Identity matching method, device, computer equipment and readable storage medium

Abstract

The present application relates to an identity matching method, apparatus, computer device, and computer-readable storage medium. By clustering the human body images of the identities to be matched, a human body image file corresponding to the identity in the identity database is obtained; according to the image feature similarity between the human body images in the human body image file and the identity card images in the identity database, a first identity matching is performed on the human body image file; if there are human body image files that have not been matched after the first identity matching, the trajectory association relationship between the personnel trajectories corresponding to the human body image files that have been matched with identities and the personnel trajectories corresponding to the human body image files that have not been matched with identities is obtained; according to the trajectory association relationship, a second identity matching is performed on the human body image files that have not been matched with identities, effectively improving the probability of identity matching.

Description

Technical Field

[0001] The present application relates to the technical field of image processing, and in particular, to an identity matching method, apparatus, computer device, and computer-readable storage medium. Background Art

[0002] With the progress of technology and the needs of humans for a convenient life, various face-swiping applications have begun to spread, including face-swiping for unlocking, opening doors, security checks, boarding planes, checking into hotels, and paying for medical treatment. They have taken root in all walks of life and generated a large number of human body images. How to make good use of these human body images and make them have a positive impact on social progress and stability? The most important step is to match the identities of the people corresponding to these human body images so that these human body images can be filed in the corresponding personal identity files.

[0003] In the prior art, the identity of a human body image is matched by comparing the human body image with the ID card image in the ID card database one by one. However, affected by the image capture scene, the capture quality of human body images varies greatly. In order to ensure accuracy in identity matching, the similarity threshold between the human body image and the ID card image in the ID card database is usually very high. Some human body images with poor quality are difficult to match the identity, resulting in a low identity matching rate of human body images. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide an identity matching method, apparatus, computer device, and computer-readable storage medium to solve the problem of low identity matching rate of human body images in the related art.

[0005] In a first aspect, an embodiment of the present application provides an identity matching method, and the method includes the following steps:

[0006] Cluster the human body images to be matched for identity to obtain a human body image file corresponding to the identity in the ID card database;

[0007] Perform a first identity match on the human body image file according to the image feature similarity between the human body image in the human body image file and the ID card image in the ID card database;

[0008] If there are human body image files that have not been matched to an identity after the first identity match, obtain the trajectory association relationship between the trajectory of the person corresponding to the human body image file that has been matched to an identity and the trajectory of the person corresponding to the human body image file that has not been matched to an identity;

[0009] Perform a second identity match on the human body image file that has not been matched to an identity according to the trajectory association relationship.

[0010] In some of these embodiments, the trajectory association relationship includes an associated file; if, after the first identity match, there are human body image files for which no identity is matched, obtaining the trajectory association relationship between the personnel trajectory corresponding to the human body image file for which an identity has been currently matched and the personnel trajectory corresponding to the human body image file for which no identity is matched includes:

[0011] Based on the human body image files, obtain the personnel trajectories corresponding to each of the human body image files;

[0012] According to the personnel trajectories, determine the human body image files that are associated with the personnel trajectory corresponding to the human body image file for which an identity has been currently matched and for which no identity is matched, and use them as the associated files of the human body image file for which an identity has been currently matched.

[0013] In some of these embodiments, the secondary identity matching of the human body image files for which no identity is matched according to the trajectory association relationship includes the following steps:

[0014] Obtain the ID card images of the personnel who have an association relationship with the personnel corresponding to the human body image file for which an identity has been matched, as the associated ID card images;

[0015] Perform secondary identity matching on the associated files according to the image feature similarity between the human body images in the associated files and the associated ID card images.

[0016] In some of these embodiments, the step of determining the human body image files that are associated with the personnel trajectory corresponding to the human body image file for which an identity has been currently matched and for which no identity is matched, and using them as the associated files of the human body image file for which an identity has been currently matched includes the following steps:

[0017] According to the personnel trajectories corresponding to all the human body image files, calculate the number of spatio-temporal concomitant points between the personnel trajectory corresponding to the human body image file for which no identity is matched and the personnel trajectory corresponding to the human body image file for which an identity has been currently matched;

[0018] Use the human body image files for which no identity is matched and for which the number of spatio-temporal concomitant points with the personnel trajectory corresponding to the human body image file for which an identity has been currently matched exceeds a first set threshold as the associated files of the human body image file for which an identity has been currently matched.

[0019] In some of these embodiments, the step of obtaining the ID card images of the personnel who have an association relationship with the personnel corresponding to the human body image file for which an identity has been matched, as the associated ID card images includes the following steps:

[0020] Obtain the ID card images of persons belonging to the same administrative division as the person corresponding to the human body image file with the matched identity, and the ID card images of persons having a social relationship with the person corresponding to the human body image file with the matched identity, as associated ID card images; the social relationships include: friendship, kinship or work relationship.

[0021] In some embodiments, the obtaining the ID card images of persons belonging to the same administrative division as the person corresponding to the human body image file with the matched identity, and the ID card images of persons having a social relationship with the person corresponding to the human body image file with the matched identity, as associated ID card images, includes the following steps:

[0022] Obtain the ID card images of persons belonging to the same administrative division as the person corresponding to the human body image file with the matched identity, and the ID card images of persons having a social relationship with the person corresponding to the human body image file with the matched identity;

[0023] Screen out the ID card images of persons whose number of spatio-temporal companion points of the mobile phone mac trajectory of the person corresponding to the human body image file with the matched identity exceeds a second set threshold and whose number of spatio-temporal companion points of the vehicle trajectory of the person corresponding to the human body image file with the matched identity exceeds a third set threshold, as the associated ID card images.

[0024] In some embodiments, the first identity matching of the human body image file according to the image feature similarity between the human body image in the human body image file and the ID card image in the ID card database includes the following steps:

[0025] According to the image quality, screen out a preset number of human body images from each human body image file as preferred human body images;

[0026] Perform identity matching on each human body image file according to the image feature similarity between the preferred human body image and the ID card image in the ID card database.

