Threshold determination method, target person identification method, device, equipment and medium

Abstract

The embodiments of the present application disclose a threshold determination method, a target person identification method, an apparatus, a device and a medium. The method comprises: determining at least one candidate similarity between a standard image of at least one candidate person in an information database and a standard image of a target person; selecting at least one target similarity from the at least one candidate similarity; and determining a similarity threshold associated with the target person based on the at least one target similarity. The above scheme can adaptively determine different similarity thresholds for different target persons, and is used to determine the target person from the monitored persons based on the similarity, thereby improving the accuracy and reference of the similarity threshold, and further improving the recognition rate and accuracy of the target person identification.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of image recognition technology, and in particular to a threshold determination method, a target person recognition method, a device, a equipment and a medium. Background Art

[0002] Face recognition is a biometric technology that identifies people based on their facial features. It uses a camera or camcorder to capture images or video streams containing faces, and automatically detects and tracks faces in the images, and then performs face recognition on the detected faces.

[0003] Currently, facial recognition is limited by factors such as algorithm accuracy, the number of personnel database lists, and the number of monitored personnel. As a result, the accuracy of personnel identification and monitoring is not high enough, which can easily lead to misidentification and mis-monitoring. For example, the similarity between each person and the person themselves may be different, making it impossible to accurately identify the target person through image similarity matching. Summary of the invention

[0004] The embodiments of the present invention provide a threshold determination method, a target person identification method, an apparatus, a device and a medium to determine a similarity threshold according to the adaptability of different target persons, thereby improving the recognition rate and accuracy of target person identification.

[0005] In one embodiment, the present application provides a threshold determination method, the method comprising:

[0006] determining at least one candidate similarity between a standard image of at least one candidate person in the information database and a standard image of a target person;

[0007] Selecting at least one target similarity from the at least one candidate similarity;

[0008] A similarity threshold associated with the target person is determined based on the at least one target similarity.

[0009] In another embodiment, the present application provides a method for identifying a target person, the method comprising:

[0010] Determining the real-time similarity between the monitored image of the monitored person and the standard image of the target person;

[0011] If the real-time similarity is greater than the similarity threshold, determining that the monitored person is a target person;

[0012] The similarity threshold is determined according to the threshold determination method described in any embodiment of the present application.

[0013] In yet another embodiment, the present application also provides a threshold determination device, the device comprising:

[0014] a candidate similarity determination module, configured to determine at least one candidate similarity between a standard image of at least one candidate person in the information database and a standard image of a target person;

[0015] A target similarity selection module, used to select at least one target similarity from the at least one candidate similarity;

[0016] The similarity threshold determination module is used to determine the similarity threshold associated with the target person according to the at least one target similarity.

[0017] In yet another embodiment, the present application also provides a target person identification device, the device comprising:

[0018] A real-time similarity determination module, used to determine the real-time similarity between the monitoring image of the monitored person and the standard image of the target person;

[0019] A target person determination module, configured to determine that the monitored person is a target person if the real-time similarity is greater than the similarity threshold;

[0020] The similarity threshold is determined according to the threshold determination method described in any embodiment of the present application.

[0021] In one embodiment, the present application also provides an electronic device, including: one or more processors;

[0022] A memory for storing one or more programs;

[0023] When the one or more programs are executed by the one or more processors, the one or more processors implement the threshold determination method described in any embodiment of the present application, or implement the target person identification method described in any embodiment of the present application.

[0024] In one embodiment, the present application also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, it implements the threshold determination method described in any embodiment of the present application, or implements the target person identification method described in any embodiment of the present application.

[0025] In an embodiment of the present application, by determining at least one candidate similarity between a standard image of at least one candidate person in an information database and a standard image of a target person; selecting at least one target similarity from the at least one candidate similarity; and determining a similarity threshold associated with the target person based on the at least one target similarity, the similarity threshold can be set adaptively and specifically for different target persons, thereby solving the problem of inaccurate target person identification caused by setting a fixed single similarity threshold, improving the accuracy and referenceability of the similarity, and thereby improving the recognition rate and accuracy of target person identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A flowchart of a threshold determination method provided by an embodiment of the present invention;

[0027] Figure 2 A flowchart of a threshold determination method provided by another embodiment of the present invention;

[0028] Figure 3 A flow chart of a target person identification method provided by an embodiment of the present invention;

[0029] Figure 4 A schematic diagram of the structure of a threshold determination device provided by an embodiment of the present invention;

