Personnel supervision method and device based on face recognition, equipment and medium

Abstract

The invention relates to the technical field of personnel supervision, in particular to a personnel supervision method and device based on face recognition, equipment and a medium, and the method comprises the steps: recognizing a positioning identifier contained in positioning data, and determining a face identifier corresponding to the positioning identifier according to a first mapping relation; acquiring a field image, and determining staying images of the target person in different divided areas according to the positioning identifier, the face identifier and the field image; according to each staying image, determining intersection personnel information corresponding to the target personnel in different divided areas; acquiring intersection audio data corresponding to each intersection duration from the field audio data according to the intersection personnel information; and according to the staying duration of the target personnel in each divided area and a preset flow chart, generating a staying duration flow chart, and superposing the intersection personnel information and the intersection audio data corresponding to different divided areas into the staying duration flow chart to obtain full-flow supervision data. The validity of the supervision data can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of personnel supervision, and in particular to a personnel supervision method, device, equipment and medium based on face recognition. Background Art

[0002] The case handling area refers to a special area where the case handling agency conducts security checks, information collection, questioning, identification and other case handling activities in accordance with the law during the case handling process. The case handling area is often used to interrogate suspects, defendants, witnesses, victims or other persons involved in the case, and is also used to receive relevant informed persons and obtain audio-visual materials.

[0003] The flow of personnel in the case handling area is complex, and it is difficult to determine the identity and number of personnel. There may also be multiple events that are cross-processed, so it is difficult to manage the personnel in the case handling area. In related technologies, wristband positioning is generally used to locate personnel in the case handling area, that is, a positioning wristband is assigned to each person entering the case handling area, and the positioning data generated by the positioning wristband is used to monitor the personnel in the case handling area. However, the positioning wristband can only record and monitor the places where the personnel in the case handling area pass, and other behavioral data cannot be obtained. In addition, the signal of the positioning wristband may be interfered by obstacles such as buildings and metal objects in the case handling area. Therefore, the supervision data generated by the positioning wristband is less effective. Summary of the invention

[0004] In order to improve the effectiveness of supervision data and thus increase the case handling rate, the present application provides a personnel supervision method, device, equipment and medium based on face recognition.

[0005] In the first aspect, the present application provides a personnel supervision method based on face recognition, which adopts the following technical solution: A personnel supervision method based on face recognition, comprising: Identify a positioning identifier included in the positioning data, and determine a face identifier corresponding to the positioning identifier according to a first mapping relationship, where the first mapping relationship is a correspondence relationship between the positioning identifier and the face identifier; Acquire a scene image, and determine the stay image of a target person in different divided areas according to the positioning mark, the face mark and the scene image, wherein the target person is a person corresponding to the face mark, and a plurality of the divided areas constitute a case handling area; According to each stay image, determine the intersection personnel information corresponding to the target person in different divided areas, wherein the intersection personnel information includes the intersection personnel and the intersection duration; Obtaining intersection audio data corresponding to each intersection duration from the on-site audio data according to the intersection personnel information; According to the target person's stay time in each divided area and the preset flowchart, a stay time flowchart is generated, and the intersection personnel information and intersection audio data corresponding to different divided areas are superimposed to the corresponding positions in the stay time flowchart to obtain full process supervision data.

[0006] By adopting the above-mentioned technical solution, face positioning is added on the basis of bracelet positioning, that is, the personnel in the case handling area are tracked and supervised by jointly using positioning data and face identification, so as to accurately determine the position and activity trajectory of the target personnel in the case handling area. At the same time, the intersection of the target personnel with other personnel in the case handling area is monitored and recorded. The recording results are helpful to analyze the interaction and correlation between the target personnel and the intersection personnel, and can provide data support for subsequent case decisions. Finally, the intersection personnel information and the intersection audio data are superimposed and then fed back. By combining the audio information, the target personnel are more comprehensively supervised and analyzed, which is convenient to improve the effectiveness of the supervision data, thereby reducing the workload of relevant case handling personnel and thus improving the case handling rate.

[0007] In a possible implementation, determining, according to each stop image, intersection personnel information corresponding to the target person in different divided areas includes: Identify the stay location of the target person in each stay image, where the stay location is the location where the target person stays in the corresponding divided area for a time period exceeding a first preset stay time period; According to each stop location and the preset stop interval, determine the observation area corresponding to each stop location, and identify the intersection persons appearing in each observation area, wherein the intersection persons are persons who stay in the observation area with the target person for a stay time longer than a second preset stay time; According to the intersection personnel corresponding to each divided area and the intersection duration corresponding to the intersection personnel, the intersection personnel information corresponding to each divided area is generated.

[0008] By adopting the above technical solution, by dividing the case handling area into multiple areas, and then counting and analyzing the intersection personnel information corresponding to each divided area, it is convenient to improve the accuracy of determining the intersection personnel information. Since the personnel in the case handling area are complex and the number of personnel is large, personnel screening is performed by preset stay time and preset stay interval, which can further improve the accuracy of determining the intersection personnel information.

[0009] In a possible implementation, obtaining the intersection audio data corresponding to each intersection duration from the on-site audio data according to the intersection personnel information includes: Identifying an intersection person image corresponding to the intersection person information from the scene image; Determine whether there is a preset feature in the intersection person image, and if so, determine the audio interception time period corresponding to each intersection person information according to the preset feature; According to the intersection personnel information and each audio interception time period, the intersection audio data corresponding to each intersection duration is determined from the on-site audio data.

[0010] By adopting the above technical solution, the conversation time between the target person and the intersection person is determined through feature recognition. Instead of superimposing all the on-site audio data, the audio interception time period is determined through feature recognition. By superimposing the audio data corresponding to the audio interception time period, it is convenient to improve the effectiveness of the whole process supervision data.

