Face recognition security management system based on data analysis

Abstract

The present invention discloses a face recognition security management system based on data analysis, relates to the field of face recognition technology, and comprises a data acquisition terminal, a data analysis module, an adjustment height acquisition module, an adjustment sequence table acquisition module and a time period recommended height acquisition module; the technical problem that when a face recognition machine performs face recognition on a plurality of persons of different heights at the same time, the camera height is adjusted frequently and disorderly, resulting in a burden on camera adjustment is solved; by analyzing historical recognition data, the optimal recognition height can be automatically determined for each recognition height interval, which helps the camera capture face images at the optimal position and angle, thereby improving the efficiency and accuracy of face recognition; the adjustment control module adjusts the camera height in sequence according to the adjustment sequence table, gives priority to processing height intervals with a large number of people, and further improves the overall security management efficiency.

Description

Technical Field

[0001] The present invention relates to the field of face recognition technology, and in particular to a face recognition security management system based on data analysis. Background Art

[0002] The face recognition security management system based on data analysis is a comprehensive system that integrates advanced face recognition technology and big data analysis capabilities. This system collects images containing faces through cameras, automatically detects and tracks faces, and then recognizes the detected faces. It also uses big data analysis technology to process, analyze and mine face image data to achieve more efficient, accurate and safe security management. The face monitoring system can perform real-time face recognition in public places, transportation hubs and commercial areas, is responsible for the collection, storage and management of face information, and provides interfaces for other systems to call to ensure data security and availability;

[0003] The patent publication number CN220829734U discloses a face recognition campus security management system, which relates to a security system, including a sub-access control device 1, a sub-access control device 2, a face recognition machine, and a main control module and a storage module arranged in the sub-access control device 1, wherein the face recognition machine and the storage module are both electrically connected to the main control module; the face recognition machine is installed on the front side of the sub-access control device 1 through a lifting drive mechanism, and the front side of the sub-access control device 1 is provided with a plurality of infrared receivers installed at different heights, and the front side of the sub-access control device 2 is provided with a plurality of infrared transmitters whose installation heights correspond to the infrared receivers one by one, and the infrared receivers are electrically connected to the main control module. The patent realizes automatic adaptive adjustment of the height of the face recognition machine, improves the practicality of the face recognition campus security management system, and solves the problem that the installation height of the face recognition machine in the current system is fixed and cannot be applied to students and faculty and staff groups on the same system;

[0004] However, although the face recognition machine can adjust the height of the camera when performing face recognition on people of different heights, when face recognition is required for multiple people of different heights, the camera height needs to be adjusted frequently, resulting in frequent and disorderly adjustment of the camera height, which imposes a complex burden on the camera height control program. Based on this, a face recognition security management system based on data analysis is proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a face recognition security management system based on data analysis, which solves the technical problem that when a face recognition machine performs face recognition on multiple people of different heights at the same time, the camera height is adjusted frequently and disorderly, causing a burden on camera adjustment.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The face recognition security management system based on data analysis includes:

[0008] The data analysis module is used to analyze the historical recognition data of the face recognition camera, thereby obtaining the recognition height interval corresponding to the camera, and at the same time obtaining the optimal recognition height corresponding to each recognition height interval;

[0009] The height adjustment acquisition module acquires the people to be identified and analyzes their height data to obtain the height to be adjusted, and controls the height of the face recognition camera according to the height to be adjusted;

[0010] An adjustment sequence table acquisition module is used to obtain an adjustment sequence table of the face recognition machine camera according to the adjustment heights corresponding to each adjustment interval and the number of persons to be recognized corresponding to each adjustment interval;

[0011] The module for obtaining the recommended height for each time period is used to obtain the historical recognition data of the face recognition camera within a preset time period T. At the same time, a day is divided into multiple time periods, and the historical recognition data of the face recognition camera in each time period is analyzed to obtain the recommended height corresponding to the face recognition camera in each time period.