[0030] Figure 5 A schematic diagram of the structure of a target person identification device provided by an embodiment of the present invention;

[0031] Figure 6 A schematic diagram of the structure of a threshold determination device provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0032] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0033] Figure 1 A flowchart of a threshold determination method provided by an embodiment of the present invention. The threshold determination method provided by this embodiment can be applied to the case where a similarity threshold is determined to determine a target person based on the similarity. Typically, this method can be applied to the case where, for different target persons, a similarity threshold associated with each target person is adaptively determined to screen the target person from the monitored persons based on the similarity threshold. The method can be specifically executed by a threshold determination device, which can be implemented by software and / or hardware, and the device can be integrated in a threshold determination device. See Figure 1 , the method of the embodiment of the present application specifically includes:

[0034] S110, determining at least one candidate similarity between a standard image of at least one candidate person in the information database and a standard image of a target person.

[0035] Among them, the candidate person can be any person in the information database, and the candidate person can include the target person. The number of candidate persons can be set according to the actual situation. In the embodiment of the present application, in order to improve the similarity between the target person and the person in the information database more comprehensively, the candidate person takes all the persons in the information database. . The target person can be a person who needs to be identified and tracked, or all the persons in the information database can be taken as target persons in turn, and the scheme in the embodiment of the present application is executed to determine the similarity threshold associated with each target person. The standard image can be an image with obvious facial features such as an ID photo. In the embodiment of the present application, an information database can be established in advance, which includes detailed information of the target person, such as age, height, weight, gender and standard image, etc., and is stored in the information database. The personnel in the information database can be teachers and students of the same school, employees of the same company, etc. For each person in the information database, it can be used as a target person, and at least one candidate similarity between the standard image of at least one candidate person in the information database and the standard image of the target person is determined, so as to establish a multidimensional information database for each target person in the information database.

[0036] Exemplarily, for each standard image of a candidate person, the similarity between the standard image of the target person and the standard image of the target person is calculated one by one to obtain at least one candidate similarity. Since the target person may have feature similarities with multiple persons and is difficult to identify, in the embodiment of the present application, at least one standard image of a candidate person in the information library is obtained in advance, and the candidate similarity between the standard image of the target person and the standard image is calculated, so as to analyze the similarity between the target person and the candidate person, so as to facilitate the subsequent identification of the target person.

[0037] S120: Select at least one target similarity from the at least one candidate similarity.

[0038] Specifically, not all candidates in the information database have similarities with the target person. Only some of the candidates may have similarities with the target person, while other candidates are obviously not the same person as the target person. Therefore, not all candidate similarities are suitable for similarity analysis. Selecting at least one target similarity from at least one candidate similarity can meet the requirements of similarity analysis between the candidate and the target person.

[0039] In the embodiment of the present application, a candidate similarity with a higher similarity can be selected as the target similarity. Exemplarily, a candidate similarity with a similarity greater than a preset value can be selected as the target similarity, or a preset number of candidate similarities with a larger similarity can be selected as the target similarity. The number of target similarities can be less than the number of candidate similarities.

[0040] Table 1

[0041] Target Personnel gender height Body shape color Candidates Target Similarity Xiao Ming male B M yellow Xiao Ming (myself) S1 Xiao Ming male B M yellow Xiaowen S2 Xiao Ming male B M yellow Xiao Cheng S3 Xiao Ming …… …… …… …… …… ……

[0042] Exemplarily, the selected target similarity can be as shown in Table 1, and there may be multiple candidates for similarity comparison with the target person. For the target person Xiao Ming, not all candidates may have similarity with him, so only the candidates with greater similarity with the target person Xiao Ming and the corresponding target similarity are selected. Among them, the height and body shape can be represented by letters to indicate their corresponding degree. For example, when the height is greater than or equal to 2 meters, it is represented by the letter A, when the height is greater than or equal to 1.9 meters and less than 2 meters, it is represented by the letter B, when the height is greater than or equal to 1.8 meters and less than 1.9 meters, it is represented by the letter C, when the height is greater than or equal to 1.7 meters and less than 1.8 meters, it is represented by the letter D, when the height is greater than or equal to 1.65 meters and less than 1.7 meters, it is represented by the letter E, when the height is greater than or equal to 1.6 meters and less than 1.65 meters, it is represented by the letter F... For body shape, it can be set to M for thin, MM for normal, MMM for fat, MMMM for obesity... The skin color can be yellow, black, white or other colors. Among them, the target similarity S1 can be 95%, S2 can be 94%, and S3 can be 92%. It should be noted that the above specific numbers and tables are just examples, and the specific numerical expression form can be determined according to actual conditions. The specific values ​​in the table can be determined according to actual image recognition and similarity calculation.

