Personnel tracking and marking method, device, electronic device and storage medium

Abstract

The present invention relates to the field of public health technology, and provides a personnel tracking and marking method, device, electronic device and storage medium, the method comprising: obtaining captured image information of a traveler, initializing the captured image information of the traveler to obtain target traveler image information; extracting target attribute feature value of the target traveler image information, calculating similarity between the target attribute feature value and the first attribute feature value in a first database, and determining suspicious marked persons from the traveler according to the similarity; tracking the suspicious marked persons based on a preset tracking model, determining the target location area, matching the target attribute feature value of the suspicious marked person with the second attribute feature value in a second database of the target location area; marking the suspicious marked person according to the matching result, obtaining the target marked person and pushing it. The present invention can improve the efficiency of identity identification of personnel.

Description

Technical Field

[0001] The present invention relates to the field of public health, and in particular to a personnel tracking and marking method, device, electronic equipment and storage medium. Background Art

[0002] As the country attaches great importance to public health services, it is required that public health services can achieve comprehensive monitoring, detection, reporting and management of public areas with large traffic flow, so that dynamic information of people can be queried. However, it is difficult to obtain the activity tracks of some special people, especially those who cannot obtain clear facial images, such as those who appear in shopping malls, stations and other areas with large traffic flow. It can be seen that in the existing technology, there is a problem of low efficiency in identifying the identity of people. Summary of the invention

[0003] The embodiment of the present invention provides a personnel tracking and marking method, which can improve the efficiency of personnel identity identification.

[0004] In a first aspect, an embodiment of the present invention provides a personnel tracking and marking method, the method comprising the following steps:

[0005] Acquire captured image information of travelers, and initialize the captured image information of the travelers to obtain target traveler image information of target travelers;

[0006] Extracting target attribute feature values ​​from the target traveler image information, calculating similarity between the target attribute feature values ​​from the target traveler image information and first attribute feature values ​​in a first database, and determining suspicious marked persons from the traveler according to the similarity;

[0007] Tracking the suspicious marked person based on a preset tracking model to determine a target location area, and matching a target attribute feature value of the suspicious marked person with a second attribute feature value in a second database of the target location area;

[0008] The suspicious marked person is marked according to the matching result, and the target marked person is obtained and pushed.

[0009] In a second aspect, an embodiment of the present invention further provides a personnel tracking marking device, comprising:

[0010] An initialization module, used for acquiring captured image information of travelers, and initializing the captured image information of the travelers to obtain target traveler image information of target travelers;

[0011] a calculation module, configured to extract a target attribute feature value from the target traveler image information, perform similarity calculation between the target attribute feature value from the target traveler image information and a first attribute feature value in a first database, and determine a suspicious marked person from the traveler according to the similarity;

[0012] A tracking module, used to track the suspicious marked person based on a preset tracking model, determine a target location area, and match a target attribute feature value of the suspicious marked person with a second attribute feature value in a second database of the target location area;

[0013] The marking module is used to mark the suspicious marked person according to the matching result, obtain the target marked person and push it.

[0014] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in the personnel tracking and marking method provided in the embodiment when executing the computer program.

[0015] In a fourth aspect, a computer-readable storage medium is provided, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps in the personnel tracking and marking method provided in the embodiment are implemented.

[0016] In the embodiment of the present invention, the present invention initializes the captured image information of the travelers, performs preliminary screening to obtain the target travelers, and calculates the similarity between the target attribute feature value of the target travelers and the first attribute feature value in the first database. Since the target travelers and the people in the first database are both travelers, it can be determined whether the same traveler has passed through different stations based on the similarity between the first attribute feature value and the target attribute feature value. In the case where a clear face cannot be obtained, the suspicious marked person can be quickly determined by comparing the attribute feature values; and the suspicious marked person is tracked by a preset tracking model to obtain the target location area, and the second attribute feature value in the second database of the target location area is matched with the target attribute feature value of the suspicious marked person, so that non-suspicious marked persons among the suspicious marked persons can be excluded. By screening and determining the target marked person layer by layer, the accuracy of tracking the suspicious marked person can be improved, and at the same time, blind tracking of unnecessary suspicious marked persons and waste of data processing resources can be avoided, which is conducive to improving the efficiency of identifying the target marked person. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 is a flow chart of a personnel tracking and marking method provided by an embodiment of the present invention;

[0019] Figure 2 is a flow chart of another personnel tracking and marking method provided by an embodiment of the present invention;

[0020] Figure 3 is a flow chart of another personnel tracking and marking method provided by an embodiment of the present invention;