[0011] In a possible implementation manner, when the number of intersecting persons in the observation area is higher than a preset number of persons, the method further includes: Obtaining the interval distance, intersection duration, and preset features between each intersection person and the target person in the observation area; Determine the distance score corresponding to each intersection person according to the interval distance between each intersection person and the target person and the second mapping relationship, wherein the second mapping relationship is the correspondence between the interval distance and the distance score; Determine the intersection score corresponding to each intersection person according to the intersection duration between each intersection person and the target person and a third mapping relationship, wherein the third mapping relationship is a correspondence between the intersection duration and the intersection score; Identify the preset feature type of each intersection person, and determine the feature score corresponding to each intersection person according to the preset feature type of each intersection person and a fourth mapping relationship, wherein the fourth mapping relationship is a correspondence relationship between the preset feature type and the feature score; According to the distance score, intersection score and feature score corresponding to each intersection person, multiple intersection persons are screened to obtain target intersection persons, and the number of intersection persons included in the target intersection persons is not higher than the preset number of persons.

[0012] By adopting the above technical solution, multiple intersection persons are screened by interval distance, intersection duration and preset feature type to obtain target intersection persons, instead of identifying all intersection persons appearing in the observation area as target intersection persons. Personnel screening facilitates improving the effectiveness of the full-process supervision data.

[0013] In one possible implementation, the method further includes: Determine the appearance frequency of each intersection person according to the full-process supervision data, and determine the intersection person whose appearance frequency is higher than the preset frequency as the first initial associated person, wherein the appearance frequency is used to characterize the number of times the intersection person appears in all divided areas; According to the full-process supervision data, the intersection time ratio of each remaining intersection person is determined, and the remaining intersection persons whose intersection time ratio is higher than the preset time ratio are determined as the second initial associated persons, and the remaining intersection persons are the intersection persons other than the first initial associated persons, and the intersection time ratio is used to characterize the time that the intersection persons spend with the target person in all divided areas; Determine the first initially associated person and the second initially associated person as associated suspicious persons of the target person; Acquire the to-be-verified information of the associated suspicious person and the target verification information of the target person, respectively identify the abnormal feature information contained in the to-be-verified information and the target verification information, and compare the to-be-verified information with the target verification information to obtain a matching value corresponding to each abnormal feature, wherein the abnormal feature information includes movement trajectory information, economic information, and work and rest information; Generate verification feedback information based on the matching value corresponding to each abnormal feature and each abnormal feature By adopting the above technical solution, by analyzing the number of appearances and the proportion of intersection time of each intersection person in different divided areas, the intersection persons corresponding to the target person can be further screened, and after comparing with the pending verification information corresponding to the finally determined related suspicious persons, verification feedback information is generated to facilitate the relevant case handling personnel to verify the identity of the target person. By analyzing the pending verification information of the related suspicious persons, it can also be verified whether the target person has related cases or cases with historical case records.

[0014] In one possible implementation, the method further includes: Identify an abnormal feature in the verification feedback information whose matching value is higher than a preset matching value, and determine the abnormal feature as a target abnormal feature; Traverse the historical triggering rule setting cases according to the target abnormal characteristics to determine the personnel associated with the historical triggering rule setting; The historical verification information corresponding to the historical trigger rule setting personnel is obtained, and the historical verification information is fed back together with the verification feedback information.

[0015] By adopting the above technical solution, instead of determining the associated triggering rule setting personnel based on any abnormal feature, the target abnormal feature is determined from the abnormal feature through a preset matching value, and then the associated historical triggering rule setting personnel is determined based on the target abnormal feature. This can improve the accuracy in determining the associated historical triggering rule setting personnel, and by providing feedback on the historical verification information of the historical triggering rule setting personnel, it is convenient to reduce the workload of the relevant case handling personnel in the case handling process. In addition, by analyzing and viewing the relevant historical cases when handling the current case, the case handling rate of the relevant case handling personnel may be improved.

[0016] In the second aspect, the present application provides a personnel supervision device based on face recognition, which adopts the following technical solution: A personnel monitoring device based on face recognition, comprising: A positioning identifier recognition module, used to recognize a positioning identifier contained in the positioning data, and determine a face identifier corresponding to the positioning identifier according to a first mapping relationship, wherein the first mapping relationship is a correspondence relationship between the positioning identifier and the face identifier; A module for determining a stay image is used to obtain an on-site image, and determine a stay image of a target person in different divided areas according to the positioning mark, the face mark and the on-site image, wherein the target person is a person corresponding to the face mark, and a plurality of the divided areas constitute a case handling area; A module for determining intersection personnel information is used to determine, according to each stop image, intersection personnel information corresponding to the target person in different divided areas, wherein the intersection personnel information includes intersection personnel and intersection duration; An intersection audio data acquisition module is used to acquire the intersection audio data corresponding to each intersection duration from the on-site audio data according to the intersection personnel information; Determine the full-process supervision data module, which is used to generate a stay time flowchart based on the target person's stay time in each divided area and the preset flowchart, and superimpose the intersection personnel information and intersection audio data corresponding to different divided areas to the corresponding positions in the stay time flowchart to obtain the full-process supervision data.

[0017] By adopting the above-mentioned technical solution, face positioning is added on the basis of bracelet positioning, that is, the personnel in the case handling area are tracked and supervised by jointly using positioning data and face identification, so as to accurately determine the position and activity trajectory of the target personnel in the case handling area. At the same time, the intersection of the target personnel with other personnel in the case handling area is monitored and recorded. The recording results are helpful to analyze the interaction and correlation between the target personnel and the intersection personnel, and can provide data support for subsequent case decisions. Finally, the intersection personnel information and the intersection audio data are superimposed and then fed back. By combining the audio information, the target personnel are more comprehensively supervised and analyzed, which is convenient to improve the effectiveness of the supervision data, thereby reducing the workload of relevant case handling personnel and thus improving the case handling rate.

[0018] In a third aspect, the present application provides an electronic device, which adopts the following technical solution: An electronic device, comprising: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the above-mentioned personnel supervision method based on face recognition.

[0019] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium includes: a computer program that can be loaded by a processor and execute the above-mentioned personnel supervision method based on face recognition.