[0043] In the embodiment of the present application, the target similarity may include facial similarity and body similarity, as shown in Table 2.

[0044] Table 2

[0045] Target Personnel gender height Body shape color Candidates Facial similarity Body similarity Xiao Ming male B M yellow Xiao Ming (myself) S1 L1 Xiao Ming male B M yellow Xiaowen S2 L2 Xiao Ming male B M yellow Xiao Cheng S3 L3 Xiao Ming …… …… …… …… …… …… ……

[0046] In the embodiment of the present application, the target similarity may be facial similarity, body similarity, or a similarity obtained by combining facial similarity and body similarity.

[0047] S130: Determine a similarity threshold associated with the target person according to the at least one target similarity.

[0048] At present, the scheme for identifying target persons generally sets a fixed similarity threshold in advance, and then compares the similarity between the monitored image of the monitored person and the standard image of the target person with the pre-set similarity threshold, so as to determine whether the monitored person is the target person. However, for different target persons, the similarity between the monitored image of the target person collected by the monitoring equipment and the standard image of the target person may be different, and setting a unified similarity threshold standard may not be able to accurately screen out the target person. In addition, for the target person, there may be multiple candidate persons similar to him, and setting a unified and fixed similarity threshold cannot accurately determine which candidate person is the target person.

[0049] In the embodiment of the present application, for different target persons, the similarity threshold associated with the target person is adaptively determined according to the similarity of the candidate persons similar to the target person, so that the similarity threshold can be applicable to the screening of the target person. Specifically, according to at least one target similarity, the similarity threshold associated with the target person is determined, and the average value of at least one target similarity can be taken as the similarity threshold. It can also be to take other combined values ​​of at least one target similarity as the similarity threshold. It can also be to determine the appropriate similarity threshold in advance according to the recognition accuracy of the target person, and then fit the calculation formula of the similarity threshold according to the relationship between the determined similarity threshold and the target similarity. The beneficial effect of the above scheme is that, for different target persons, the similarity threshold associated with them is determined for the identification of the target person, thereby meeting the different requirements of the similarity of different target persons, and for different target persons, the target person can be accurately screened from the monitored persons according to the associated similarity threshold.

[0050] In an embodiment of the present application, determining the similarity threshold associated with the target person according to the at least one target similarity includes: determining the similarity threshold associated with the target person according to the target similarity and the following formula:

[0051] K=P1-[P1-(P2+P3+……+Pn) / n]*V;

[0052] Wherein, K is the similarity threshold, P1 is the maximum target similarity, P2-Pn is the target similarity less than the maximum target similarity, n is the number of target similarities, and V is the optimal offset coefficient.

[0053] Among them, V can be determined according to actual conditions. For example, in the actual detection process, the similarity threshold is determined according to the recognition rate and accuracy of the target person, and the relationship between the similarity threshold and the target similarity is fitted to calculate the optimal offset coefficient V to form the above formula. Then, according to the above formula and other target similarities, the similarity threshold associated with the target person is determined, as shown in Table 3.

[0054] Table 3

[0055] Name Identification threshold Xiao Ming K1 Xiaolan K2 Little Dragon K3 Xiaojie K4 …… ……

[0056] In an embodiment of the present application, by determining at least one candidate similarity between a standard image of at least one candidate person in an information database and a standard image of a target person; selecting at least one target similarity from the at least one candidate similarity; and determining a similarity threshold associated with the target person based on the at least one target similarity, the similarity threshold can be set adaptively and specifically for different target persons, thereby solving the problem of inaccurate target person identification caused by setting a fixed single similarity threshold, improving the accuracy and referenceability of the similarity, and thereby improving the recognition rate and accuracy of target person identification.

[0057] Figure 2 This is a flowchart of a threshold determination method provided by another embodiment of the present invention. This embodiment of the present application optimizes the above embodiment based on the above embodiment. For details not described in detail in this embodiment, please refer to the above embodiment. Figure 2 , the threshold determination method provided in this embodiment may include:

[0058] S210: Determine at least one candidate similarity between a standard image of at least one candidate person in the information database and a standard image of a target person.

[0059] S220: Determine a maximum candidate similarity among at least one candidate similarity.