[0027] In some embodiments, the clustering of the human body images with the identity to be matched includes the following steps:

[0028] Cluster the human body images with the identity to be matched according to the image feature similarity between the human body images with the identity to be matched.

[0029] In a second aspect, in the present embodiment, an identity matching device is provided, the device includes: a clustering module, a first matching module, an association module and a second matching module:

[0030] The clustering module is used to cluster the human body images of the identities to be matched, and obtain human body image files corresponding to the identities in the ID card database;

[0031] The first matching module is used to perform a first identity match on the human body image file according to the image feature similarity between the human body image in the human body image file and the ID card image in the ID card database;

[0032] The association module is used to, if there are human body image files that have not been matched after the first identity match, obtain the trajectory association relationship between the personnel trajectory corresponding to the human body image file that has been matched with an identity and the personnel trajectory corresponding to the human body image file that has not been matched with an identity;

[0033] The second matching module is used to perform a second identity match on the human body image file that has not been matched with an identity according to the trajectory association relationship.

[0034] In a third aspect, a computer device is provided in this embodiment, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the method described in the first aspect above are implemented.

[0035] In a fourth aspect, a computer-readable storage medium is provided in this embodiment, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in the first aspect above are implemented.

[0036] For the above identity matching method, device, computer device, and computer-readable storage medium, by clustering the human body images of the identities to be matched, human body image files corresponding to the identities in the ID card database are obtained; according to the image feature similarity between the human body images in the human body image files and the ID card images in the ID card database, a first identity match is performed on the human body image files; if there are human body image files that have not been matched after the first identity match, the trajectory association relationship between the personnel trajectory corresponding to the human body image file that has been matched with an identity and the personnel trajectory corresponding to the human body image file that has not been matched with an identity is obtained; according to the trajectory association relationship, a second identity match is performed on the human body image file that has not been matched with an identity. By mining the trajectory association relationship between the personnel trajectory corresponding to the human body image file that has been matched with an identity and the personnel trajectory corresponding to the human body image file that has not been matched with an identity, a second identity match is performed on the human body image file that has not been matched with an identity, effectively improving the probability of identity matching. Description of the Drawings

[0037] The accompanying drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The schematic embodiments and descriptions thereof of the present application are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0038] Figure 1 is an application scenario diagram of the identity matching method provided according to an embodiment of the present application;

[0039] Figure 2 is a flowchart of the identity matching method provided according to an embodiment of the present application;

[0040] Figure 3 is a flowchart of the first identity matching in the identity matching method provided according to an embodiment of the present application;

[0041] Figure 4 is a schematic structural diagram of the identity matching device provided according to an embodiment of the present application;

[0042] Figure 5 is a schematic structural diagram of the computer device provided according to an embodiment of the present application. Detailed implementation manners

[0043] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be described and explained below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments provided in the present application without creative efforts belong to the scope of protection of the present application.

[0044] Obviously, the accompanying drawings in the following description are only some examples or embodiments of the present application. For those of ordinary skill in the art, without creative efforts, the present application can also be applied to other similar scenarios based on these drawings. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for those of ordinary skill in the art related to the content disclosed in the present application, some design, manufacturing or production changes based on the technical content disclosed in the present application are only conventional technical means and should not be understood that the content disclosed in the present application is insufficient.

[0045] Referring to "embodiment" in the present application means that the specific features, structures or characteristics described in connection with the embodiment may be included in at least one embodiment of the present application. The phrase appears in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those of ordinary skill in the art explicitly and implicitly understand that the embodiments described in the present application can be combined with other embodiments without conflict.

[0046] Unless otherwise defined, the technical terms or scientific terms involved in this application shall have the ordinary meanings understood by those with ordinary skills in the technical field to which this application belongs. The words such as "a", "an", "one kind", "the" and the like involved in this application do not indicate a limitation in quantity and may represent a singular or plural number. The terms "including", "containing", "having" and any variations thereof involved in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product or device containing a series of steps or modules (units) is not limited to the listed steps or units, but may further include steps or units not listed, or may further include other steps or units inherent to these processes, methods, products or devices. The similar words such as "connected", "linked", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "plurality" involved in this application means two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships may exist. For example, "A and / or B" may represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the front and rear associated objects. The terms "first", "second", "third" and the like involved in this application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0047] Figure 1 It is an application scenario diagram of the identity matching method provided for an embodiment of this application. As Figure 1 shown, data can be transmitted between the server 101 and the mobile terminal 102 through a network. Among them, the mobile terminal 102 is used to collect a human body image of the identity to be matched and transmit the collected human body image of the identity to be matched to the server 101. After the server 101 receives the human body image of the identity to be matched, it clusters the human body image of the identity to be matched to obtain a human body image file corresponding to the identity in the ID card database; according to the image feature similarity between the human body image in the human body image file and the ID card image in the ID card database, it performs a first identity match on the human body image file; if there is a human body image file that fails to match the identity after the first identity match, it obtains the trajectory association relationship between the personnel trajectory corresponding to the human body image file that has been matched with the identity and the personnel trajectory corresponding to the human body image file that fails to match the identity; according to the trajectory association relationship, it performs a second identity match on the human body image file that fails to match the identity. Among them, the server 101 can be implemented by an independent server or a server cluster composed of multiple servers, and the mobile terminal 102 can be an image acquisition device such as a camera or a mobile phone.

[0048] This embodiment provides an identity matching method, which can be used for identity matching of human body images in the field of image processing technology, such as Figure 2 As shown, this method includes the following steps:

[0049] Step S210: Cluster the human body images with the identity to be matched to obtain human body image files corresponding to the identities in the ID card database.