[0021] Figure 4 It is a structural schematic diagram of a personnel tracking marking device provided by an embodiment of the present invention;

[0022] Figure 5 is a schematic structural diagram of another personnel tracking marking device provided by an embodiment of the present invention;

[0023] Figure 6 is a schematic structural diagram of another personnel tracking marking device provided by an embodiment of the present invention;

[0024] Figure 7 is a schematic structural diagram of another personnel tracking marking device provided by an embodiment of the present invention;

[0025] Figure 8 is a schematic structural diagram of another personnel tracking marking device provided by an embodiment of the present invention;

[0026] Fig. 9 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] The terms "including" and "having" and any variations thereof in the specification, claims and drawings of the present application are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification, claims or drawings of the present application are used to distinguish different objects rather than to describe a specific order. Reference to "embodiment" in this article means that the specific features, structures or characteristics described in conjunction with the embodiment may be included in at least one embodiment of the present application. The appearance of this phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0029] like Figure 1 As shown, Figure 1 1 is a flow chart of a personnel tracking and marking method provided by an embodiment of the present invention, and the personnel tracking and marking method comprises the following steps:

[0030] S101. Acquire captured image information of travelers, and initialize the captured image information of travelers to obtain target traveler image information of target travelers.

[0031] In this embodiment, the scenarios in which the personnel tracking and marking method is applied include but are not limited to shopping malls, stations, road traffic stations, etc., and the electronic device on which the above-mentioned personnel tracking and marking method is running can obtain the captured image information of the travelers through a wired connection method or a wireless connection method. The above-mentioned wireless connection method may include but is not limited to 3G / 4G connection, WiFi (Wireless-Fidelity) connection, Bluetooth connection, WiMAX (Worldwide Interoperability for Microwave Access) connection, Zigbee (low-power LAN protocol, also known as Zigbee protocol) connection, UWB (ultrawideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0032] The above-mentioned travelers may include any mobile personnel, and the above-mentioned captured image information may refer to image information corresponding to the travelers, and the captured image information may include human bodies, faces, clothing, and accessories, etc. The captured image information of the above-mentioned travelers can be captured by an image acquisition device, and the captured image information is sent to the background algorithm for data processing, wherein the image acquisition device may be an electronic device capable of image capture, recognition, storage, etc., including but not limited to a camera, a scanner, a camera, a video camera, and other devices with cameras (mobile phones, tablet computers, etc.). In an embodiment of the present invention, the above-mentioned image acquisition device may be a camera, which may refer to cameras in various occasions, and different numbers of cameras may be set in different occasions, so that more complete image information in different occasions can be obtained. Of course, this will also make the stored image data more and require more storage space.

[0033] The above-mentioned initialization processing of the captured image information of travelers can be to classify travelers with the same characteristics and classify and manage travelers with different characteristics for easy search. For example, there are masks, sunglasses, hats and other obstructions in the captured image information of travelers. After the initialization processing, travelers with certain preset characteristics can be tracked as target travelers, and the target traveler image information of the target travelers can be obtained.

[0034] S102, extracting target attribute feature values ​​in the target traveler image information, calculating similarity between the target attribute feature values ​​in the target traveler image information and first attribute feature values ​​in the first database, and determining suspicious marked persons from the travelers based on the similarity.

[0035] Among them, the above-mentioned target attribute feature values ​​may include feature values ​​used to identify people, especially when facial image information or clear facial image information cannot be collected, such as: clothes, pants, masks, wearing, shoes and the color and style of clothes, pants, masks, wearing, shoes and other information.

[0036] The above-mentioned first database may refer to a database of ordinary travelers. Both the target travelers and the ordinary travelers in the database of ordinary travelers are travelers whose data needs to be collected, and the corresponding data are collected within a preset time period. The database of ordinary travelers may be a portrait library obtained by initializing and summarizing travelers captured at ordinary stations in a preset time period. The above-mentioned first attribute feature value may be a target attribute feature value stored in the above-mentioned database of ordinary travelers, that is, the attribute feature value of the travelers obtained at ordinary stations. And the first attribute feature value may be classified according to categories and stored in different data tables in the database of ordinary travelers, for example: clothes are stored in data table No. 1, hats are placed in data table No. 2, etc.