[0060] Exemplarily, among the candidate similarities, the candidate similarities with greater similarity have a greater reference for determining the similarity threshold, while the candidate similarities with smaller similarity may not be used as reference data. Therefore, in the embodiment of the present application, the maximum candidate similarity among the candidate similarities is selected as the reference data for determining the similarity threshold, so as to more accurately determine the similarity threshold.

[0061] S230, determining whether the number of best similarities among at least one candidate similarity whose difference with the maximum candidate similarity is less than a preset similarity difference is greater than or equal to a preset number, if so, executing S240; if not, executing S250.

[0062] Specifically, the target similarity is selected according to the difference between the candidate similarity and the maximum candidate similarity. The preset similarity difference is set in advance according to the actual situation, and the target similarity is selected from the candidate similarities whose difference with the maximum candidate similarity is within the preset similarity difference. The number of the specifically selected target similarities can be determined according to the number of the candidate similarities whose difference with the maximum candidate similarity is within the preset similarity difference.

[0063] S240: Select a preset number of best similarities as target similarities.

[0064] Among them, the preset number can be set according to actual conditions. Exemplarily, if the preset number is 5, the preset similarity difference is set to 5%, the maximum candidate similarity is 95%, and the best similarities with a difference of less than 5% from the maximum candidate similarity include P1=95%, P2=94%, P3=93%, P4=92.5%, P5=92%, and P6=91%. The number of best similarities with a difference of less than 5% from the maximum candidate similarity is 6, which is greater than 5. Therefore, 5 best similarities with a difference of less than 5% from the maximum candidate similarity are selected as target similarities. The beneficial effect of the above scheme is that a person with a higher similarity and a close similarity to the target person is selected as the target similarity, so that the similarity threshold determined according to the target similarity is more referenceable and can more effectively separate the target person from other persons.

[0065] In the embodiment of the present application, selecting a preset number of best similarities as the target similarities includes: sorting the best similarities in descending order; and determining the preset number of best similarities at the front of the sort as the target similarities.

[0066] For example, in the above example, the similarity is ranked as P1>P2>P3>P4>P5>P6, so the top five best similarities, namely P1, P2, P3, P4 and P5, are selected as the target similarities. By selecting the best similarity with the largest similarity as the target similarity, the similarity threshold determined according to the target similarity is more referenceable.

[0067] S250: Taking all the best similarities as target similarities.

[0068] Exemplarily, in the above example, if the candidate similarities whose difference with the maximum candidate similarity is less than 5% include P1=95%, P2=94%, P3=93%, P4=92.5%, which is less than the preset number 5, then all the best similarities, namely P1, P2, P3 and P4, are used as target similarities.

[0069] S260: Determine a similarity threshold associated with the target person according to the at least one target similarity.

[0070] This step is performed after S240 or S250 is completed.

[0071] The technical solution of the embodiment of the present application selects a person with a high similarity and a close similarity to the target person as the target similarity, thereby making the similarity threshold determined according to the target similarity more referenceable and more able to effectively separate the target person from other persons.

[0072] Figure 3 A flowchart of a target person identification method provided by an embodiment of the present invention. The target person identification method provided by this embodiment can be applied to the situation of determining the target person from the monitoring scene. Typically, this method can be applied to the situation of adaptively determining the similarity threshold associated with each target person for different target persons, so as to screen the target person from the monitored persons according to the similarity threshold. The method can be specifically executed by a target person identification device, which can be implemented by software and / or hardware, and the device can be integrated in a similarity target person identification device. See Figure 3 , the method of the embodiment of the present application specifically includes:

[0073] S310: Determine the real-time similarity between the monitoring image of the monitored person and the standard image of the target person.

[0074] Exemplarily, in an actual monitoring scenario, a monitoring image of a monitored person in a monitoring screen is obtained, and a real-time similarity between the monitoring image and a standard image of the target person is determined, thereby determining whether the monitored person is the target person based on the real-time similarity.

[0075] S320: If the real-time similarity is greater than the similarity threshold, determine that the monitored person is a target person.

[0076] The similarity threshold is determined according to the threshold determination method provided in any of the above embodiments.

[0077] If the real-time similarity is greater than or equal to the similarity threshold, it means that the monitored person is the target person. If the real-time similarity is less than the similarity threshold, it is determined that the monitored person is not the target person.

[0078] In the embodiment of the present application, the similarity threshold can be set adaptively and specifically for different target persons, which solves the problem of inaccurate target person identification caused by setting a fixed single similarity threshold, improves the accuracy and referenceability of the similarity, and further improves the recognition rate and accuracy of target person identification.