[0050] Specifically, in daily life, each person will be captured a series of human body images. Before these human body images are filed, identity matching needs to be performed so that these human body images are filed in the corresponding personnel identity files. The human body images with the identity to be matched can be the initial human body images directly captured by a camera or a camera, or the initial human body images obtained from a human body image library. In order to obtain a better identity matching effect, preprocessing operations can be performed on the initial human body images, such as denoising and enhancement processing, so as to obtain the human body images with the identity to be matched. Cluster all the human body images with the identity to be matched, and cluster the human body images belonging to the activity track of a person into the same human body image file, so as to obtain one or more human body image files corresponding to the identities in the ID card database. The specific way to cluster the human body images with the identity to be matched can be to compare the feature values of the human body images to be matched one by one, so as to cluster the human body images with the identity to be matched; or use existing human body image clustering methods, such as Kmeans clustering, etc., so as to cluster the human body images with the identity to be matched.

[0051] Step S220: Perform the first identity matching on the human body image file according to the image feature similarity between the human body image in the human body image file and the ID card image in the ID card database.

[0052] Specifically, the ID card images in the ID card database can be obtained from the human body image database. Through existing human feature extraction networks, such as CNN convolutional networks, etc., the human body images in the human body image file and the ID card images in the ID card database can be subjected to feature extraction, so as to obtain the image features of the human body images in the human body image file and the ID card images in the ID card database. The image features can represent the concentrated information of the image and are represented by multi-dimensional vectors. After obtaining the image features of the human body images in the human body image file and the ID card images in the ID card database, the similarity comparison between the image features of the human body images in the human body image file and the ID card images in the ID card database can be carried out, so as to perform the first identity matching on the human body image file. For example, when the similarity of the image features between the human body image in the human body image file and the ID card image in the ID card database exceeds the fourth set threshold, it is determined that the current human body image matches the identity. Since the above step S210 clusters the human body images to be matched for identity, ensuring that the human body images in the same human body image file correspond to the same personnel identity, when the current human body image matches the identity, it can be considered that the human body image file where the current human body image is located matches the identity. Specifically, the similarity conditions can be set according to actual needs, and the first identity matching of the human body image file can be performed according to the similarity of the image features between the human body images in the human body image file and the ID card images in the ID card database.

[0053] Step S230, if there are human body image files that have not been matched for identity after the first identity matching, obtain the trajectory association relationship between the personnel trajectory corresponding to the human body image file that has been matched for identity and the personnel trajectory corresponding to the human body image file that has not been matched for identity.

[0054] Specifically, since identity matching requires a usually high similarity threshold between the human body image and the ID card image in the ID card database to ensure accuracy, after the first identity matching of the human body image file through the above step S220, there may be human body image files that have not been matched for identity. According to the personnel trajectory corresponding to the human body image file that has been matched for identity and the personnel trajectory corresponding to the human body image file that has not been matched for identity, the trajectory association relationship between the personnel trajectory corresponding to the human body image file that has been matched for identity and the personnel trajectory corresponding to the human body image file that has not been matched for identity can be obtained. For example, at which time points and locations the personnel trajectory corresponding to the human body image file that has been matched for identity coincides with the personnel trajectories corresponding to which human body image files that have not been matched for identity.

[0055] Step S240, perform secondary identity matching on the human body image files that have not been matched for identity according to the trajectory association relationship.

[0056] Specifically, if there is a trajectory association relationship between the personnel trajectory corresponding to the human body image file whose identity has been currently matched and the personnel trajectory corresponding to the human body image file whose identity has not been currently matched, the human body image file whose identity has not been matched can be secondarily matched for identity according to this trajectory association relationship. For example, according to the time point and location of the trajectory association, other trajectories of the person corresponding to the human body image file whose identity has been currently matched at the time point and location of the trajectory association can be obtained, such as the mobile phone mac trajectory, vehicle trajectory, etc., to find the identity of the associated person who has other trajectory associations with the person corresponding to the human body image file whose identity has been currently matched at the time point and location of the trajectory association, and secondarily match the human body image file whose identity has not been matched with the identity of the associated person corresponding to the human body image file whose identity has been currently matched.

[0057] In the related art, the identity of the human body image is matched by comparing the human body image with the identity card image in the identity card database one by one. However, affected by the image capture scenario, the capture quality of the human body image is uneven. In order to ensure accuracy in identity matching, the similarity threshold between the human body image and the identity card image in the identity card database is usually very high, and it is difficult to match the identity of the human body image with relatively poor quality, resulting in a low identity matching rate of the human body image. In this application, through the above steps S210 to S240, by clustering the human body images to be matched for identity, human body image files corresponding to the identities in the identity card database are obtained; according to the image feature similarity between the human body images in the human body image files and the identity card images in the identity card database, the human body image files are first matched for identity; if there are human body image files whose identities have not been matched after the first identity matching, the trajectory association relationship between the personnel trajectory corresponding to the human body image file whose identity has been currently matched and the personnel trajectory corresponding to the human body image file whose identity has not been matched is obtained; according to the trajectory association relationship, the human body image files whose identities have not been matched are secondarily matched for identity. By mining the trajectory association relationship between the personnel trajectory corresponding to the human body image file whose identity has been matched and the personnel trajectory corresponding to the human body image file whose identity has not been matched, the human body image files whose identities have not been matched are secondarily matched for identity, effectively increasing the probability of identity matching.

[0058] As one implementation manner, the trajectory association relationship includes associated files. If there are human body image files whose identities have not been matched after the first identity matching in the above step S230, the steps for obtaining the trajectory association relationship between the personnel trajectory corresponding to the human body image file whose identity has been currently matched and the personnel trajectory corresponding to the human body image file whose identity has not been matched include the following steps:

[0059] Step S231, based on the human body image files, obtain the personnel trajectories corresponding to each human body image file;

[0060] Step S232: Determine the human body image files that are associated with the personnel trajectory corresponding to the human body image file whose identity has been matched currently and whose identities have not been matched, and use them as the associated files of the human body image file whose identity has been matched currently.