[0020] In summary, the present application includes at least one of the following beneficial technical effects: On the basis of bracelet positioning, face positioning is added, that is, the personnel in the case handling area are tracked and supervised through positioning data and face identification, so as to accurately determine the position and activity trajectory of the target personnel in the case handling area. At the same time, the intersection of the target personnel with other personnel in the case handling area is monitored and recorded. The recording results are helpful to analyze the interaction and relationship between the target personnel and the intersection personnel, and can provide data support for subsequent case decisions. Finally, the intersection personnel information and the intersection audio data are superimposed and then fed back. By combining the audio information, the target personnel are more comprehensively supervised and analyzed, which is convenient to improve the effectiveness of the supervision data, thereby reducing the workload of relevant case handling personnel and thus improving the case handling rate.

[0021] The conversation time between the target person and the intersection person is determined through feature recognition. This does not mean that all on-site audio data are superimposed. Instead, the audio interception time period is determined through feature recognition. By superimposing the audio data corresponding to the audio interception time period, it is convenient to improve the effectiveness of the full-process supervision data. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flowchart of a personnel supervision method based on face recognition in an embodiment of the present application; Figure 2 This is a schematic diagram of full-process supervision data in an embodiment of the present application; Figure 3 It is a schematic diagram of a method for determining verification feedback information in an embodiment of the present application; Figure 4 It is a flow chart of a personnel supervision device based on face recognition in an embodiment of the present application; Figure 5 It is a structural schematic diagram of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following is combined with Figure 1-5 This application is described in further detail.

[0024] After reading this specification, those skilled in the art may make non-creative modifications to this embodiment as needed, but such modifications are protected by patent law as long as they are within the scope of the claims of this application.

[0025] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0026] Specifically, the embodiment of the present application provides a personnel supervision method based on face recognition, which is executed by an electronic device, which can be a server or a terminal device, wherein the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present application.

[0027] refer to Figure 1 , Figure 1 : is a flow chart of a personnel supervision method based on face recognition in an embodiment of the present application, the method comprises steps S110 to S150, wherein: Step S110: identifying a positioning identifier included in the positioning data, and determining a face identifier corresponding to the positioning identifier according to a first mapping relationship, where the first mapping relationship is a correspondence relationship between the positioning identifier and the face identifier.

[0028] Specifically, the positioning data can be sent by the positioning device and uploaded to the Dianzi device. The positioning device can be a device with positioning function such as a positioning bracelet. When the relevant personnel enter the case handling area, the relevant staff will configure a positioning device for each relevant personnel to monitor the actions of the relevant personnel in the case handling area, and the positioning device will be taken back after the relevant personnel leave the case handling area.

[0029] Different positioning devices generate different positioning identifiers, that is, different people correspond to different positioning identifiers. Different positioning devices are distinguished by using different positioning identifiers, that is, different people are distinguished by using different positioning identifiers, which facilitates improving the accuracy of determining facial identification based on positioning data. The positioning identifier can be a device number or a special character. The specific positioning identifier is not specifically limited in the embodiments of the present application and can be set by relevant technical personnel.

[0030] The first mapping relationship includes multiple positioning identifiers and corresponding facial identifiers. The facial identifier corresponding to each positioning identifier can be determined through the first mapping relationship. The specific content of the first mapping relationship is not specifically limited in the embodiment of the present application, and can be updated and modified in real time by relevant technical personnel based on the actual personnel entering and leaving the case handling area.

[0031] Step S120: Acquire the scene image, and determine the target person's stay image in different divided areas according to the positioning mark, face mark and scene image. The target person is the person corresponding to the face mark, and multiple divided areas constitute the case handling area.

[0032] Specifically, the scene image can be collected by the image acquisition device set in the case handling area and uploaded to the electronic device. The case handling area is usually equipped with a command center, an interrogation room, an inquiry room, a polygraph room, an identification room, a lounge, a waiting room, a toilet and other facilities. Therefore, the case handling area can be divided according to the function, and the case handling area can also be divided by area. The specific division method is not specifically limited in the embodiment of the present application. By dividing the case handling area into regions, and then further identifying or analyzing each divided area, it is convenient to improve the accuracy of the analysis result. The stay image corresponding to the different divided areas is the image of the target person in the divided area. When determining the stay image corresponding to the different divided areas from the scene image according to the positioning mark and the face mark, it can be first determined according to the positioning mark whether the target person has entered the divided area. If it is detected that the target person has entered the divided area, the face mark is used to re-identify the person in the divided area to determine the target person in the divided area. Since the divided area may contain multiple people at the same time, when the person is detected to enter the divided area, the face mark is used to identify again, which can improve the accuracy of determining the stay image, that is, it can ensure that at least the target person is included in the stay image.

[0033] Step S130: determining the intersection personnel information corresponding to the target person in different divided areas according to each stay image, where the intersection personnel information includes the intersection personnel and the intersection duration.

[0034] Specifically, the intersection personnel information corresponding to the target person in different divided areas is the intersection personnel corresponding to the target person in different divided areas and the intersection duration with each intersection personnel. Since different divided areas correspond to different area types or area areas, the number of personnel corresponding to different divided areas is also different. Correspondingly, the intersection personnel information corresponding to the target person in different divided areas may be different. The intersection personnel are personnel who have been in the same location with the target person for a period of time. The intersection duration is the length of time the intersection personnel and the target person have been together. Furthermore, in order to improve the accuracy of determining the intersection personnel information, the intersection personnel information corresponding to the target person in different divided areas is determined according to each stop image. The intersection personnel information includes the intersection personnel and the intersection duration, specifically including: Identify the target person's stay location in each stay image, where the stay location is the location where the target person stays in the corresponding divided area for a period of time exceeding the first preset stay time; determine the observation area corresponding to each stay location based on each stay location and the preset stay interval, and identify the intersection persons appearing in each observation area, where the intersection persons are persons who are in the observation area with the target person and whose stay time is longer than the second preset stay time; generate the intersection person information corresponding to each divided area based on the intersection persons corresponding to each divided area and the intersection time corresponding to the intersection persons.