[0037] The similarity calculation can be expressed as the degree of similarity between the target attribute feature value and the first attribute feature value, for example, the similarity between A and B's clothes can be calculated, and the calculation can be performed from the styles (long, short, short sleeves, long sleeves, collars, decorations, etc.), colors, patterns, etc. The above-mentioned travelers include target travelers and ordinary travelers, so the captured image information of travelers includes the image information of target travelers and the image information of ordinary travelers. When capturing images, the capture time period of ordinary travelers and target travelers is the same. The fewer the change parameters for calculating the similarity, the less the amount of calculated data. According to the calculated similarity, it can be determined whether the travelers collected at different stations are the same person. If yes, it is determined that the person can be marked. Of course, facial feature values ​​can also be stored in the database of ordinary travelers. In addition to calculating the similarity between the target attribute feature value and the first attribute feature value, if the target attribute feature value also obtains a face image (including front face, side face, etc.), the face feature value can also be matched with data.

[0038] After the similarity calculation between the target attribute characteristic value and the first attribute characteristic value in the target traveler's image information is completed, the above-mentioned suspicious marked persons can be screened out according to the similarity calculation result. The suspicious marked persons may include the target traveler and ordinary travelers whose images match the target traveler's images. As a feasible implementation method, the information of the screened suspicious marked persons and the marked persons can be summarized to generate a suspicious marked person file. The suspicious marked person file may include information such as the capture time and capture location of the suspicious marked person's captured image, and the attribute characteristic value of the suspicious marked person. Generating a suspicious marked person file is conducive to unified management.

[0039] S103: Track the suspicious marked person based on a preset tracking model, determine the target location area, and match the target attribute feature value of the suspicious marked person with the second attribute feature value in the second database of the target location area.

[0040] Among them, the preset tracking model can be the use of cross-border tracking technology. The cross-camera tracking (Person Re-Identification, referred to as ReID) technology mainly solves the identification and retrieval of people across cameras and scenes. This technology can be used as an important supplement to face recognition technology. It can continuously track the heads of people who cannot obtain clear face images across borders and enhance the spatiotemporal connectivity of data. In addition, the cross-camera tracking (ReID) technology can recognize pedestrians based on their clothing, body shape, hairstyle and other information. Combined with face recognition technology, it can be applied to more new application scenarios.

[0041] The above-mentioned target location area may refer to every location passed by the suspicious marked person who is tracked across borders based on cross-border tracking (ReID) technology, such as the starting point and terminal of the vehicle. The above-mentioned second database may refer to an image database of the area where the target location area is located, such as the city-level portrait library corresponding to site A. The above-mentioned second attribute feature values ​​may also include information such as clothes, pants, masks, wearing, shoes, and the color and style of clothes, pants, masks, wearing, shoes, etc. The second attribute feature values ​​can be classified and stored in the above-mentioned second database.

[0042] Specifically, by tracking the suspicious marked persons through cross-border tracking (ReID), the target location area corresponding to the suspicious marked persons can be determined, and then the second attribute feature value in the second database corresponding to the target location area can be extracted, and the target attribute feature value of the suspicious marked persons is matched with the second attribute feature value. Based on the matching of the target attribute feature value with the second attribute feature value, it can be determined whether an impossible event occurs between the two. When an impossible event occurs, the marked person can be excluded, which is conducive to quickly determining the identity of each traveler among the suspicious marked persons.

[0043] S104. Mark the suspicious marked persons according to the matching results, obtain the target marked persons and push them.

[0044] Among them, the matching result may include the similarity value between the target attribute feature value of the suspicious marked person and the above-mentioned second attribute feature value, and the suspicious marked person can be marked according to the size of the similarity value. The target marked person obtained after marking can be regarded as an object that needs to be focused on, and the target marked person and the image information corresponding to the target marked person are pushed to the statistical terminal, and the marked traveler can be marked and displayed on the statistical terminal. The above-mentioned statistical terminal may include but is not limited to electronic devices with display functions such as mobile phones, desktop computers, notebooks and handheld computers.

[0045] In the embodiment of the present invention, the captured image information of the travelers is initialized, and the target travelers are initially screened to obtain the target travelers, and the target attribute feature value of the target travelers is similarly calculated with the first attribute feature value in the first database. Since the target travelers and the people in the first database are both travelers, it can be determined whether the same traveler has passed through different stations based on the similarity between the first attribute feature value and the target attribute feature value. In the case where a clear face cannot be obtained, the suspicious marked person can be quickly determined by comparing the attribute feature values; and the suspicious marked person is tracked by a preset tracking model to obtain the target location area, and the second attribute feature value in the second database of the target location area is matched with the target attribute feature value of the suspicious marked person, so that non-suspicious marked persons among the suspicious marked persons can be excluded. By screening and determining the target marked person layer by layer, the accuracy of tracking the suspicious marked person can be improved, and at the same time, blind tracking of unnecessary suspicious marked persons and waste of data processing resources can be avoided, which is conducive to improving the efficiency of identifying the target marked person.