[0012] As a further solution of the present invention: the specific method of obtaining the optimal recognition height corresponding to each recognition height interval is:

[0013] The height range acquisition unit in the data analysis module analyzes the height values ​​of people in the historical recognition data of the face recognition camera, and then obtains the corresponding height range of the camera;

[0014] Through the optimal recognition height acquisition unit in the data analysis module, the height of the face recognition machine camera in each recognition height interval is analyzed each time it is adjusted in the preset n recognitions, and then the optimal recognition height corresponding to each recognition height interval is obtained, where n is a preset value, and n≥1.

[0015] As a further solution of the present invention: the specific method of obtaining the identified height range is:

[0016] Obtain the height values ​​of the identified persons corresponding to the preset n identifications by the face recognition camera, and mark them as Gn respectively. Here, the same values ​​in the person's height values ​​Gn are marked as the same height state value, and then obtain the corresponding height state value Ga in the person's height value Gn, where a refers to the number corresponding to the height state value Ga, and n≥a≥1;

[0017] According to the numerical values ​​corresponding to each height status value Ga, the height status values ​​Ga are sorted in ascending order, and every two adjacent height status values ​​are used as an identification height interval, thereby obtaining a-1 identification height intervals.

[0018] As a further solution of the present invention: the optimal recognition height corresponding to each recognition height interval is obtained in the following specific manner:

[0019] S1: randomly select a height interval from each identified height interval as the target interval;

[0020] S2: Obtain the recognition height and recognition time corresponding to the face recognition machine camera when the face recognition machine camera recognizes the person whose height value is within the target interval for a preset b times, and take the recognition height with the shortest recognition time as the best recognition height corresponding to the target interval, and mark it as H1, where b is a preset value, b≥1;

[0021] S3: Repeat the above steps S1-S2 to obtain the optimal recognition height He corresponding to each recognition height interval, where e is the number of recognition height intervals, e=a-1.

[0022] As a further solution of the present invention: the specific way to obtain the height to be adjusted is:

[0023] When face recognition is required for multiple persons, the height values ​​corresponding to each person to be identified are obtained, and the identification height ranges corresponding to each person to be identified are obtained according to the height values ​​corresponding to each person to be identified, and used as the adjustment range, and the optimal identification height corresponding to each adjustment range is obtained, and marked as the adjustment height Hc, where c is the number corresponding to the adjustment range, e≥c≥1.

[0024] As a further solution of the present invention: the specific method of obtaining the adjustment sequence table of the face recognition camera is:

[0025] The number of persons to be identified corresponding to each adjustment interval is obtained, and is used as Rc to obtain the real-time height E of the face recognition camera;

[0026] The priority coefficient Fc corresponding to each adjustment height is obtained by the formula Rc×β1+(1 / |Hc-E|)×β2=Fc, where β1 and β2 are both preset fixed coefficients;

[0027] According to the values ​​of the priority coefficients Fc corresponding to the respective adjustment heights, the respective adjustment heights Hc are sorted in order from large to small, thereby obtaining an adjustment sequence table of the face recognition machine camera.

[0028] As a further solution of the present invention: the specific method of obtaining the recommended heights corresponding to the face recognition camera in each time period is:

[0029] S01: randomly select a time period from multiple time periods as the target time period;

[0030] S02: Obtain the height values ​​of the identified persons corresponding to the face recognition camera in the target time period every day within the preset time period T, and obtain the number of the height values ​​of the identified persons in each identification height interval according to the height values ​​of the identified persons corresponding to the target time period every day, and mark them as LTe respectively, where T is the preset time period, and T≥1;

[0031] S001: Select a height interval as the analysis interval;

[0032] S002: Obtain the number MT of height values ​​of people recognized by the face recognition camera every day within the preset time period T and falling within the analysis interval, perform discrete analysis on MT, and obtain the human flow value F1 corresponding to the analysis interval within the target time period;

[0033] S003: Repeat steps S001-S002 to obtain the crowd flow values ​​Fe corresponding to each identified height interval in the target time period, obtain the identified height interval corresponding to the maximum value, and use the corresponding adjustment height as the recommended height J1 of the face recognition camera in the target time period;

[0034] S03: Repeat steps S01-S02 to obtain the recommended height Jv corresponding to the face recognition camera in each time period, and bind it to the corresponding time period one by one, where v is the number of time periods, v≥1.