[0035] Specifically, since the target person may move around in the divided area, if each position in the divided area is used as a stopover location, it is necessary to judge whether the person appearing in the entire divided area is an intersection person of the target person, so a large amount of data calculation may be generated. By limiting the stay time, the stopover location of the target person in the divided area can be determined, which can improve the accuracy of determining the stopover location, thereby reducing the probability of generating invalid data. The stopover location can be a specific location, or a location area with a specific location as the center and a preset distance as the radius. The specific form of the stopover location is not specifically limited in the embodiment of the present application, and can be set by relevant technical personnel. When the stopover location is a location area, the specific preset distance can be 0.5 meters or 1 meter. The specific distance is not specifically limited in the embodiment of the present application, and can be set by relevant technical personnel. The first preset stay time can be 1 minute or 1.5 minutes. The specific time is not specifically limited in the embodiment of the present application, and can be set by relevant technical personnel. When it is detected that the target person stays at a certain location or a location area for a period of time longer than the first preset stay time, the location or the location area is determined as the stopover location.

[0036] The preset stop spacing is greater than the preset distance, and the preset stop spacing can be 2 meters, or 2.5 meters. The specific spacing is not specifically limited in the embodiment of the present application. The circular area formed by taking the stop location as the center and the preset stop spacing as the radius is the observation area of ​​the stop location. Since the observation area belongs to a certain position in the divided area, and the divided area may contain other moving personnel, when determining the intersection personnel of the target personnel, the intersection personnel of the target personnel are also determined from multiple personnel by adding a limit on the length of stay, that is, by recording the length of stay corresponding to each person entering the observation area, the person whose length of stay in the observation area is longer than the second preset length of stay is determined as the intersection personnel of the target personnel, wherein the second preset length of stay is similar to the first preset length of stay, and can be 1 minute, or can be 1.5 minutes. The specific length of stay is not specifically limited in the embodiment of the present application. The intersection duration corresponding to the intersection personnel is the coexistence time of the intersection personnel and the related personnel in the observation area. When determining the intersection duration, the timing can be started from the time when the intersection personnel is identified, and the timing is stopped when the target personnel or the intersection personnel is detected to leave. The intersection personnel information is composed of the intersection personnel and the intersection duration corresponding to the intersection personnel, among which the number of intersection personnel can be 0, 1 or more, and the specific number is not specifically limited in the embodiment of this application. Due to the complexity of the personnel in the case handling area and the large number of personnel, the accuracy of determining the intersection personnel information can be further improved by screening the personnel by presetting the stay time and the preset stay interval.

[0037] Step S140: Obtain intersection audio data corresponding to each intersection duration from the on-site audio data according to the intersection personnel information.

[0038] Specifically, the on-site audio data can be collected by a sound receiving device set in the case handling area and uploaded to an electronic device. It can also be collected by a sound receiving device worn by personnel entering the case handling area and uploaded to an electronic device. The method of obtaining the on-site audio data is not specifically limited in the embodiment of the present application and can be set by relevant technical personnel. When determining the intersection audio data corresponding to the intersection duration, the intersection time period can be first determined based on the intersection duration, and then intercepted from the on-site audio data based on the intersection time period. Furthermore, in order to improve the effectiveness of the full-process supervision data, the intersection audio data corresponding to each intersection duration is obtained from the on-site audio data based on the intersection personnel information, specifically including: Identify the intersection person image corresponding to the intersection person information from the on-site image; determine whether there is a preset feature in the intersection person image, and if so, determine the audio interception time period corresponding to each intersection person information according to the preset feature; determine the intersection audio data corresponding to each intersection duration from the on-site audio data according to the intersection person information and each audio interception time period.

[0039] Specifically, the intersection person image is the time period corresponding to the intersection duration, and the corresponding scene image, that is, in the intersection person image, the target person and the intersection person are in the same observation area. The preset features are lip movements, facial movements, etc. When the preset features appear in the intersection person image, it indicates that the intersection person or the target person is speaking. For example, when a person is speaking, the lips will open and close and move in a series of ways. These movements can be identified and analyzed through image processing technology. For example, the classification of lip movements using machine learning algorithms can identify whether the intersection person image contains lip movements; when a person is speaking, other facial muscles will also undergo corresponding changes, such as changes in muscle movements in the eyebrows, nose, and cheeks. These changes can be identified and analyzed through image processing and machine learning technology. The method for determining whether the intersection person image contains preset features is not specifically limited in the embodiments of the present application, and can be selected and set by relevant technical personnel.

[0040] The moment when the preset feature is detected is determined as the starting moment of the audio interception time period, and the moment when the preset feature disappears is determined as the end moment of the audio interception time period. If the preset feature is not detected within the preset time period, it means that the preset feature has disappeared. The preset time period can be 30 seconds or 50 seconds. The specific duration corresponding to the preset time period is not specifically limited in the embodiment of the present application.

[0041] After the audio interception time period is determined, the audio data on the scene can be intercepted according to the audio interception time period. Since there can be multiple intersecting persons, the intersection durations corresponding to different intersecting persons may be different, and correspondingly, the intersection audio data corresponding to different intersecting persons may also be different. The conversation moment between the target person and the intersecting person is determined by feature recognition. This does not mean that all the on-site audio data is superimposed. Instead, the audio interception time period is determined by feature recognition. By superimposing the audio data corresponding to the audio interception time period, the effectiveness of the full-process supervision data can be improved.

[0042] Step S150: Generate a stay time flowchart based on the target person's stay time in each divided area and the preset flowchart, and superimpose the intersection personnel information and intersection audio data corresponding to different divided areas to the corresponding positions in the stay time flowchart to obtain full process supervision data.

[0043] Specifically, the number of nodes in the preset flow chart is a preset number, and each node corresponds to a divided area. Since the case handling area is generally divided into multiple areas, in order to reduce the modification operation steps of the preset flow chart in the process of generating the stay time flow chart, the preset flow chart generally does not pre-set a part of the nodes, and the preset number can be 5 or 6. The specific number is not specifically limited in the embodiment of the present application, and can be set by relevant technical personnel according to the historical regional division situation. Identify each divided area in the preset flow chart, and mark the target person's stay time in each divided area at the corresponding node, wherein the target person's stay time in each divided area is the sum of the stay time of the target person at each stay place in each divided area. For example, the divided area a contains two stay places, namely, stay place a1 and stay place a2. If the target person stays at stay place a1 for 5 minutes and stays at stay place a2 for 4 minutes, it is determined that the target person stays at the divided area a for 9 minutes. In addition, the intersection personnel information corresponding to each divided area needs to be superimposed and marked at the corresponding node in the preset flow chart. The display form of the intersection personnel information in the preset flow chart can be a list display or an information list display, such as Figure 2 As shown, Figure 2 The information is displayed in a listed manner, and the specific display method is not specifically limited in the embodiments of this application.