[0046] like Figure 2 As shown, Figure 2 Another method flow chart provided by an embodiment of the present invention specifically includes the following steps:

[0047] S201. Acquire travelers captured in a preset time period and at a preset location.

[0048] The preset time period can be selected according to the situation, and the period of the preset time period can be 1 day, 12 hours, etc. The limitation of the preset time period can keep the appearance of the travelers consistent as much as possible. The preset location can refer to a key site, such as a station, a checkpoint, etc. The travelers can be captured by an image acquisition device at a preset time period and a preset location, for example, from 10 am to 12 pm, the image of the travelers appearing at Station A is captured by a camera.

[0049] S202: extracting captured image information of travelers, clustering the captured image information, and marking captured image information that fails in clustering as target traveler image information.

[0050] Among them, the above clustering can mean archiving of portraits, screening the captured image information that can be clearly identified from the captured image information and can be used to determine the identity of travelers, and aggregating them as a regular image type, for example: travelers whose facial images can be clearly obtained. That is, the captured image information that is successfully clustered and the captured image information that fails to cluster are stored and processed differently, which can avoid the subsequent secondary identity confirmation of travelers with clear identities. The above clustering failure can mean that the identity cannot be determined from the captured image information obtained. The captured image information that failed to cluster can be marked as the image information of the target traveler, for example: the captured image of the traveler contains obstructions such as masks.

[0051] S203, extracting target attribute feature values ​​in the target traveler image information, calculating similarity between the target attribute feature values ​​in the target traveler image information and the first attribute feature values ​​in the first database, and determining suspicious marked persons from the travelers according to the similarity.

[0052] S204: Track the suspicious marked person based on a preset tracking model, determine the target location area, and match the target attribute feature value of the suspicious marked person with the second attribute feature value in the second database of the target location area.

[0053] S205. Mark the suspicious marked persons according to the matching results, obtain the target marked persons and push them.

[0054] Optionally, the above step S203 includes:

[0055] A first attribute characteristic value in a first database is obtained, where the first attribute characteristic value in the first database is a characteristic value obtained in a preset time period and a non-preset location.

[0056] When extracting the target attribute feature value of the target traveler image information, the first attribute feature value in the first database can also be called. The above target attribute feature value and the first attribute feature value are extracted in a preset time period, for example, from 10:00 to 12:00 on the same day. The above non-preset location can mean that it is not a focus point, but belongs to an ordinary site, such as a road detection camera, etc.

[0057] The target attribute feature value in the target traveler image information is calculated similarly to the first attribute feature value to obtain a similarity value.

[0058] Among them, because the captured image information of the travelers (including the images of the target travelers and the images in the database of ordinary travelers) is captured at a time within the above-mentioned preset time period, the attribute feature values ​​of each target traveler, such as the body, clothing, and mask style, are likely to remain unchanged. If the target attribute feature values ​​include some non-frontal feature values, they can also serve as facial feature values ​​for data matching with the facial feature values ​​in the first database. The above-mentioned similarity calculation can be a 1:N calculation, and after the calculation, the similarity value between each type of attribute feature value can be obtained, for example: the similarity value between clothes and clothes, the similarity value between masks and masks, etc.

[0059] Determine whether the similarity value reaches a preset similarity threshold.

[0060] A similarity threshold may be preset, and all calculated similarity values ​​are compared with the preset similarity threshold.

[0061] If the similarity value reaches a preset similarity threshold, the travelers whose similarity values ​​reach the preset similarity threshold are extracted as suspicious marked persons.

[0062] Among them, if there is a similarity value reaching the preset similarity threshold, it can be said that the target traveler and the ordinary traveler whose similarity reaches the similarity threshold are likely to become suspicious persons, and the target traveler and the ordinary traveler whose similarity value does not reach the preset similarity threshold can be screened out. Therefore, the traveler whose similarity value reaches the above similarity threshold can be further tracked as a suspicious marked person to confirm the identity.