[0061] Specifically, the human body images in the human body image files usually contain information such as the capture time and capture location. According to the human body images in the human body image files, the personnel trajectory corresponding to the human body image file can be obtained. Based on the personnel trajectory corresponding to the human body image file whose identity has been matched and the personnel trajectory corresponding to the human body image file whose identity has not been matched, the trajectory association relationship between the personnel trajectory corresponding to the human body image file whose identity has been matched currently and the personnel trajectory corresponding to the human body image file whose identity has not been matched can be obtained, so that the human body image files that are associated with the personnel trajectory corresponding to the human body image file whose identity has been matched currently and whose identities have not been matched can be obtained, and they are used as the associated files of the human body image file whose identity has been matched currently.

[0062] Through the above Step S231 and Step S232, the human body image files that are associated with the personnel trajectory corresponding to the human body image file whose identity has been matched and whose identities have not been matched can be effectively obtained, laying a foundation for re-matching the identity of the human body image files whose identities have not been matched subsequently.

[0063] In one embodiment, the above Step S240 performs secondary identity matching on the human body image files whose identities have not been matched according to the trajectory association relationship, including the following steps:

[0064] Step S241: Obtain the ID card images of the personnel who have an association relationship with the personnel corresponding to the human body image file whose identity has been matched as the associated ID card images;

[0065] Step S242: Perform secondary identity matching on the associated files according to the image feature similarity between the human body images in the associated files and the associated ID card images.

[0066] Specifically, since the associated ID card image is used for secondary identity matching with the associated file, and since the associated file has a trajectory association relationship with the human body image file whose identity has been matched, then the associated ID card image must be the ID card image corresponding to the person who has a trajectory association relationship with the person corresponding to the human body image whose identity has been matched. Since the identity of the person corresponding to the human body image file whose identity has been matched is already determined, based on this person's identity, the identity of the person having a trajectory association relationship with it can be determined. For example, the two persons having a trajectory association relationship can be colleagues, relatives, etc. Once the identity of the person having an association relationship is determined, the ID card image of the person having an association relationship with the person corresponding to the human body image file whose identity has been matched can be obtained and used as the associated ID card image. Compare the similarity between the image features of the human body image in the associated file and the associated ID card image to perform secondary identity matching on the associated file. For example, when the similarity of the image features between the human body image in the associated file and the associated ID card image exceeds the fifth set threshold, it is determined that the human body image in the current associated file has been matched to an identity, that is, the associated file has been matched to an identity. As one implementation, the fifth set threshold is less than the fourth set threshold. Since through the confirmation of the association relationship, the ID card image for secondary matching is controlled to be the associated ID card image, the problem of the decrease in accuracy caused by reducing the similarity threshold can be basically ignored.

[0067] Furthermore, in one embodiment, the above step S232 determines, according to the personnel trajectory, the human body image file that is associated with the personnel trajectory corresponding to the currently identity-matched human body image file and whose identity has not been matched, and uses it as the associated file of the currently identity-matched human body image file, including the following steps:

[0068] According to the personnel trajectories corresponding to all human body image files, calculate the number of spatio-temporal accompanying points of the personnel trajectory corresponding to the human body image file whose identity has not been matched and the personnel trajectory corresponding to the currently identity-matched human body image file;

[0069] Use the human body image file whose identity has not been matched and whose number of spatio-temporal accompanying points of the personnel trajectory corresponding to the currently identity-matched human body image file exceeds the first set threshold as the associated file of the currently identity-matched human body image file.

[0070] Specifically, spatio-temporal companionship means that if another person appears within a preset time range of the time point when a person appears at a certain place, then these two people are considered to have spatio-temporal companionship. Since the human body images in the image archives contain the capture time and the capture location, based on the personnel trajectories corresponding to the human body image archives with known identities and the personnel trajectories corresponding to the human body image archives without known identities, the number of spatio-temporal companion points between the personnel trajectory corresponding to the human body image archive without known identity and the personnel trajectory corresponding to the currently known human body image archive can be calculated. Based on the number of spatio-temporal companion points between the personnel trajectory corresponding to the human body image archive without known identity and the personnel trajectory corresponding to the human body image archive with known identity, the associated archives of the currently known human body image archive can be determined.

[0071] In one embodiment, the above step S241 of obtaining the ID card images of the personnel having an associated relationship with the personnel corresponding to the human body image archive with known identity as the associated ID card images includes the following steps:

[0072] Obtain the ID card images of the personnel belonging to the same administrative region as the personnel corresponding to the human body image archive with known identity, and the ID card images of the personnel having a social relationship with the personnel corresponding to the human body image archive with known identity as the associated ID card images; the social relationships include: friendship, family relationship or work relationship.

[0073] Specifically, the personnel belonging to the same administrative region as the personnel corresponding to the human body image archive with known identity and the personnel having a social relationship with the personnel corresponding to the human body image archive with known identity can be obtained through the registration information of the personnel corresponding to the human body image archive with known identity from various channels. The personnel belonging to the same administrative region as the personnel corresponding to the human body image archive with known identity and the personnel having a social relationship with the personnel corresponding to the human body image archive with known identity are necessarily the personnel having a trajectory association relationship with the personnel corresponding to the human body image archive with known identity. Therefore, the ID card images of the personnel belonging to the same administrative region as the personnel corresponding to the human body image archive with known identity and the ID card images of the personnel having a social relationship with the personnel corresponding to the human body image archive with known identity are effective choices as the associated ID card images.

[0074] Furthermore, in one embodiment, in the above steps, obtaining the ID card images of the personnel belonging to the same administrative region as the personnel corresponding to the human body image archive with known identity, and the ID card images of the personnel having a social relationship with the personnel corresponding to the human body image archive with known identity as the associated ID card images includes the following steps:

[0075] Obtain the ID card images of the persons who belong to the same administrative division as the person corresponding to the human body image file with the matched identity, and the ID card images of the persons who have social relations with the person corresponding to the human body image file with the matched identity;

[0076] From these, screen out the ID card images of the persons whose number of spatio-temporal companion points of the mobile phone mac trajectory of the person corresponding to the human body image file with the matched identity exceeds the second set threshold and whose number of spatio-temporal companion points of the vehicle trajectory of the person corresponding to the human body image file with the matched identity exceeds the third set threshold, and use them as associated ID card images.