[0044] In addition, when the target person has a conversation with other people in a certain designated area, the intersection audio data generated during the conversation must be superimposed. After superimposing the intersection audio data in the stay time flow chart, full-process supervision data is generated. The generated full-process supervision data is fed back to the terminal device of the relevant staff so that the relevant staff can effectively supervise each person entering the case handling area.

[0045] On the basis of bracelet positioning, face positioning is added, that is, the personnel in the case handling area are tracked and supervised through positioning data and face identification, so as to accurately determine the position and activity trajectory of the target personnel in the case handling area. At the same time, the intersection of the target personnel with other personnel in the case handling area is monitored and recorded. The recording results are helpful to analyze the interaction and relationship between the target personnel and the intersection personnel, and can provide data support for subsequent case decisions. Finally, the intersection personnel information and the intersection audio data are superimposed and then fed back. By combining the audio information, the target personnel are more comprehensively supervised and analyzed, which is convenient to improve the effectiveness and reliability of the supervision data, thereby reducing the workload of relevant case handling personnel and thus improving the case handling rate.

[0046] Furthermore, when the number of intersecting persons in the observation area is higher than the preset number of persons, the embodiment of the present application further includes: Obtain the interval distance, intersection duration and preset features between each intersection person and the target person in the observation area; determine the distance score corresponding to each intersection person according to the interval distance between each intersection person and the target person and the second mapping relationship, the second mapping relationship being the correspondence between the interval distance and the distance score; determine the intersection score corresponding to each intersection person according to the intersection duration between each intersection person and the target person and the third mapping relationship, the third mapping relationship being the correspondence between the intersection duration and the intersection score; identify the preset feature type of each intersection person, determine the feature score corresponding to each intersection person according to the preset feature type of each intersection person and the fourth mapping relationship, the fourth mapping relationship being the correspondence between the preset feature type and the feature score; screen multiple intersection persons according to the distance score, intersection score and feature score corresponding to each intersection person to obtain the target intersection person, the number of intersection persons included in the target intersection person is not higher than the preset number of persons.

[0047] Specifically, in the embodiment of the present application, the intersection personnel are determined from the observation area. However, when there are a large number of people coming and going in the observation area, not all the people coming and going are determined as target intersection personnel, but it is necessary to screen from multiple intersection personnel to obtain the target intersection personnel. The interval distance, intersection duration, and preset features between each intersection person and the target person can be identified through the on-site image, and then each intersection person is scored based on the interval distance, intersection duration, and preset feature type corresponding to each intersection person, and finally the screening is performed according to the score of each intersection person.

[0048] The second mapping relationship includes distance scores corresponding to different interval distances, and the shorter the interval distance, the higher the corresponding distance score. According to the second mapping relationship, the distance score of each intersection person can be determined; the third mapping relationship includes intersection scores corresponding to different intersection durations, and the longer the intersection duration, the higher the corresponding intersection score. According to the third mapping relationship, the intersection score of each intersection person can be determined; the fourth mapping relationship includes feature scores corresponding to different preset feature types, and the feature score of each intersection person can be determined according to the fourth mapping relationship. The specific contents of the second mapping relationship, the third mapping relationship and the fourth mapping relationship are not specifically limited in the embodiments of the present application and can be set by relevant technical personnel.

[0049] According to the distance score, intersection score and feature score of each intersection person, the total score corresponding to each intersection person is calculated, and the total scores corresponding to multiple intersection persons are sorted in order from small to large or from large to small to obtain a total score sequence, and according to the total score sequence, the total score sequence is eliminated from the end with the smallest total score until the number of remaining intersection persons is no more than the preset number of persons, wherein the preset number of persons may be 3 or 4, and the specific number is not specifically limited in the embodiments of the present application and may be set by relevant technical personnel. Multiple intersection persons are screened by interval distance, intersection duration and preset feature type to obtain target intersection persons, rather than determining all intersection persons appearing in the observation area as target intersection persons, and personnel screening facilitates the effectiveness of the whole process supervision data.

[0050] Furthermore, in order to facilitate the relevant case handling personnel to verify the identity of the target person, the method also includes steps S1 to S5, such as Figure 3 As shown, where: Step S1: According to the full-process supervision data, the appearance frequency of each intersection person is determined, and the intersection person whose appearance frequency is higher than the preset frequency is determined as the first initial associated person. The appearance frequency is used to characterize the number of times the intersection person appears in all divided areas.

[0051] Specifically, since the full-process supervision data contains all the information of the intersection personnel, that is, the intersection personnel appearing in each divided area can be identified through the full-process supervision data, and the number of times each intersection personnel appears in all divided areas is determined by counting the intersection personnel in each divided area. The higher the number of appearances, the higher the frequency of appearance of the intersection personnel, that is, the greater the probability that the intersection personnel becomes the first initial associated person. Among them, the specific preset frequency is not specifically limited in the embodiment of the present application and can be set by relevant technical personnel.

[0052] Step S2: Based on the full-process supervision data, determine the intersection time ratio of each remaining intersection person, and determine the remaining intersection persons whose intersection time ratio is higher than the preset time ratio as the second initial associated persons. The remaining intersection persons are the intersection persons other than the first initial associated persons. The intersection time ratio is used to characterize the length of time that the intersection persons spend with the target person in all divided areas.

[0053] Specifically, multiple intersection persons can be preliminarily screened according to the preset frequency, and the intersection persons whose appearance frequency is higher than the preset frequency are the first initial associated persons, and the intersection persons whose appearance frequency is not higher than the preset frequency are the remaining intersection persons. According to the full-process supervision data, the total intersection time of each remaining intersection person with the target person in all divided areas can be determined, and the intersection time ratio of the remaining intersection persons is the ratio of the total intersection time to the total stay time, wherein the total stay time is the sum of the stay time of the target person in all divided areas. The higher the intersection time ratio, the higher the probability that the remaining intersection person becomes the second initial associated person. When the intersection time ratio of the remaining intersection person is higher than the preset time ratio, the remaining intersection person is determined as the second initial associated person. The specific preset time ratio is not specifically limited in the embodiment of the present application and can be set by relevant technical personnel.