[0063] In the embodiment of the present invention, the captured image information of the travelers is clustered and archived, and the captured image information that failed to be clustered is marked as the target traveler image information. At the same time, the secondary identity confirmation of the travelers who were successfully clustered (travelers whose identities can be confirmed) can be avoided. The target attribute feature value in the target traveler image information is extracted and the similarity is calculated with the first attribute feature value. In the case where a clear face cannot be obtained, the suspicious marked person can be quickly determined by comparing the attribute feature value, and the suspicious marked person is continued to be tracked and matched with the second attribute feature value in the target location area, so that non-suspicious marked persons can be excluded from the suspicious marked persons. By screening and determining the target marked person layer by layer, the accuracy of tracking the suspicious marked person can be improved, and at the same time, the blind tracking of unnecessary suspicious marked persons and the waste of data processing resources can be avoided, which is conducive to improving the efficiency of identifying the target marked person.

[0064] like Figure 3 As shown, Figure 3Another method flow chart provided by an embodiment of the present invention includes the following steps:

[0065] S301, acquiring captured image information of travelers, and initializing the captured image information of travelers to obtain target traveler image information of target travelers.

[0066] S302, extracting target attribute feature values ​​in the target traveler image information, calculating similarity between the target attribute feature values ​​in the target traveler image information and first attribute feature values ​​in the first database, and determining suspicious marked persons from the travelers based on the similarity.

[0067] S303: Obtain the capture time, capture location and train number information of the suspicious marked person.

[0068] Among them, the above-mentioned capture time corresponds to the capture location, and the corresponding capture time at each capture location may be different, for example: the capture time at location C is 8:00 a.m., and the corresponding capture time at location D is 12:00 noon. Of course, the above-mentioned capture time belongs to the above-mentioned preset time period, and the above-mentioned capture location may include key focus sites and ordinary sites. The above-mentioned train number information may be the information of the vehicle taken by the travelers, and may include the departure time, suspension time, arrival time, license plate number, vehicle model, passing locations and other information of the train number.

[0069] S304: Screen suspicious marked persons according to the capture time, capture location and train number information.

[0070] Among them, after the suspicious marked persons are extracted, they can be summarized to form a suspicious marked person file, in which all the captured image information of the suspicious marked persons can be stored. The above screening of suspicious marked persons can mean the elimination of information related to unreasonable imagination, for example: the traveler corresponding to the captured image in the suspicious marked person file passes through stations A, B, C, D, and E, but according to the capture time and capture location, it can be judged that the traveler cannot appear at stations B and C at the same time, so the corresponding train information of the day is extracted. If it can be concluded from the train information that there are trains that pass through A, C, D, and E at the same time and there is no conflict in the passing time, the captured image of the traveler at station B in the above suspicious marked person file can be eliminated to achieve the optimization of the suspicious marked person file, reduce the amount of data for subsequent tracking, and improve the efficiency of identity discrimination.

[0071] S305: Track the suspicious marked person based on a preset tracking model, determine the target location area, and match the target attribute feature value of the suspicious marked person with the second attribute feature value in the second database of the target location area.

[0072] S306. Mark the suspicious marked persons according to the matching results, obtain the target marked persons and push them.

[0073] Optionally, the above step S304 includes:

[0074] Get all the captured locations of suspicious marked persons.

[0075] Among them, images of all places where the suspicious marked person has passed can be captured to obtain all the captured locations of the suspicious marked person.

[0076] According to the train information, the capture locations that the suspicious marked person actually needs to pass through are determined, and the capture locations that the suspicious marked person actually needs to pass through correspond to the capture time.

[0077] Among them, when a suspicious marked person takes a bus, the riding information of the suspicious marked person can be obtained from the ticket purchasing system, and the riding information can include the above-mentioned train number information. If you want to obtain information in the ticket purchasing system, you can do it under the condition of obtaining permission, by sending an information query request to the ticket purchasing system, and the ticket purchasing system will respond to the information query request and feedback the corresponding train number information. After obtaining the train number information, you can determine the places that the vehicle actually needs to pass (image capture can be performed at each place).

[0078] Determine whether the capture locations that the suspicious marked person actually needs to pass through include all the captured capture locations of the suspicious marked person.

[0079] After obtaining the actual capture location that needs to be passed, it can be compared with all the capture locations of the currently captured suspicious marked persons to find out whether there are the same or different locations.

[0080] If the capture locations that the suspicious marked person actually needs to pass do not include all the captured locations of the suspicious marked person collected, the captured image information corresponding to the capture locations that are not actually needed to be passed by the suspicious marked person is deleted.

[0081] Among them, if the capture locations that the suspicious marked person actually needs to pass through do not include all the capture locations of the collected suspicious marked persons, for example: the traveler corresponding to the capture image in the suspicious marked person file passes through stations A, B, C, D, and E, but according to the capture time and capture location, it can be judged that the traveler cannot appear at stations B and C at the same time, then the corresponding train number information of the day is extracted. If it can be concluded from the train number information that there are trains that pass through A, C, D, and E at the same time and there is no conflict in the passing time, then the capture image information of the traveler at station B in the above-mentioned suspicious marked person file can be eliminated.