[0077] Specifically, if the identity of the person corresponding to the human body image file with the matched identity is determined, the mobile phone mac trajectory of the person corresponding to the human body image file with the matched identity can be obtained according to the mac signal collected by the mobile phone mac collector, so that the mobile phone mac trajectory whose number of spatio-temporal companion points of the mobile phone mac trajectory of the person corresponding to the human body image file with the matched identity exceeds the second set threshold can be obtained. According to the registration information of the mobile phone, the identity of the person corresponding to this mac trajectory can be obtained. In addition, the vehicle trajectory of the person corresponding to the human body image file with the matched identity can be obtained according to the data collected by the traffic management department, so that the vehicle trajectory whose number of spatio-temporal companion points of the vehicle trajectory of the person corresponding to the human body image file with the matched identity exceeds the third set threshold can be obtained. According to the vehicle registration information, the identity of the person corresponding to this vehicle trajectory can be obtained.

[0078] Since the number of ID card images of the persons who belong to the same administrative division as the person corresponding to the human body image file with the matched identity, and the number of ID card images of the persons who have social relations with the person corresponding to the human body image file with the matched identity is too large, the workload of secondary identity matching is too large. Screening the ID card images for secondary matching according to the mobile phone mac trajectory and the vehicle trajectory can effectively reduce the number of associated ID card images, thereby effectively improving the identity matching efficiency.

[0079] In one embodiment, the above step S220 performs a first identity matching on the human body image files according to the image feature similarity between the human body images in the human body image files and the ID card images in the ID card database, including the following steps:

[0080] Step S221, screen out a preset number of human body images from each human body image file as preferred human body images according to the image quality;

[0081] Step S222, perform identity matching on each human body image file according to the image feature similarity between the preferred human body images and the ID card images in the ID card database.

[0082] Specifically, in order to better avoid the large quality difference of human body images in the same human body image file during the first identity matching of human body image files, a preset number of human body images are screened from each human body image file as preferred human body images according to the image quality. The preferred human body images in each human body image file are used as the selection for the first identity matching. Identity matching is performed on each human body image file according to the image feature similarity between the preferred human body images and the identity card images in the identity card database. As one implementation, Figure 3 is the flowchart of the first identity matching in the identity matching method, as Figure 3 shown. The first identity matching includes the following steps:

[0083] Step S310, according to the image quality, screen out a preset number of human body images from each human body image file as preferred human body images.

[0084] Step S320, compare the similarity between the features of the preferred human body images and the identity card images in the identity card database one by one, obtain the identity card image with the highest similarity and the highest similarity value corresponding to each preferred human body image, and obtain the highest value among all the highest similarity values, which is the optimal comparison value. The identity card image with the highest similarity corresponding to the optimal comparison value is the optimal matching identity card image.

[0085] Step S330, determine whether the optimal comparison value corresponding to the preferred human body image in the human body image file exceeds the sixth set threshold. If so, execute step S340; if not, execute step S350.

[0086] Step S340, determine that the identity matched by the current human body image file is the optimal matching identity card image.

[0087] Step S350, determine whether the optimal comparison value corresponding to the preferred human body image in the current human body image file exceeds the seventh set threshold but does not exceed the sixth set threshold. If so, execute step S360.

[0088] Step S360, determine whether the number of the highest similarity values that exceed the seventh set threshold but do not exceed the sixth set threshold among all the highest similarity values is greater than 2. If so, execute step S370.

[0089] Step S370, determine whether the identity card images with the highest similarity corresponding to the highest similarity values that exceed the seventh set threshold but do not exceed the sixth set threshold are the same identity card image. If so, execute step S380; if not, execute step S390.

[0090] Step S380: Determine that the identity corresponding to the currently matched human body image file is the identity card image with the highest similarity value corresponding to the similarity value that exceeds the seventh set threshold but does not exceed the sixth set threshold.

[0091] Step S390: Determine which identity card images are the identity card images with the highest similarity value corresponding to the similarity value that exceeds the seventh set threshold but does not exceed the sixth set threshold, and determine from them the identity card image with the largest number of the same identity card images corresponding to the preferred human body image. This identity card image is the identity corresponding to the currently matched human body image file.

[0092] For example, the current human body image file is File A. Five preferred human body images are selected from File A, namely a, b, c, d, and e. Among them, the sixth set threshold is 90%, and the seventh set threshold is 80%. If the similarity between a and the ID card image m is the highest, with a similarity of 92%, the similarity between b and the ID card image m is the highest, with a similarity of 85%, the similarity between c and the ID card image m is the highest, with a similarity of 83%, the similarity between d and the ID card image m is the highest, with a similarity of 84%, and the similarity between e and the ID card image n is the highest, with a similarity of 86%. Since the highest value among all the highest similarity values is 92%, which is the optimal comparison value, the ID card image m with the highest similarity corresponding to the optimal comparison value is the optimal matching ID card image. If the similarity between a and the ID card image m is the highest, with a similarity of 82%, the similarity between b and the ID card image m is the highest, with a similarity of 85%, the similarity between c and the ID card image m is the highest, with a similarity of 83%, the similarity between d and the ID card image m is the highest, with a similarity of 84%, and the similarity between e and the ID card image m is the highest, with a similarity of 86%, then the number of the highest similarity values that exceed the seventh set threshold of 80% but do not exceed the sixth set threshold of 90% is greater than 2, and the ID card image m with the highest similarity corresponding to the highest similarity value that exceeds the seventh set threshold of 80% but does not exceed the sixth set threshold of 90% is the same ID card image m, then it is determined that the identity matched by the current human body image file is the ID card image m. If the similarity between a and the ID card image m is the highest, with a similarity of 82%, the similarity between b and the ID card image m is the highest, with a similarity of 85%, the similarity between c and the ID card image m is the highest, with a similarity of 83%, the similarity between d and the ID card image n is the highest, with a similarity of 84%, and the similarity between e and the ID card image n is the highest, with a similarity of 86%, then the ID card images m and n correspond to the highest similarity values that exceed the seventh set threshold of 80% but do not exceed the sixth set threshold of 90%. There are three preferred human body images a, b, and c corresponding to the ID card image m, and two preferred human body images d and e corresponding to the ID card image n. Then the ID card image m is the ID card image with the largest number of preferred human body images corresponding to the same ID card image. Then it is determined that the ID card image m is the identity matched by the current human body image file.