[0054] Step S3: Determine the first initially associated person and the second initially associated person as associated suspicious persons of the target person.

[0055] Step S4: Obtain the information to be verified of the associated suspicious person and the target verification information of the target person, identify the abnormal feature information contained in the information to be verified and the target verification information respectively, and compare the information to be verified with the target verification information to obtain the matching value corresponding to each abnormal feature. The abnormal feature information includes movement trajectory information, economic information, and work and rest information.

[0056] Specifically, the number of times the first initially associated person appears in the divided area is related to the intersection person, and the length of time the second initially associated person spends together in the divided area is related to the intersection person. By treating both the first initially associated person and the second initially associated person as associated suspicious persons, the probability of omissions when determining associated suspicious persons is reduced.

[0057] When obtaining the information to be verified of the associated suspicious person, face recognition can be performed from the on-site image, and then based on the face recognition result of the associated suspicious person, the corresponding information to be verified can be retrieved from the electronic device. It is also possible to identify the location identification of the associated suspicious person and retrieve the corresponding information to be verified from the electronic device through the location identification of the associated person. The specific method of obtaining the information to be verified is not specifically limited in the embodiments of this application and can be set by relevant technical personnel. The specific method of obtaining the target verification information corresponding to the target person can refer to the above method of obtaining the information to be verified, which will not be repeated here.

[0058] The information to be verified and the target verification information at least include movement trajectory information, economic information and work and rest information. The abnormal feature information in the information to be verified and the target verification information can be identified by feature recognition. For each abnormal feature information, the abnormal feature information in the information to be verified is matched with the corresponding abnormal feature information in the target verification information, and the matching value corresponding to each abnormal feature information can be obtained, that is, the matching value of each abnormal feature. The specific matching method is not specifically limited in the embodiment of the present application, as long as the matching value corresponding to each abnormal feature can be obtained after the abnormal feature information is compared. Among them, the action trajectory information includes the departure place, destination, departure time, arrival time, etc., and the action trajectory information in the information to be verified is matched with the action trajectory information in the target verification information to determine the matching value corresponding to the action trajectory information; the economic information includes employment record information, payment bill information, housing information, etc., and the economic information in the information to be verified is matched with the economic information in the target verification information to determine the matching value corresponding to the economic information; the work and rest information includes Internet record information and housing record information, and the work and rest information in the information to be verified is matched with the work and rest information in the target verification information to determine the matching value corresponding to the work and rest information.

[0059] Step S5: Generate verification feedback information according to the matching value corresponding to each abnormal feature and each abnormal feature.

[0060] Specifically, the verification feedback information includes each abnormal feature and the matching value corresponding to each abnormal feature. After the verification feedback information is generated, the verification feedback information can be fed back to the terminal device of the relevant case handling personnel to facilitate the relevant case handling personnel to verify the identity of the target person. By analyzing the information to be verified of related suspicious persons, it can also be verified whether the target person has related cases or cases with historical case records.

[0061] Furthermore, in order to reduce the workload of relevant case handlers and thus improve the case handling rate of relevant case handlers, the method further includes: Identify abnormal features in the verification feedback information whose matching values ​​are higher than the preset matching values, and determine the abnormal features as target abnormal features; traverse the historical trigger setting rule cases according to the target abnormal features, and determine the associated historical trigger setting rule personnel; obtain the historical verification information corresponding to the historical trigger setting rule personnel, and feed back the historical verification information together with the verification feedback information.

[0062] Specifically, the preset matching value can be 89% or 95%. The specific preset matching value is not specifically limited in the embodiment of the present application and can be set by relevant technical personnel. When the matching value corresponding to a certain abnormal feature is detected to be higher than the preset matching value, the abnormal feature is determined as the target abnormal feature. By identifying the associated historical trigger setting rule personnel from the historical trigger setting rule personnel corresponding to the historical trigger setting rule case according to the target abnormal feature, and feeding back the historical verification information corresponding to the associated historical trigger setting rule personnel, it is convenient to provide more comprehensive case handling information for the relevant case handling personnel. At the same time, screening by the target abnormal feature is conducive to reducing the workload of the relevant case handling personnel when analyzing and screening the suspicious persons in the target trigger setting rule case, thereby improving the work efficiency of the case handling personnel when handling the case.

[0063] The above embodiment introduces a personnel supervision method based on face recognition from the perspective of method flow, and the following embodiment introduces a personnel supervision device based on face recognition from the perspective of a virtual module or a virtual unit. For details, please refer to the following embodiment.

[0064] The present application embodiment provides a personnel monitoring device based on face recognition, such as Figure 4 As shown, the device may specifically include a positioning identification module 410, a stop image determination module 420, an intersection personnel information determination module 430, an intersection audio data acquisition module 440, and a full-process supervision data determination module 450, wherein: A positioning identifier identification module 410 is used to identify the positioning identifier contained in the positioning data, and determine the face identifier corresponding to the positioning identifier according to a first mapping relationship, where the first mapping relationship is a correspondence relationship between the positioning identifier and the face identifier; The module 420 for determining the image of a target person is used to obtain an image of the scene, and to determine the image of a target person in different divided areas according to the positioning mark, the face mark and the scene image. The target person is the person corresponding to the face mark, and the multiple divided areas constitute a case handling area. The intersection personnel information determination module 430 is used to determine the intersection personnel information corresponding to the target personnel in different divided areas according to each stay image, and the intersection personnel information includes the intersection personnel and the intersection duration; The intersection audio data acquisition module 440 is used to acquire the intersection audio data corresponding to each intersection duration from the on-site audio data according to the intersection personnel information; Determine the full-process supervision data module 450, which is used to generate a stay time flowchart based on the target person's stay time in each divided area and the preset flowchart, and superimpose the intersection personnel information and intersection audio data corresponding to different divided areas to the corresponding positions in the stay time flowchart to obtain the full-process supervision data.