[0082] In the embodiment of the present invention, the captured image information of the suspicious marked person is screened by the capture time, capture location and train number information, so that unreasonable phenomena can be eliminated and non-suspicious marked persons can be excluded. By determining the target marked person through layer-by-layer screening, the accuracy of tracking the suspicious marked person can be improved, and at the same time, it can avoid blindly tracking unnecessary suspicious marked persons and wasting data processing resources, reduce the amount of data for subsequent tracking, and achieve more accurate and rapid tracking to improve the efficiency of identity discrimination.

[0083] like Figure 4 As shown, Figure 4 4 is a schematic diagram of a personnel tracking and marking device provided by an embodiment of the present invention. The personnel tracking and marking device 400 includes:

[0084] Initialization module 401, used to obtain captured image information of travelers, and initialize the captured image information of travelers to obtain target traveler image information of target travelers;

[0085] The calculation module 402 is used to extract the target attribute feature value in the target traveler image information, calculate the similarity between the target attribute feature value in the target traveler image information and the first attribute feature value in the first database, and extract the suspicious marked person according to the similarity;

[0086] Tracking module 403, used to track the suspicious marked person based on a preset tracking model, determine the target location area, and match the target attribute feature value of the suspicious marked person with the second attribute feature value in the second database of the target location area;

[0087] The marking module 404 is used to mark the suspicious marked persons according to the matching results, obtain the target marked persons and push them.

[0088] Optional, such as Figure 5 As shown, Figure 5 4 is a schematic diagram of the structure of another personnel tracking and marking device provided by an embodiment of the present invention. The initialization module 401 includes:

[0089] The first acquisition unit 4011 is used to acquire travelers captured at a preset time period and at a preset location;

[0090] The clustering unit 4012 is used to extract the captured image information of the travelers, cluster the captured image information, and mark the captured image information that fails to be clustered as the target traveler image information.

[0091] Optional, such as Figure 6 As shown, Figure 6 4 is a schematic diagram of the structure of another personnel tracking and marking device provided by an embodiment of the present invention, wherein the calculation module 402 includes:

[0092] The second acquisition unit 4021 is used to acquire a first attribute characteristic value in the first database, where the first attribute characteristic value in the first database is a characteristic value acquired in a preset time period and a non-preset location;

[0093] The calculation unit 4022 is used to calculate the similarity between the target attribute feature value in the target traveler image information and the first attribute feature value to obtain a similarity value;

[0094] The first judging unit 4023 is used to judge whether the similarity value reaches a preset similarity threshold;

[0095] The extraction unit 4024 is used to extract the travelers whose similarity values ​​reach the preset similarity threshold as suspicious marked persons if the similarity value reaches the preset similarity threshold.

[0096] Optional, such as Figure 7 As shown, Figure 7 400 is a schematic diagram of the structure of another personnel tracking and marking device provided by an embodiment of the present invention. The device 400 also includes:

[0097] The acquisition module 405 is used to obtain the capture time, capture location and train number information of the suspicious marked person;

[0098] The screening module 406 is used to screen suspicious marked persons according to the capture time, capture location and train number information.

[0099] Optional, such as Figure 8 As shown, Figure 8 4 is a schematic diagram of the structure of another personnel tracking and marking device provided by an embodiment of the present invention, wherein the screening module 406 includes:

[0100] The third acquisition unit 4061 acquires all captured locations of the suspicious marked persons;

[0101] The determination unit 4062 is used to determine the capture location that the suspicious marked person actually needs to pass through according to the train number information, and the capture location that the suspicious marked person actually needs to pass through corresponds to the capture time;

[0102] The second judgment unit 4063 is used to judge whether the capture locations that the suspicious marked person actually needs to pass through include all the captured capture locations of the suspicious marked person;

[0103] The deleting unit 4064 is used to delete the captured image information corresponding to the captured locations that are not actually passed by the suspicious marked person if the captured locations that the suspicious marked person actually needs to pass do not include all the captured locations of the suspicious marked person collected.

[0104] like Fig. 9 As shown, Fig. 9 It is a structural diagram of an electronic device provided in an embodiment of the present invention. The electronic device 900 includes: a processor 901, a memory 902, a network interface 903, and a computer program stored in the memory 902 and executable on the processor 901. When the processor 901 executes the computer program, the steps in the personnel tracking and marking method provided in the embodiment are implemented.