[0093] As one of the implementation manners, step S210 clusters the human body images of the identity to be matched, including the following steps:

[0094] Cluster the human body images of the identity to be matched according to the image feature similarity between the human body images of the identity to be matched.

[0095] Specifically, the human body images to be matched for identity can be compared pairwise for the similarity of image features, and the human body images with the image feature similarity exceeding the eighth set threshold between pairwise are clustered into the same human body image file. It is also possible to obtain the previously existing human body image files, including the human body image files with the identity already matched and the human body image files with the identity not yet matched, compare the image features of the human body images to be matched for identity with the file features of the previously existing human body image files, and add the human body images with the similarity greater than the ninth set threshold to the current human body image file. If the similarity threshold between the image features of the current human body image and the file features of all the previously existing human body image files does not exceed the ninth set threshold, a new human body image file is created, and the current human body image is added to the newly created human body image file. Among them, the file feature of the human body image file is the average image feature of all the human body images in the human body image file or the image feature of the clearest human body image in the human body image file is used as the file feature of the current human body image file.

[0096] This embodiment also provides an identity matching method, and the method includes the following steps:

[0097] Step S410: Cluster the human body images to be matched for identity according to the similarity of image features between the human body images to obtain the human body image files corresponding to the identities in the ID card database.

[0098] Step S420: Screen out a preset number of human body images from each human body image file as the preferred human body images according to the image quality.

[0099] Step S430: Perform the first identity matching on each human body image file according to the similarity of image features between the preferred human body images and the ID card images in the ID card database.

[0100] Step S440: Based on the human body image files, obtain the personnel trajectories corresponding to each human body image file.

[0101] Step S450: According to the personnel trajectories corresponding to all the human body image files, calculate the number of spatio-temporal companion points between the personnel trajectory corresponding to the human body image file with the identity not yet matched and the personnel trajectory corresponding to the human body image file with the identity already matched currently.

[0102] Step S460: Use the human body image files with the identity not yet matched and the number of spatio-temporal companion points exceeding the first set threshold between the personnel trajectory corresponding to the human body image file with the identity already matched currently as the associated files of the human body image file with the identity already matched currently.

[0103] Step S470: Obtain the ID card images of the persons who belong to the same administrative division as the person corresponding to the human body image file with the matched identity, and the ID card images of the persons who have social relations with the person corresponding to the human body image file with the matched identity.

[0104] Step S480: Screen out the ID card images of the persons whose number of spatio-temporal accompanying points of the mobile phone mac trajectory of the person corresponding to the human body image file with the matched identity exceeds the second set threshold and whose number of spatio-temporal accompanying points of the vehicle trajectory of the person corresponding to the human body image file with the matched identity exceeds the third set threshold as the associated ID card images.

[0105] Step S490: Perform secondary identity matching on the associated files according to the image feature similarity between the human body images in the associated files and the associated ID card images.

[0106] Figure 4 It is a schematic diagram of the identity matching device in an embodiment of the present invention. As Figure 4 shown, an identity matching device 30 is provided. The device includes a clustering module 31, a first matching module 32, an association module 33, and a second matching module 34:

[0107] The clustering module 31 is configured to cluster the human body images to be matched for identity to obtain human body image files corresponding to the identities in the ID card database.

[0108] The first matching module 32 is configured to perform a first identity matching on the human body image files according to the image feature similarity between the human body images in the human body image files and the ID card images in the ID card database.

[0109] The association module 33 is configured to, if there are human body image files that have not been matched for identity after the first identity matching, obtain the trajectory association relationship between the trajectory of the person corresponding to the currently matched human body image file and the trajectory of the person corresponding to the human body image file that has not been matched for identity.

[0110] The second matching module 34 is configured to perform secondary identity matching on the human body image files that have not been matched for identity according to the trajectory association relationship.

[0111] The above identity matching device 30 clusters the human body images of the identities to be matched to obtain human body image files corresponding to the identities in the ID card database; based on the image feature similarity between the human body images in the human body image files and the ID card images in the ID card database, performs a first identity match on the human body image files; if there are human body image files that have not been matched after the first identity match, obtains the trajectory association relationship between the personnel trajectories corresponding to the human body image files that have been matched to identities and the personnel trajectories corresponding to the human body image files that have not been matched to identities; and based on the trajectory association relationship, performs a second identity match on the human body image files that have not been matched to identities. By mining the trajectory association relationship between the personnel trajectories corresponding to the human body image files that have been matched to identities and the personnel trajectories corresponding to the human body image files that have not been matched to identities, and performing a second identity match on the human body image files that have not been matched to identities, the probability of identity matching is effectively improved.

[0112] In one embodiment, the trajectory association relationship includes associated files, and the association module 33 is further configured to obtain the personnel trajectories corresponding to each human body image file based on the human body image files;

[0113] Based on the personnel trajectories, determines the human body image files that are associated with the personnel trajectory corresponding to the human body image file that has been matched to an identity and have not been matched to an identity, and uses them as the associated files of the human body image file that has been matched to an identity.

[0114] In one embodiment, the second matching module 34 is further configured to obtain the ID card images of the personnel associated with the personnel corresponding to the human body image file that has been matched to an identity as the associated ID card images;

[0115] Based on the image feature similarity between the human body images in the associated files and the associated ID card images, performs a second identity match on the associated files.

[0116] In one embodiment, the association module 33 is further configured to calculate the number of spatio-temporal accompanying points between the personnel trajectory corresponding to the human body image file that has not been matched to an identity and the personnel trajectory corresponding to the human body image file that has been matched to an identity based on the personnel trajectories corresponding to all human body image files;

[0117] Uses the human body image files that have not been matched to an identity and the number of spatio-temporal accompanying points of the personnel trajectory corresponding to the human body image file that has been matched to an identity exceeds a first set threshold as the associated files of the human body image file that has been matched to an identity.