[0079] In an embodiment of the present application, the similarity threshold is a facial similarity threshold; accordingly, after determining the similarity threshold associated with the target person based on the at least one target similarity, the method further includes: determining the facial similarity between the monitored person and the target person based on the monitored image of the monitored person and the standard image of the target person; if there are at least two monitored persons whose facial similarities with the target person are greater than the facial similarity threshold, determining the target person from at least two monitored persons based on the body similarity between the monitored person and the target person.

[0080] Specifically, in the embodiment of the present application, the similarity threshold may be facial similarity, and correspondingly, the candidate similarity may be candidate facial similarity, the target similarity may be target facial similarity, and the best similarity may be best facial similarity. In an actual monitoring scenario, a monitoring image of the monitored person in the monitoring screen is obtained, and the facial similarity between the monitoring image and the standard image of the target person is determined, so as to determine whether the monitored person is the target person according to the facial similarity. If the facial similarity is greater than or equal to the facial similarity threshold, it means that the monitored person is the target person. If the facial similarity is less than the facial similarity threshold, it is determined that the monitored person is not the target person. Exemplarily, if there are at least two monitored persons with a large similarity to the target person, and it is impossible to accurately determine which one is the target person, the target person can be further screened and determined according to the body similarity between the monitored person and the target person. For example, if the facial similarity between the monitored person B and the monitored person C and the target person A is greater than the facial similarity threshold, the body similarity between the monitored person B and the monitored person C and the target person A is determined. For example, if the body similarity between monitored person B and target person A is 89%, and the body similarity between monitored person C and target person A is 82%, if the preset body similarity threshold is 85%, since the body similarity between monitored person B and target person A is greater than the preset body similarity threshold, it can be determined that monitored person B is the target person. If the body similarity between monitored person B and target person A is 89%, and the body similarity between monitored person C and target person A is 86%, both of which are greater than the preset body similarity threshold, such as 85%, since the body similarity between monitored person B and target person A is greater, it can be determined that monitored person B is the target person.

[0081] In an embodiment of the present application, before determining the target person from at least two monitored persons based on the body similarity between the monitored person and the target person, the method also includes: acquiring physical feature data of the target person; determining the physical feature data of the monitored person based on a monitoring image of the monitored person; and determining the body similarity between the monitored person and the target person based on the physical feature data of the target person and the physical feature data of the monitored person.

[0082] Exemplarily, as shown in Table 4, for each monitored person appearing in the monitoring screen, the facial similarity and body similarity with the target person are calculated one by one. The similarity calculation can be completed by a similarity calculation algorithm, which will not be described in detail here.

[0083] Table 4

[0084] Name gender height Body shape color Person under surveillance Facial similarity Body similarity Xiao Ming male B M yellow Xiao Ming (myself) 95％ 95％ Xiao Ming male B M yellow Xiaowen 84％ 86％ Xiao Ming male B M yellow Xiao Cheng 82％ 75％ Xiao Ming …… …… …… …… …… …… ……

[0085] In an embodiment of the present application, if there is a monitored person who is similar to the target person but whose facial similarity is less than the facial similarity threshold, the target person is determined from at least two monitored persons based on the body similarity between the monitored person and the target person. For example, the facial similarities between monitored person B and monitored person C and target person A are 89% and 85% respectively, both of which are less than the facial similarity threshold of 90%, and the target person is further determined based on the body similarity. If the body similarity between monitored person B and target person A is 89%, and the body similarity between monitored person C and target person A is 82%, if the preset body similarity threshold is 85%, since the body similarity between monitored person B and target person A is greater than the preset body similarity threshold, monitored person B can be determined as the target person. If the body similarity between monitored person B and target person A is 89%, and the body similarity between monitored person C and target person A is 86%, both of which are greater than a preset body similarity threshold, such as 85%, the monitored person B can be determined to be the target person because the body similarity between monitored person B and target person A is large. If there is no monitored person similar to the target person, and the facial similarity between the monitored person and the target person in the monitoring screen is less than 50%, continue to capture more monitoring images for screening and identification to determine the target person.

[0086] The beneficial effect of the above scheme is that when there are at least two monitored persons and the facial similarity between the target person is greater than the facial similarity threshold, the target person is determined from at least two monitored persons based on the body similarity between the monitored person and the target person, thereby more comprehensively screening and determining the target person based on the body similarity to improve the recognition rate and recognition accuracy of the target person.