[0105] Specifically, the processor 901 is configured to perform the following steps:

[0106] Acquire captured image information of travelers, and initialize the captured image information of travelers to obtain target traveler image information of target travelers;

[0107] Extracting target attribute feature values ​​in the target traveler image information, calculating similarity between the target attribute feature values ​​in the target traveler image information and the first attribute feature values ​​in the first database, and determining suspicious marked persons from the travelers according to the similarity;

[0108] Tracking the suspicious marked person based on a preset tracking model, determining the target location area, and matching the target attribute feature value of the suspicious marked person with the second attribute feature value in the second database of the target location area;

[0109] Mark the suspicious marked persons according to the matching results, obtain the target marked persons and push them.

[0110] Optionally, after acquiring the image information, the processor 901 executes the steps of acquiring the captured image information of the travelers and initializing the captured image information of the travelers to obtain the target traveler image information of the target travelers, including:

[0111] Obtain travelers captured in a preset time period and at a preset location;

[0112] The captured image information of the travelers is extracted, and the captured image information is clustered, and the captured image information that fails in clustering is marked as the image information of the target travelers.

[0113] Optionally, after inputting into a preset face re-detection model, the processor 901 performs similarity calculation on the target attribute feature value in the target traveler image information and the first attribute feature value in the first database, and determines the suspicious marked person from the traveler according to the similarity, including:

[0114] Obtain a first attribute characteristic value in a first database, where the first attribute characteristic value in the first database is a characteristic value obtained in a preset time period and a non-preset location;

[0115] Calculate the similarity between the target attribute feature value in the target traveler image information and the first attribute feature value to obtain a similarity value;

[0116] Determine whether the similarity value reaches a preset similarity threshold;

[0117] If the similarity value reaches a preset similarity threshold, the travelers whose similarity values ​​reach the preset similarity threshold are extracted as suspicious marked persons.

[0118] Optionally, after extracting the suspicious marked persons according to the similarity, the processor 901 is further used to execute the acquisition of the capture time, capture location and train number information of the suspicious marked persons;

[0119] Suspicious marked persons are screened based on the capture time, capture location and train number information.

[0120] Optionally, the step of screening suspicious marked persons according to the capture time, capture location and train number information performed by the processor 901 includes:

[0121] Obtain all captured locations of suspicious marked persons;

[0122] According to the train information, the capture location that the suspicious marked person actually needs to pass through is determined, and the capture location that the suspicious marked person actually needs to pass through corresponds to the capture time;

[0123] Determine whether the capture locations that the suspicious marked person actually needs to pass through include all the captured capture locations of the suspicious marked person;

[0124] If the capture locations that the suspicious marked person actually needs to pass do not include all the captured locations of the suspicious marked person collected, the captured image information corresponding to the capture locations that are not actually needed to be passed by the suspicious marked person is deleted.

[0125] The electronic device 900 provided in the embodiment of the present invention can implement various implementation methods in the personnel tracking and marking method embodiment, as well as corresponding beneficial effects, which will not be described again here to avoid repetition.

[0126] It should be noted that the figure only shows 901-903 with components, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the electronic device 900 here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASIC), programmable gate arrays (FPGA), digital signal processors (DSP), embedded devices, etc.

[0127] The electronic device 900 may be a computing device such as a desktop computer, a notebook, a palm computer, etc. The electronic device 900 may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0128] The memory 902 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 902 can be an internal storage unit of the electronic device 900, such as a hard disk or memory of the electronic device 900. In other embodiments, the memory 902 can also be an external storage device of the electronic device 900, such as a plug-in hard disk equipped on the electronic device 900, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Of course, the memory 902 can also include both the internal storage unit of the electronic device 900 and its external storage device. In this embodiment, the memory 902 is generally used to store the operating system and various application software installed on the electronic device 900, such as the program code of the personnel tracking and marking method, etc. In addition, the memory 902 can also be used to temporarily store various types of data that have been output or are to be output.

[0129] In some embodiments, the processor 901 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 901 is generally used to control the overall operation of the electronic device 900. In this embodiment, the processor 901 is used to run the program code stored in the memory 902 or process data, such as the program code for running the personnel tracking tag method.

[0130] The network interface 903 may include a wireless network interface or a wired network interface, and the network interface 903 is generally used to establish a communication connection between the electronic device 900 and other electronic devices.