[0065] In a possible implementation, when the intersection personnel information determining module 430 determines the intersection personnel information corresponding to the target person in different divided areas according to each stay image, it is specifically used to: Identify the target person's stay location in each stay image, where the stay location is the location where the target person stays in the corresponding divided area for a period exceeding a first preset stay time; According to each stop location and the preset stop interval, determine the observation area corresponding to each stop location, and identify the intersection persons appearing in each observation area, where the intersection persons are persons who stay in the observation area with the target person for a period longer than the second preset stay time; According to the intersection personnel corresponding to each divided area and the intersection duration corresponding to the intersection personnel, the intersection personnel information corresponding to each divided area is generated.

[0066] In a possible implementation, when the intersection audio data acquisition module 440 acquires the intersection audio data corresponding to each intersection duration from the on-site audio data according to the intersection personnel information, it is specifically used to: Identify intersection personnel images corresponding to intersection personnel information from the scene images; Determine whether there is a preset feature in the intersection person image, and if so, determine the audio interception time period corresponding to each intersection person information according to the preset feature; According to the intersection personnel information and each audio interception time period, the intersection audio data corresponding to each intersection duration is determined from the on-site audio data.

[0067] In a possible implementation, when the number of intersecting persons in the observation area is higher than a preset number of persons, the device further includes: A feature acquisition module is used to obtain the interval distance, intersection duration and preset features between each intersection person and the target person in the observation area; A distance score determination module is used to determine the distance score corresponding to each intersection person according to the interval distance between each intersection person and the target person and the second mapping relationship, where the second mapping relationship is the correspondence between the interval distance and the distance score; An intersection score determination module is used to determine the intersection score corresponding to each intersection person according to the intersection duration between each intersection person and the target person and a third mapping relationship, where the third mapping relationship is the correspondence between the intersection duration and the intersection score; A feature score determination module is used to identify the preset feature type of each intersection person, and determine the feature score corresponding to each intersection person according to the preset feature type of each intersection person and a fourth mapping relationship, where the fourth mapping relationship is a correspondence between the preset feature type and the feature score; The module for determining target intersection personnel is used to screen multiple intersection personnel according to the distance score, intersection score and feature score corresponding to each intersection personnel to obtain target intersection personnel. The number of intersection personnel included in the target intersection personnel is not higher than the preset number of personnel.

[0068] In a possible implementation, the device further includes: A module for determining the first initial associated personnel is used to determine the frequency of occurrence of each intersection personnel based on the full-process supervision data, and to determine the intersection personnel whose frequency of occurrence is higher than the preset frequency as the first initial associated personnel. The frequency of occurrence is used to characterize the number of times the intersection personnel appear in all divided areas. A module for determining the second initially associated persons is used to determine the intersection time ratio of each remaining intersection person according to the full-process supervision data, and to determine the remaining intersection persons whose intersection time ratio is higher than the preset time ratio as the second initially associated persons. The remaining intersection persons are the intersection persons other than the first initially associated persons, and the intersection time ratio is used to characterize the time that the intersection persons spend with the target person in all divided areas; A module for determining associated suspicious persons, used for determining the first initial associated person and the second initial associated person as associated suspicious persons of the target person; A matching value determination module is used to obtain the to-be-verified information of the associated suspicious person and the target verification information of the target person, respectively identify the abnormal feature information contained in the to-be-verified information and the target verification information, and compare the to-be-verified information with the target verification information to obtain the matching value corresponding to each abnormal feature. The abnormal feature information includes action trajectory information, economic information, and work and rest information; The verification feedback information generating module is used to generate verification feedback information according to the matching value corresponding to each abnormal feature and each abnormal feature.

[0069] In a possible implementation, the device further includes: A target abnormal feature determination module is used to identify abnormal features in the verification feedback information whose matching values ​​are higher than a preset matching value, and determine the abnormal features as target abnormal features; A module for determining personnel associated with historical trigger setting rules is used to traverse historical trigger setting rule cases according to target abnormal characteristics and determine personnel associated with historical trigger setting rules; A historical verification information module is determined to obtain the historical verification information corresponding to the historical trigger setting rule personnel, and to feed back the historical verification information together with the verification feedback information.

[0070] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the personnel supervision device based on face recognition described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0071] An electronic device is provided in an embodiment of the present application, such as Figure 5 As shown, Figure 5 The electronic device 500 shown includes: a processor 501 and a memory 503. The processor 501 and the memory 503 are connected, such as through a bus 502. Optionally, the electronic device 500 may also include a transceiver 504. It should be noted that in actual applications, the transceiver 504 is not limited to one, and the structure of the electronic device 500 does not constitute a limitation on the embodiments of the present application.

[0072] Processor 501 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 501 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0073] The bus 502 may include a path to transmit information between the above components. The bus 502 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 502 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0074] The memory 503 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0075] The memory 503 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 501. The processor 501 is used to execute the application code stored in the memory 503 to implement the contents shown in the above method embodiment.

[0076] The electronic devices include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and fixed terminals such as digital TVs, desktop computers, etc. It can also be a server, etc. Figure 5 The electronic device 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.

[0077] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment.

[0078] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0079] The above description is only a partial implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A personnel supervision method based on face recognition, characterized in that: include: Identify a positioning identifier included in the positioning data, and determine a face identifier corresponding to the positioning identifier according to a first mapping relationship, where the first mapping relationship is a correspondence relationship between the positioning identifier and the face identifier; Acquire a scene image, and determine the stay image of a target person in different divided areas according to the positioning mark, the face mark and the scene image, wherein the target person is a person corresponding to the face mark, and a plurality of the divided areas constitute a case handling area; According to each stay image, determine the intersection personnel information corresponding to the target person in different divided areas, wherein the intersection personnel information includes the intersection personnel and the intersection duration; Obtaining intersection audio data corresponding to each intersection duration from the on-site audio data according to the intersection personnel information; According to the target person's stay time in each divided area and the preset flowchart, a stay time flowchart is generated, and the intersection personnel information and intersection audio data corresponding to different divided areas are superimposed to the corresponding positions in the stay time flowchart to obtain full process supervision data.

