictions here.

[0055] Specifically, after matching two targets in a target pair whose spatial similarity distance is less than a first preset distance, the target matching device will also mark the two matched targets as the same target.

[0056] In practical applications of target fusion, the camera mounting angle can cause target occlusion issues when taking photos. Therefore, the same target needs to be observed by cameras from multiple directions. Ideally, the target seen by each camera should overlap in the projection coordinate system. However, due to the aforementioned issues, the projected positions differ significantly. Therefore, this application proposes a target matching method based on camera viewpoint.

[0057] See Figure 4 and Figure 5 , Figure 4 This is a flowchart illustrating the second embodiment of the target matching method provided in this application. Figure 5This is a schematic diagram of the process of applying the target matching method provided in this application to the target matching device.

[0058] Step 41: Obtain the shooting areas of several cameras and the images captured by several cameras at the same time.

[0059] like Figure 6 As shown, Figure 6 This is a schematic diagram of the region segmentation in the target matching method provided in this application. Four cameras observe a region: Camera 1, Camera 2, Camera 3, and Camera 4. Camera 1 is responsible for capturing the regions C1B2OD1B1 and B8B9B10D4; Camera 2 is responsible for capturing the regions C2B5OD2B4 and D1B11B12B1; Camera 3 is responsible for capturing the regions C3B8OD3B6 and B2B3B4D2; and Camera 4 is responsible for capturing the regions C4B11OD4B10 and B5B6B7D4. The regions captured by each camera can be set by the user and are not limited here. The images captured by the cameras are images taken by the cameras at different angles at the same time.

[0060] Step 42: Filter the images captured by each camera using the area to be captured.

[0061] Specifically, each camera's captured image may include several targets. The target matching device extracts targets belonging to the corresponding camera's responsible shooting area and places them into a responsible target set. As shown in the example above, there are a total of four responsible target sets, denoted as cam1_main_set, cam2_main_set, cam3_main_set, and cam4_main_set. The target set in each camera's captured image is denoted as cam1_obj_set, cam2_obj_set, cam3_obj_set, and cam4_obj_set.

[0062] Specifically, the target matching device acquires targets within the responsible shooting area from the captured images of each camera as the filtering results.

[0063] Step 43: Use a preset target matching method to match the filtered images from each camera with images from other cameras; wherein, the preset target matching method is the target matching method described above.

[0064] Specifically, the target matching device uses the preset target matching method to match the filtered images with the images captured by the other cameras in pairs until the images captured by all other cameras are matched, which is taken as the target matching result.

[0065] Specifically, the target matching device uses the target matching method described in steps 11 to 14 above to perform target matching between the filtered images from each camera and the images captured by other cameras. Continuing with camera 1 in the example above, the target set cam1_main_set in the regions C1B2OD1B1 and B8B9B10D4 is first matched with the images captured by other cameras.

[0066] Specifically, the target matching device fuses the images captured by each camera with the target set that camera 1 is responsible for in pairs, resulting in three fused target sets: cam1_main_set-cam2_obj_set (fusion of the region that camera 1 is responsible for with the images captured by camera 2), cam1_main_set-cam3_obj_set (fusion of the region that camera 1 is responsible for with the images captured by camera 3), and cam1_main_set-cam4_obj_set (fusion of the region that camera 1 is responsible for with the images captured by camera 4).

[0067] Because of the common value cam1_main_set, the target matching device takes the intersection of these three sets and performs a second fusion: cam3_obj_set - cam1_main_set - cam2_obj_set (the area handled by camera 1 is fused with the image taken by camera 2 and then fused with the image taken by camera 3) cam4_obj_set - cam1_main_set - cam3_obj_set (the area handled by camera 1 is fused with the image taken by camera 4 and then fused with the image taken by camera 3).

[0068] The target matching device then merges the two sets again to obtain the final fusion result am1_main_set-cam3_obj_set-cam2_obj_set-cam4_obj_set (the area handled by camera 1 is fused with the image taken by camera 4, then with the image taken by camera 3, and finally with the image taken by camera 2).

[0069] Step 44: Match the images captured by all cameras to obtain the final matching result.

[0070] Specifically, the target matching device filters the images captured by each of the cameras; matches the filtering results with the images captured by other cameras to obtain target matching results; and merges all the target matching results to obtain the final matching result.

[0071] Specifically, the target matching device performs target matching on the images selected by all cameras and the images captured by other cameras to obtain an initial matching result. Then, the initial matching results obtained by each camera are matched again to obtain the final matching result.

[0072] See Figure 7 and Figure 8 , Figure 7 This is a schematic diagram of the target before fusion in one embodiment of the target matching method provided in this application; Figure 8 This is a schematic diagram of the fused target in one embodiment of the target matching method provided in this application. Figure 7 The middle section shows the target set captured by each sensor (S1, S2, S3, S4) before the target matching device fuses the matching results. As can be seen from the figure, there is overlap in the targets captured by each sensor. (See also...) Figure 8 , Figure 8 To use the target matching method provided in this application for Figure 7 The target diagram obtained by matching and fusing the targets in the middle is used to fuse overlapping targets together based on the target matching method provided in this application, thereby realizing target matching and fusion of multiple cameras and multiple views, and enabling accurate target matching under multiple views.

[0073] Unlike existing technologies, the target matching method provided by this invention is applied to a target matching device. The device acquires first and second images of the same monitored area from different cameras; obtains a first target set from the first image and a second target set from the second image; calculates the spatial similarity distance between each pair of targets in the first and second target sets, where each target pair includes one target in the first target set and one target in the second target set; and matches the two targets in the corresponding target pairs whose spatial similarity distance is less than a first preset distance. Compared to conventional target matching methods, this invention, which acquires different images of the same area from different cameras in the target matching device and matches and associates targets in the images based on a spatial geometric similarity matching algorithm to fuse the same targets from multiple cameras, can accurately calculate the spatial similarity distance between two targets based on the target's own information, thereby improving the accuracy of target matching and fusion.