[0118] In one embodiment, the second matching module 34 is further configured to obtain the ID card images of the persons belonging to the same administrative division as the person corresponding to the human body image file with the matched identity, and the ID card images of the persons having a social relationship with the person corresponding to the human body image file with the matched identity, as associated ID card images; the social relationships include: friendship, family relationship or work relationship.

[0119] In one embodiment, the second matching module 34 is further configured to obtain the ID card images of the persons belonging to the same administrative division as the person corresponding to the human body image file with the matched identity, and the ID card images of the persons having a social relationship with the person corresponding to the human body image file with the matched identity;

[0120] From these, the ID card images of the persons whose number of spatio-temporal companion points of the mobile phone mac trajectory corresponding to the person corresponding to the human body image file with the matched identity exceeds a second set threshold and whose number of spatio-temporal companion points of the vehicle trajectory corresponding to the person corresponding to the human body image file with the matched identity exceeds a third set threshold are selected as associated ID card images.

[0121] In one embodiment, the first matching module 32 is further configured to screen out a preset number of human body images from each human body image file as preferred human body images according to the image quality;

[0122] Perform identity matching on each human body image file according to the image feature similarity between the preferred human body images and the ID card images in the ID card database.

[0123] In one embodiment, the clustering module 31 is further configured to cluster the human body images to be matched according to the image feature similarity between the human body images to be matched.

[0124] It should be noted that the above-mentioned each module can be a functional module or a program module, and can be implemented either by software or by hardware. The above-mentioned each module can be embedded in the processor in the computer device in the form of hardware or be independent of it, or can be stored in the memory in the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to the above-mentioned each module.

[0125] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 5As shown in the figure. The computer device includes a processor, a memory, a network interface, and a database connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store a set of preset configuration information. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the above-mentioned identity matching method is implemented.

[0126] In one embodiment, a computer device is provided. The computer device may be a terminal. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, an identity matching method is implemented. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device may be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0127] Those skilled in the art can understand that Figure 5 the structure shown in the figure is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have a different component layout.

[0128] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0129] Cluster the human body images to be matched for identity to obtain human body image files corresponding to the identities in the identity database of ID cards;

[0130] Perform a first identity match on the human body image files according to the image feature similarity between the human body images in the human body image files and the ID card images in the identity database of ID cards;

[0131] If there are human body image files for which identities are not matched after the first identity match, obtain the trajectory association relationship between the personnel trajectory corresponding to the human body image file for which the identity has been matched and the personnel trajectory corresponding to the human body image file for which the identity has not been matched;

[0132] According to the trajectory association relationship, perform a secondary identity match on the human body image file for which the identity has not been matched.

[0133] In one embodiment, the trajectory association relationship includes associated files, and when the processor executes the computer program, the following steps are further implemented:

[0134] Based on the human body image files, obtain the personnel trajectories corresponding to each human body image file;

[0135] According to the personnel trajectories, determine the human body image files that are associated with the personnel trajectory corresponding to the human body image file for which the identity has been matched and for which the identity has not been matched, and use them as the associated files of the human body image file for which the identity has been matched.

[0136] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0137] Obtain the ID card images of the personnel associated with the personnel corresponding to the human body image file for which the identity has been matched as the associated ID card images;

[0138] According to the image feature similarity between the human body images in the associated files and the associated ID card images, perform a secondary identity match on the associated files.

[0139] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0140] According to the personnel trajectories corresponding to all human body image files, calculate the number of spatio-temporal accompanying points between the personnel trajectory corresponding to the human body image file for which the identity has not been matched and the personnel trajectory corresponding to the human body image file for which the identity has been matched;

[0141] Use the human body image files for which the identity has not been matched and for which the number of spatio-temporal accompanying points with the personnel trajectory corresponding to the human body image file for which the identity has been matched exceeds the first set threshold as the associated files of the human body image file for which the identity has been matched.

[0142] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0143] Obtain the ID card images of the persons who belong to the same administrative division as the person corresponding to the human body image file with the matched identity, and the ID card images of the persons who have social relationships with the person corresponding to the human body image file with the matched identity, as associated ID card images; the social relationships include: friendship, family relationship or work relationship.

[0144] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0145] Obtain the ID card images of the persons who belong to the same administrative division as the person corresponding to the human body image file with the matched identity, and the ID card images of the persons who have social relationships with the person corresponding to the human body image file with the matched identity;

[0146] From them, screen out the ID card images of the persons whose number of spatio-temporal companion points of the mobile phone mac trajectory of the person corresponding to the human body image file with the matched identity exceeds the second set threshold and whose number of spatio-temporal companion points of the vehicle trajectory of the person corresponding to the human body image file with the matched identity exceeds the third set threshold, as associated ID card images.

[0147] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0148] According to the image quality, screen out a preset number of human body images from each human body image file as preferred human body images;

[0149] Perform identity matching on each human body image file according to the image feature similarity between the preferred human body images and the ID card images in the ID card database.

[0150] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0151] Cluster the human body images to be matched for identity according to the image feature similarity between the human body images to be matched for identity.

[0152] The above storage medium clusters the human body images to be matched by identity, and obtains human body image files corresponding to the identities in the identity card database; performs a first identity match on the human body image files according to the image feature similarity between the human body images in the human body image files and the identity card images in the identity card database; if there are human body image files that have not been matched by identity after the first identity match, obtains the trajectory association relationship between the personnel trajectories corresponding to the human body image files that have been matched by identity and the personnel trajectories corresponding to the human body image files that have not been matched by identity; and performs a second identity match on the human body image files that have not been matched by identity according to the trajectory association relationship. By mining the trajectory association relationship between the personnel trajectories corresponding to the human body image files that have been matched by identity and the personnel trajectories corresponding to the human body image files that have not been matched by identity, and performing a second identity match on the human body image files that have not been matched by identity, the probability of identity match is effectively increased.