[0087] Figure 4 This is a schematic diagram of the structure of a threshold determination device provided by an embodiment of the present invention. The device can be applied to the case where a similarity threshold is determined to determine a target person according to the similarity. Typically, the method can be applied to the case where a similarity threshold associated with each target person is adaptively determined for different target persons, so as to screen the target person from the monitored persons according to the similarity threshold. The device can be implemented by software and / or hardware, and the device can be integrated in a threshold determination device. See Figure 4 , the device specifically comprises:

[0088] A candidate similarity determination module 410 is used to determine at least one candidate similarity between a standard image of at least one candidate person in the information database and a standard image of a target person;

[0089] A target similarity selection module 420, configured to select at least one target similarity from the at least one candidate similarity;

[0090] The similarity threshold determination module 430 is used to determine the similarity threshold associated with the target person according to the at least one target similarity.

[0091] In the embodiment of the present application, the target similarity selection module 420 includes:

[0092] a maximum candidate similarity determination unit, configured to determine a maximum candidate similarity among at least one candidate similarity;

[0093] A first best similarity selection unit, configured to select a preset number of best similarities as target similarities if the number of best similarities whose difference with the maximum candidate similarity is less than a preset similarity difference among at least one candidate similarity is greater than or equal to a preset number;

[0094] The second best similarity selection unit is configured to take all the best similarities as target similarities if the number of best similarities whose difference with the maximum candidate similarity is less than a preset similarity difference among at least one candidate similarity is less than a preset number.

[0095] In the embodiment of the present application, the first best similarity selection unit includes:

[0096] A sorting subunit, used for sorting the best similarities in descending order;

[0097] The sorting and selecting subunit is used to determine a preset number of best similarities that are sorted first as target similarities.

[0098] In the embodiment of the present application, the similarity threshold determination module 430 is specifically used to:

[0099] The similarity threshold associated with the target person is determined based on the target similarity and the following formula:

[0100] K=P1-[P1-(P2+P3+……+Pn) / n]*V;

[0101] Wherein, K is the similarity threshold, P1 is the maximum target similarity, P2-Pn is the target similarity less than the maximum target similarity, n is the number of target similarities, and V is the optimal offset coefficient.

[0102] The threshold determination device provided in the embodiments of the present application can execute the threshold determination method provided in any embodiment of the present application, and has the corresponding functional modules and beneficial effects of the execution method.

[0103] Figure 5The schematic diagram of the structure of a target person identification device provided by an embodiment of the present invention. The device can be used to determine the situation of identifying a target person from a monitoring scene. Typically, the method can be used to adaptively determine the similarity threshold associated with each target person for different target persons, so as to screen the target person from the monitored persons according to the similarity threshold. The device can be implemented by software and / or hardware, and the device can be integrated in a similarity target person identification device. See Figure 5 , the device specifically comprises:

[0104] A real-time similarity determination module 510 is used to determine the real-time similarity between the monitoring image of the monitored person and the standard image of the target person;

[0105] A target person determination module 520, configured to determine that the monitored person is a target person if the real-time similarity is greater than the similarity threshold;

[0106] The similarity threshold is determined according to the threshold determination method described in any of the above embodiments.

[0107] In the embodiment of the present application, the similarity threshold is a facial similarity threshold;

[0108] Accordingly, the device further comprises:

[0109] A facial similarity determination module, used to determine the facial similarity between the monitored person and the target person based on the monitored image of the monitored person and the standard image of the target person;

[0110] The facial similarity comparison module is used to determine the target person from at least two monitored persons according to the body similarity between the monitored person and the target person if there are at least two monitored persons whose facial similarity with the target person is greater than the facial similarity threshold.

[0111] In the embodiment of the present application, the device further includes:

[0112] A first feature data acquisition module, used to acquire the physical feature data of the target person;

[0113] A second feature data acquisition module, used to determine the physical feature data of the monitored person according to the monitoring image of the monitored person;

[0114] The body similarity determination module is used to determine the body similarity between the monitored person and the target person based on the body feature data of the target person and the body feature data of the monitored person.

[0115] The target person identification device provided in the embodiments of the present application can execute the target person identification method provided in any embodiment of the present application, and has the corresponding functional modules and beneficial effects of the execution method.