[0074] The method described in the above embodiments can be implemented using a target matching device, as described below. Figure 9 Describe, Figure 9 This is a schematic diagram of the structure of the first embodiment of the target matching device provided in this application.

[0075] like Figure 9As shown, the target matching device 90 in this embodiment of the application includes an image acquisition module 91, a target acquisition module 92, a calculation module 93, and a matching module 94.

[0076] The image acquisition module 91 is used to acquire first and second images of the same monitored area captured by different cameras.

[0077] The target acquisition module 92 is used to acquire a first target set from a first image and a second target set from a second image.

[0078] The calculation module 93 is used to calculate the spatial similarity distance between each target pair in the first target set and the second target set, wherein the target pair includes a target in the first target set and a target in the second target set.

[0079] The matching module 94 is used to match two targets in a target pair whose spatial similarity distance is less than a first preset distance.

[0080] The method described in the above embodiments can be implemented using a target matching device, as described below. Figure 10 , Figure 10 This is a schematic diagram of the structure of the second embodiment of the target matching device provided in this application. The target matching device 100 includes a memory 101 and a processor 102. The memory 101 is used to store program data, and the processor 102 is used to execute the program data to implement the following method:

[0081] Acquire a first image and a second image of the same monitored area captured by different cameras; obtain a first target set from the first image and a second target set from the second image; calculate the spatial similarity distance between each target pair in the first target set and the second target set, wherein a target pair includes a target in the first target set and a target in the second target set; match the two targets in the corresponding target pairs whose spatial similarity distance is less than a first preset distance.

[0082] See Figure 11 , Figure 11 This is a schematic diagram of an embodiment of a computer-readable storage medium provided in this application. The computer-readable storage medium 110 stores program data 111, which, when executed by a processor, is used to implement the following method:

[0083] Acquire a first image and a second image of the same monitored area captured by different cameras; obtain a first target set from the first image and a second target set from the second image; calculate the spatial similarity distance between each target pair in the first target set and the second target set, wherein a target pair includes a target in the first target set and a target in the second target set; match the two targets in the corresponding target pairs whose spatial similarity distance is less than a first preset distance.

[0084] When the embodiments of this application are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0085] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A target matching method characterized by, The target matching method includes: Acquire first and second images of the same monitored area captured by different cameras; A first target set is obtained from the first image, and a second target set is obtained from the second image; Calculate the spatial similarity distance between each pair of targets in the first target set and the second target set, wherein the target pair includes a target in the first target set and a target in the second target set; Matching two targets in a target pair whose spatial similarity distance is less than a first preset distance includes: establishing a distance matrix based on the spatial similarity distance of each target pair, wherein the distance matrix represents the spatial similarity distance from each target in the first target set to each target in the second target set; traversing each column i in the distance matrix, finding all columns that have not been recorded with index values ​​and traversing them, finding the index j of the column with the smallest index value, and recording the index; judging the recorded index value D[i][j], and if the index value D[i][j] is less than a predetermined threshold, then the two targets corresponding to the index value D[i][j] are considered to be matched; The step of calculating the spatial similarity distance between each target pair in the first target set and the second target set includes: obtaining a first range target set for the first target and a second range target set for the second target in each target pair, wherein the center point of each first range target in the first range target set falls within a range with the center point of the first target in the corresponding target pair as the center and a preset radius; obtaining the minimum distance from each first range target in the first range target set to the second range target set; and taking the sum of the spatial distance between the first target and the second target in the target pair and the maximum value of the minimum distances from all first range targets to the second range target set as the spatial similarity distance of the target pair.

2. The target matching method according to claim 1, characterized in that, After matching two targets in the target pairs whose spatial similarity distance is less than a first preset distance, the method further includes: Mark the two matched targets as the same target.

3. A target matching method characterized by, The target matching method includes, Acquire the shooting areas of several cameras and the images captured by the cameras at the same time; The images captured by each camera are filtered using the area to be captured. A preset target matching method is used to match the filtered images captured by each camera with the images captured by other cameras. The final matching result is obtained by matching the images captured by all cameras. The preset target matching method is the target matching method according to any one of claims 1 to 2.

4. The target matching method according to claim 3, characterized in that, The step of filtering the images captured by each camera using the designated shooting area includes: The targets within the responsible shooting area in the captured images of each camera are obtained as the filtering results.

5. The target matching method according to claim 3, characterized in that, The step of matching the filtered images with other images captured by the camera using a preset target matching method includes: The preset target matching method is used to match the filtered images with the images captured by the other cameras in pairs until all images captured by the other cameras are matched, and this is taken as the target matching result.

6. The target matching method according to claim 3, characterized in that, The step of matching the images captured by all the cameras to obtain the final matching result includes: The images captured by each of the cameras are filtered separately; The filtered results are matched with images captured by other cameras to obtain target matching results; The final matching result is obtained by merging all the target matching results.

7. A target matching device, characterized by, The target matching device includes a memory and a processor coupled to the memory; The memory is used to store program data, and the processor is used to execute the program data to implement the target matching method as described in any one of claims 1 to 2 and / or the target matching method as described in any one of claims 3 to 6.

8. A computer storage medium, characterized in that The computer storage medium is used to store program data, which, when executed by the computer, is used to implement the target matching method as described in any one of claims 1 to 2 and / or the target matching method as described in any one of claims 3 to 6.

Images