[0131] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by the processor 901, each process in the personnel tracking and marking method provided in the embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0132] A person skilled in the art can understand that all or part of the processes in the personnel tracking and marking method of the embodiment can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of each method. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0133] The first, second, etc. mentioned in the embodiments of the present invention do not represent the size, but are only for the convenience of description. The above disclosure is only the preferred embodiment of the present invention, and of course it cannot be used to limit the scope of the rights of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A personnel tracking and marking method, characterized in that: The following steps are involved: Acquire captured image information of travelers, and initialize the captured image information of the travelers to obtain target traveler image information of target travelers; Extracting target attribute feature values ​​from the target traveler image information, calculating similarity between the target attribute feature values ​​from the target traveler image information and first attribute feature values ​​in a first database, and determining suspicious marked persons from the traveler according to the similarity; Tracking the suspicious marked person based on a preset tracking model to determine a target location area, and matching a target attribute feature value of the suspicious marked person with a second attribute feature value in a second database of the target location area; Mark the suspicious marked person according to the matching result, obtain the target marked person and push it; The step of acquiring the captured image information of the travelers and initializing the captured image information of the travelers to obtain the target traveler image information of the target travelers includes: Obtain travelers captured in a preset time period and at a preset location; Extracting the captured image information of the travelers, clustering the captured image information, and marking the captured image information that fails to be clustered as target traveler image information; The step of calculating the similarity between the target attribute feature value in the target traveler image information and the first attribute feature value in the first database, and determining the suspicious marked persons from the traveler according to the similarity comprises: Acquire the first attribute characteristic value in the first database, where the first attribute characteristic value in the first database is a characteristic value acquired in the preset time period and not in the preset location; Calculate the similarity between the target attribute feature value in the target traveler image information and the first attribute feature value to obtain a similarity value; Determining whether the similarity value reaches a preset similarity threshold; If the similarity value reaches the preset similarity threshold, the traveler whose similarity value reaches the preset similarity threshold is extracted as the suspicious marked person.

2. The personnel tracking and marking method according to claim 1, characterized in that: After extracting suspicious marked persons according to the similarity, the method further comprises the steps of: Obtain the capture time, capture location and train number information of the suspicious marked person; The suspicious marked persons are screened according to the capture time, the capture location and the train number information.

3. The personnel tracking and marking method according to claim 2, characterized in that: The step of screening the suspicious marked persons according to the capture time, the capture location and the train number information comprises: Obtaining all captured locations of the suspicious marked persons; Determine, based on the train number information, a capture location that the suspicious marked person actually needs to pass through, where the capture location that the suspicious marked person actually needs to pass through corresponds to the capture time; Determine whether the capture locations that the suspicious marked person actually needs to pass through include all the captured capture locations of the suspicious marked person that have been collected; If the capture locations that the suspicious marked person actually needs to pass do not include all the captured capture locations of the suspicious marked person that have been collected, the captured image information corresponding to the capture locations that are not actually needed to be passed by the suspicious marked person will be deleted.

4. A personnel tracking and marking device, characterized in that: include: An initialization module, used for acquiring captured image information of travelers, and initializing the captured image information of the travelers to obtain target traveler image information of target travelers; a calculation module, configured to extract a target attribute feature value from the target traveler image information, perform similarity calculation between the target attribute feature value from the target traveler image information and a first attribute feature value in a first database, and determine a suspicious marked person from the traveler according to the similarity; A tracking module, used to track the suspicious marked person based on a preset tracking model, determine a target location area, and match a target attribute feature value of the suspicious marked person with a second attribute feature value in a second database of the target location area; A marking module, used to mark the suspicious marked person according to the matching result, obtain the target marked person and push it; The initialization module includes: The first acquisition unit is used to acquire travelers captured at a preset time period and at a preset location; A clustering unit, used for extracting the captured image information of the travelers, clustering the captured image information, and marking the captured image information that fails in clustering as target traveler image information; The calculation module comprises: A second acquisition unit is used to acquire the first attribute characteristic value in the first database, where the first attribute characteristic value in the first database is a characteristic value acquired in the preset time period and not in the preset location; A calculation unit, used for calculating the similarity between the target attribute feature value in the target traveler image information and the first attribute feature value to obtain a similarity value; A first judging unit, configured to judge whether the similarity value reaches a preset similarity threshold; The extraction unit is used for extracting the traveler whose similarity value reaches the preset similarity threshold as the suspicious marked person if the similarity value reaches the preset similarity threshold.

5. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in the personnel tracking and marking method according to any one of claims 1 to 3 when executing the computer program.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the personnel tracking and marking method according to any one of claims 1 to 3 are implemented.

Images