[0153] It should be understood that the specific embodiments described herein are only used to explain this application and not to limit it. According to the embodiments provided in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0154] Obviously, the drawings are only some examples or embodiments of the present application. For those of ordinary skill in the art, the present application can also be applied to other similar situations based on these drawings without creative efforts. Additionally, it can be understood that although the work done during this development process may be complex and time-consuming, for those of ordinary skill in the art, certain design, manufacturing, or production changes based on the technical content disclosed in the present application are only conventional technical means and should not be regarded as insufficient disclosure of the present application.

[0155] The term "embodiment" in the present application means that the specific features, structures, or characteristics described in connection with the embodiment may be included in at least one embodiment of the present application. The phrase appears in various positions in the specification and does not necessarily mean the same embodiment, nor does it mean independence or alternative to other embodiments that are mutually exclusive. Those of ordinary skill in the art can clearly or implicitly understand that the embodiments described in the present application can be combined with other embodiments without conflict.

[0156] The above-described embodiments only represent several implementation manners of the present application. Their descriptions are relatively specific and detailed, but should not be construed as limiting the scope of patent protection. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. An identity matching method, characterized in that, The method includes the following steps: Cluster the human body images to be matched for identity to obtain human body image files corresponding to the identities in the ID card database; According to the image quality, screen out a preset number of human body images from each of the human body image files as target human body images; Perform a first identity match on each of the human body image files according to the image feature similarity between the target human body images and the ID card images in the ID card database; If after the first identity match, there are human body image files that have not been matched to an identity, obtain the trajectory association relationship between the personnel trajectories corresponding to the human body image files that have been matched to an identity and the personnel trajectories corresponding to the human body image files that have not been matched to an identity; Perform a second identity match on the human body image files that have not been matched to an identity according to the trajectory association relationship.

2. The identity matching method according to claim 1, wherein The trajectory association relationship includes associated files; if after the first identity match, there are human body image files that have not been matched to an identity, obtaining the trajectory association relationship between the personnel trajectories corresponding to the human body image files that have been matched to an identity and the personnel trajectories corresponding to the human body image files that have not been matched to an identity includes: Based on the human body image files, obtain the personnel trajectories corresponding to each of the human body image files; According to the personnel trajectories, determine the human body image files that are associated with the personnel trajectory corresponding to the human body image file that has been matched to an identity and have not been matched to an identity, and use them as the associated files of the human body image file that has been matched to an identity.

3. The identity matching method according to claim 2, wherein The performing a second identity match on the human body image files that have not been matched to an identity according to the trajectory association relationship includes the following steps: Obtain the ID card images of the personnel who have an association relationship with the personnel corresponding to the human body image files that have been matched to an identity as associated ID card images; Perform a second identity match on the associated files according to the image feature similarity between the human body images in the associated files and the associated ID card images.

4. The identity matching method according to claim 2, characterized in that The determining the human body image files that are associated with the personnel trajectory corresponding to the human body image file that has been matched to an identity and have not been matched to an identity according to the personnel trajectories and using them as the associated files of the human body image file that has been matched to an identity includes the following steps: According to the personnel trajectories corresponding to all the human body image files, calculate the number of spatio-temporal accompanying points of the personnel trajectory corresponding to the human body image file that has not been matched to an identity and the personnel trajectory corresponding to the human body image file that has been matched to an identity; Use the human body image files that have not been matched to an identity and whose number of spatio-temporal accompanying points of the personnel trajectory corresponding to the human body image file that has been matched to an identity exceeds the first set threshold as the associated files of the human body image file that has been matched to an identity.

5. The identity matching method according to claim 3, wherein The obtaining the ID card images of the personnel who have an association relationship with the personnel corresponding to the human body image files that have been matched to an identity as associated ID card images includes the following steps: Obtain the ID card images of persons belonging to the same administrative division as the person corresponding to the human body image file with the identity already matched, and the ID card images of persons having a social relationship with the person corresponding to the human body image file with the identity already matched, as associated ID card images; the social relationships include: friendship, kinship or working relationship.

6. The identity matching method according to claim 5, wherein The obtaining of the ID card images of persons belonging to the same administrative division as the person corresponding to the human body image file with the identity already matched, and the ID card images of persons having a social relationship with the person corresponding to the human body image file with the identity already matched, as associated ID card images, includes the following steps: Obtain the ID card images of persons belonging to the same administrative division as the person corresponding to the human body image file with the identity already matched, and the ID card images of persons having a social relationship with the person corresponding to the human body image file with the identity already matched; Screen out the ID card images of persons whose number of spatio-temporal companion points of the mobile phone mac trajectory of the person corresponding to the human body image file with the identity already matched exceeds a second set threshold and whose number of spatio-temporal companion points of the vehicle trajectory of the person corresponding to the human body image file with the identity already matched exceeds a third set threshold, as the associated ID card images.

7. The identity matching method according to any one of claims 1 to 6, characterized in that, The clustering of the human body images with the identity to be matched includes the following steps: Cluster the human body images with the identity to be matched according to the image feature similarity between the human body images with the identity to be matched.

8. An identity matching device, characterized in that, The device includes: a clustering module, a first matching module, an association module and a second matching module: The clustering module is used to cluster the human body images with the identity to be matched to obtain human body image files corresponding to the identities in the ID card database; The first matching module is used to screen out a preset number of human body images from each of the human body image files as target human body images according to the image quality; Perform a first identity matching on each of the human body image files according to the image feature similarity between the target human body images and the ID card images in the ID card database; The association module is used to, if there are human body image files that have not been matched with an identity after the first identity matching, obtain the trajectory association relationship between the trajectory of the person corresponding to the currently matched human body image file and the trajectory of the person corresponding to the human body image file that has not been matched with an identity; The second matching module is used to perform a secondary identity matching on the human body image files that have not been matched with an identity according to the trajectory association relationship.

9. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images