2. According to claim 1, a personnel supervision method based on face recognition is characterized in that: Determining, according to each stop image, intersection personnel information corresponding to the target person in different divided areas, includes: Identify the stay location of the target person in each stay image, where the stay location is the location where the target person stays in the corresponding divided area for a time period exceeding a first preset stay time period; According to each stop location and the preset stop interval, determine the observation area corresponding to each stop location, and identify the intersection persons appearing in each observation area, wherein the intersection persons are persons who stay in the observation area with the target person for a stay time longer than a second preset stay time; According to the intersection personnel corresponding to each divided area and the intersection duration corresponding to the intersection personnel, the intersection personnel information corresponding to each divided area is generated.

3. The personnel supervision method based on face recognition according to claim 1 is characterized in that: The step of obtaining the intersection audio data corresponding to each intersection duration from the on-site audio data according to the intersection personnel information includes: Identifying an intersection person image corresponding to the intersection person information from the scene image; Determine whether there is a preset feature in the intersection person image, and if so, determine the audio interception time period corresponding to each intersection person information according to the preset feature; According to the intersection personnel information and each audio interception time period, the intersection audio data corresponding to each intersection duration is determined from the on-site audio data.

4. The personnel supervision method based on face recognition according to claim 2 is characterized in that: When the number of intersecting persons in the observation area is higher than the preset number of persons, the method further includes: Obtaining the interval distance, intersection duration, and preset features between each intersection person and the target person in the observation area; Determine the distance score corresponding to each intersection person according to the interval distance between each intersection person and the target person and the second mapping relationship, wherein the second mapping relationship is the correspondence between the interval distance and the distance score; Determine the intersection score corresponding to each intersection person according to the intersection duration between each intersection person and the target person and a third mapping relationship, wherein the third mapping relationship is a correspondence between the intersection duration and the intersection score; Identify the preset feature type of each intersection person, and determine the feature score corresponding to each intersection person according to the preset feature type of each intersection person and a fourth mapping relationship, wherein the fourth mapping relationship is a correspondence relationship between the preset feature type and the feature score; According to the distance score, intersection score and feature score corresponding to each intersection person, multiple intersection persons are screened to obtain target intersection persons, and the number of intersection persons included in the target intersection persons is not higher than the preset number of persons.

5. The personnel supervision method based on face recognition according to claim 1 is characterized in that: Also includes: Determine the appearance frequency of each intersection person according to the full-process supervision data, and determine the intersection person whose appearance frequency is higher than the preset frequency as the first initial associated person, wherein the appearance frequency is used to characterize the number of times the intersection person appears in all divided areas; According to the full-process supervision data, the intersection time ratio of each remaining intersection person is determined, and the remaining intersection persons whose intersection time ratio is higher than the preset time ratio are determined as the second initial associated persons, and the remaining intersection persons are the intersection persons other than the first initial associated persons, and the intersection time ratio is used to characterize the time that the intersection persons spend with the target person in all divided areas; Determine the first initially associated person and the second initially associated person as associated suspicious persons of the target person; Acquire the to-be-verified information of the associated suspicious person and the target verification information of the target person, respectively identify the abnormal feature information contained in the to-be-verified information and the target verification information, and compare the to-be-verified information with the target verification information to obtain a matching value corresponding to each abnormal feature, wherein the abnormal feature information includes movement trajectory information, economic information, and work and rest information; Verification feedback information is generated based on the matching value corresponding to each abnormal feature and each abnormal feature.

6. A personnel supervision method based on face recognition according to claim 5, characterized in that: Also includes: Identify an abnormal feature in the verification feedback information whose matching value is higher than a preset matching value, and determine the abnormal feature as a target abnormal feature; Traverse the historical triggering rule setting cases according to the target abnormal characteristics to determine the personnel associated with the historical triggering rule setting; The historical verification information corresponding to the historical trigger rule setting personnel is obtained, and the historical verification information is fed back together with the verification feedback information.

7. A personnel monitoring device based on face recognition, characterized in that: include: A positioning identifier recognition module, used to recognize a positioning identifier contained in the positioning data, and determine a face identifier corresponding to the positioning identifier according to a first mapping relationship, wherein the first mapping relationship is a correspondence relationship between the positioning identifier and the face identifier; A module for determining a stay image is used to obtain an on-site image, and determine a stay image of a target person in different divided areas according to the positioning mark, the face mark and the on-site image, wherein the target person is a person corresponding to the face mark, and a plurality of the divided areas constitute a case handling area; A module for determining intersection personnel information is used to determine, according to each stop image, intersection personnel information corresponding to the target person in different divided areas, wherein the intersection personnel information includes intersection personnel and intersection duration; An intersection audio data acquisition module is used to acquire the intersection audio data corresponding to each intersection duration from the on-site audio data according to the intersection personnel information; Determine the full-process supervision data module, which is used to generate a stay time flowchart based on the target person's stay time in each divided area and the preset flowchart, and superimpose the intersection personnel information and intersection audio data corresponding to different divided areas to the corresponding positions in the stay time flowchart to obtain the full-process supervision data.

8. The personnel monitoring device based on face recognition according to claim 7, characterized in that: The intersection personnel information determination module is specifically used to determine the intersection personnel information corresponding to the target person in different divided areas according to each stay image: Identify the stay location of the target person in each stay image, where the stay location is the location where the target person stays in the corresponding divided area for a time period exceeding a first preset stay time period; According to each stop location and the preset stop interval, determine the observation area corresponding to each stop location, and identify the intersection persons appearing in each observation area, wherein the intersection persons are persons who stay in the observation area with the target person for a stay time longer than a second preset stay time; According to the intersection personnel corresponding to each divided area and the intersection duration corresponding to the intersection personnel, the intersection personnel information corresponding to each divided area is generated.

9. An electronic device, characterized in that: The electronic device includes: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute a personnel supervision method based on face recognition as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that: include: A computer program is stored which can be loaded by a processor and executes a personnel supervision method based on face recognition as described in any one of claims 1 to 6.

Images