[0116] Figure 6 A schematic diagram of the structure of a threshold determination device provided in one embodiment of the present invention. Figure 6 A block diagram of an exemplary threshold determination device 612 suitable for implementing embodiments of the present application is shown. Figure 6 The threshold determination device 612 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0117] like Figure 6 As shown, the threshold determination device 612 may include: one or more processors 616; a memory 628, which is used to store one or more programs. When the one or more programs are executed by the one or more processors 616, the one or more processors 616 implement the threshold determination method provided in the embodiment of the present application, including:

[0118] determining at least one candidate similarity between a standard image of at least one candidate person in the information database and a standard image of a target person;

[0119] Selecting at least one target similarity from the at least one candidate similarity;

[0120] A similarity threshold associated with the target person is determined based on the at least one target similarity.

[0121] Components of the threshold determination device 612 may include, but are not limited to, one or more processors or a processor 616 , a memory 628 , and a bus 618 that connects the various device components including the memory 628 and the processor 616 .

[0122] Bus 618 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0123] The threshold determination device 612 typically includes a variety of computer device readable storage media. These storage media can be any available storage media that can be accessed by the threshold determination device 612, including volatile and non-volatile storage media, removable and non-removable storage media.

[0124] The memory 628 may include computer device readable storage media in the form of volatile memory, such as random access memory (RAM) 630 and / or cache memory 632. The threshold determination device 612 may further include other removable / non-removable, volatile / non-volatile computer device storage media. By way of example only, the storage system 634 may be used to read and write non-removable, non-volatile magnetic storage media ( Figure 6 not shown, usually called a "hard drive"). Although Figure 6 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical storage medium) may be provided. In these cases, each drive may be connected to bus 618 via one or more data storage medium interfaces. Memory 628 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of various embodiments of the present invention.

[0125] A program / utility 640 having a set (at least one) of program modules 642 may be stored, for example, in the memory 628, such program modules 642 including, but not limited to, operating devices, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 642 generally perform the functions and / or methods of the embodiments described herein.

[0126] The threshold determination device 612 may also communicate with one or more external devices 614 (e.g., a keyboard, a pointing device, a display 624, etc.), may also communicate with one or more devices that enable a user to interact with the threshold determination device 612, and / or may communicate with any device that enables the threshold determination device 612 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may be performed via an input / output (I / O) interface 622. Furthermore, the threshold determination device 612 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 620. Figure 6 As shown, the network adapter 620 communicates with other modules of the threshold determination device 612 via the bus 618. It should be understood that although Figure 6 Not shown, other hardware and / or software modules may be used in conjunction with the threshold determination device 612, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID devices, tape drives, and data backup storage devices.

[0127] The processor 616 executes various functional applications and data processing by running at least one of the other programs among the multiple programs stored in the memory 628, such as implementing a threshold determination method provided in an embodiment of the present application.

[0128] The present application also provides a schematic diagram of the structure of a target person identification device, whose structure and function are the same as the threshold determination device in the above embodiment, see Figure 6 , may include: one or more processors 616; a memory 628, for storing one or more programs, when the one or more programs are executed by the one or more processors 616, the one or more processors 616 implement the target person identification method provided in the embodiment of the present application, including:

[0129] Determining the real-time similarity between the monitored image of the monitored person and the standard image of the target person;

[0130] If the real-time similarity is greater than the similarity threshold, determining that the monitored person is a target person;

[0131] The similarity threshold is determined according to the threshold determination method described in any one of the above embodiments.

[0132] The processor 616 executes various functional applications and data processing by running at least one of the other programs among the multiple programs stored in the memory 628, such as implementing a target person identification method provided in an embodiment of the present application.

[0133] An embodiment of the present invention provides a storage medium containing computer executable instructions, wherein the computer executable instructions are used to perform a threshold determination method when executed by a computer processor, including:

[0134] determining at least one candidate similarity between a standard image of at least one candidate person in the information database and a standard image of a target person;

[0135] Selecting at least one target similarity from the at least one candidate similarity;

[0136] A similarity threshold associated with the target person is determined based on the at least one target similarity.

[0137] Or perform target person identification methods, including:

[0138] Determining the real-time similarity between the monitored image of the monitored person and the standard image of the target person;

[0139] If the real-time similarity is greater than the similarity threshold, determining that the monitored person is a target person;

[0140] The similarity threshold is determined according to the threshold determination method described in any one of the above embodiments.

[0141] The computer storage medium of the embodiment of the present application can adopt any combination of one or more computer-readable storage media. The computer-readable storage medium can be a computer-readable signal storage medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, - but not limited to - electrical, magnetic, optical, electromagnetic, infrared, or semiconductor equipment, devices or devices, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In an embodiment of the present application, a computer-readable storage medium can be any tangible storage medium containing or storing a program, which can be used by an instruction execution device, device or device or used in combination with it.

[0142] A computer-readable signal storage medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal storage medium may also be any computer-readable storage medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution device, apparatus, or device.

[0143] The program code contained on the computer-readable storage medium may be transmitted using any appropriate storage medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0144] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or device. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0145] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A threshold determination method, It is characterized in that The method comprises: determining at least one candidate similarity between a standard image of at least one candidate person in the information database and a standard image of a target person; Selecting at least one target similarity from the at least one candidate similarity; According to the at least one target similarity, a similarity threshold associated with the target person is determined; wherein the similarity threshold is calculated by fitting a calculation formula of the similarity threshold based on the relationship between the similarity threshold determined by the target person recognition accuracy and the target similarity.

2. The method according to claim 1, It is characterized in that Selecting at least one target similarity from the at least one candidate similarity includes: determining a maximum candidate similarity among at least one candidate similarity; If, among at least one candidate similarity, the number of best similarities whose difference with the maximum candidate similarity is less than a preset similarity difference is greater than or equal to a preset number, then a preset number of best similarities are selected as target similarities; If, among at least one candidate similarity, the number of best similarities whose difference with the maximum candidate similarity is less than a preset similarity difference is less than a preset number, all the best similarities are used as target similarities.

3. The method according to claim 2, It is characterized in that Select a preset number of best similarities as the target similarity, including: sorting the best similarities in descending order; A preset number of best similarities that are ranked first are determined as target similarities.

4. The method according to claim 1, It is characterized in that Determining a similarity threshold associated with the target person according to the at least one target similarity includes: The similarity threshold associated with the target person is determined based on the target similarity and the following formula: K=P1-[P1-(P2+P3+……+Pn) / n]*V; Wherein, K is the similarity threshold, P1 is the maximum target similarity, P2-Pn is the target similarity less than the maximum target similarity, n is the number of target similarities, and V is the optimal offset coefficient.

5. A method for identifying a target person, It is characterized in that The method comprises: Determining the real-time similarity between the monitored image of the monitored person and the standard image of the target person; If the real-time similarity is greater than the similarity threshold, determining that the monitored person is a target person; The similarity threshold is determined according to the threshold determination method according to any one of claims 1-4.

6. The method according to claim 5, It is characterized in that The similarity threshold is a facial similarity threshold; Accordingly, after determining the similarity threshold associated with the target person according to the at least one target similarity, the method further includes: Determining the facial similarity between the monitored person and the target person based on the monitored image of the monitored person and the standard image of the target person; If there are at least two monitored persons whose facial similarity with the target person is greater than the facial similarity threshold, the target person is determined from the at least two monitored persons according to the body similarity between the monitored person and the target person.

7. The method according to claim 6, It is characterized in that Before determining the target person from at least two monitored persons based on the body similarity between the monitored person and the target person, the method further includes: Acquiring physical characteristic data of the target person; Determining physical feature data of the monitored person according to the monitored image of the monitored person; The body similarity between the monitored person and the target person is determined based on the body feature data of the target person and the body feature data of the monitored person.

8. A threshold determination device, It is characterized in that The device comprises: a candidate similarity determination module, configured to determine at least one candidate similarity between a standard image of at least one candidate person in the information database and a standard image of a target person; A target similarity selection module, used to select at least one target similarity from the at least one candidate similarity; A similarity threshold determination module is used to determine a similarity threshold associated with the target person based on the at least one target similarity; wherein the similarity threshold is calculated by fitting the calculation formula of the similarity threshold based on the relationship between the similarity threshold determined by the target person recognition accuracy and the target similarity.

9. A target person identification device, It is characterized in that The device comprises: A real-time similarity determination module, used to determine the real-time similarity between the monitored image of the monitored person and the standard image of the target person; A target person determination module, configured to determine that the monitored person is a target person if the real-time similarity is greater than the similarity threshold; The similarity threshold is determined according to the threshold determination method according to any one of claims 1-4.

10. An electronic device, It is characterized in that The electronic device comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the threshold determination method as described in any one of claims 1 to 4, or implement the target person identification method as described in any one of claims 5 to 7.

11. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the program is executed by a processor, the threshold determination method according to any one of claims 1 to 4 is implemented, or the target person identification method according to any one of claims 5 to 7 is implemented.

Images