[0035] As a further solution of the present invention: the specific method of performing discrete analysis on MT is:

[0036] By formula The discrete value U of MT is calculated. When the discrete value U is less than the preset value Y1, the mean value of MT is used as the passenger flow value F1 corresponding to the analysis interval within the target time period. When the discrete value is greater than or equal to the preset value Y1, the mean of the maximum and minimum values ​​in MT is used as the passenger flow value F1 corresponding to the analysis interval within the target time period, where Mp is the mean value of MT, and T≥r≥1.

[0037] As a further solution of the present invention: a specific method of dividing a day into multiple time periods is:

[0038] According to a preset time interval Q, 24 hours of a day are equally divided to obtain multiple time periods, where Q is a preset value.

[0039] Beneficial effects of the present invention:

[0040] (1) The present invention can automatically determine the optimal recognition height for each recognition height interval by analyzing historical recognition data, which helps the camera capture facial images at the optimal position and angle, thereby improving the efficiency and accuracy of facial recognition; the adjustment control module adjusts the camera height in sequence according to the adjustment sequence table, giving priority to height intervals with a large number of people, which further improves the recognition efficiency, reduces waiting time, ensures the accuracy and speed of the facial recognition system, and improves the overall security management efficiency;

[0041] (2) The present invention automatically adjusts the face recognition machine camera to the recommended height of the corresponding time period according to the recommended height corresponding to each time period, so that the face recognition machine can timely identify the corresponding person with the height most likely to appear in the corresponding time period, and helps the camera to capture the face image at the best position, thereby improving the accuracy of face recognition. At the same time, it can more intelligently predict and adapt to the height distribution of people in different time periods, thereby improving the accuracy and efficiency of face recognition and further enhancing the overall security management effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The present invention will be further described below in conjunction with the accompanying drawings.

[0043] Figure 1 It is a schematic diagram of the system framework structure of the present invention;

[0044] Figure 2 It is a schematic diagram of the framework structure of the method of the present invention. DETAILED DESCRIPTION

[0045] 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.

[0046] Embodiment 1

[0047] See also Figure 1-Figure 2 As shown, the present invention is a face recognition security management system based on data analysis, comprising:

[0048] The data acquisition end is used for the historical recognition data of the face recognition camera. The historical recognition data includes the height value of the person, the height value of the camera and the angle value. The height value of the person is captured, collected and recorded through person registration, camera or infrared detector. The height and angle of the camera are recorded each time it is adjusted.

[0049] The data analysis module is used to analyze the historical recognition data of the face recognition camera, and then obtain the recognition height interval corresponding to the camera, and at the same time obtain the optimal recognition height corresponding to each recognition height interval. The specific method is as follows:

[0050] The height range acquisition unit in the data analysis module is used to analyze the height values ​​of people in the historical recognition data of the face recognition camera, and then obtain the height range corresponding to the camera. The specific method of obtaining the height range is as follows:

[0051] Obtain the height values ​​of the identified persons corresponding to the preset n recognitions by the face recognition camera, and mark them as Gn, where n is a preset value, and the specific value is formulated by relevant personnel according to actual needs, n≥1;

[0052] Mark the same values ​​in the height values ​​Gn of the personnel as the same height status value, and then obtain the corresponding height status value Ga in the height values ​​Gn of the personnel, where a refers to the number corresponding to the height status value Ga, n≥a≥1;

[0053] According to the numerical values ​​corresponding to the height status values ​​Ga, the height status values ​​Ga are sorted in ascending order, and every two adjacent height status values ​​are used as an identification height interval, thereby obtaining a-1 identification height intervals;

[0054] The optimal recognition height acquisition unit in the data analysis module analyzes the height of the face recognition camera in each recognition height interval when it is adjusted in the preset n recognitions, and then obtains the optimal recognition height corresponding to each recognition height interval, specifically in the following manner:

[0055] S1: randomly select a height interval from each identified height interval as the target interval;

[0056] S2: Obtain the recognition height and recognition time corresponding to the face recognition machine camera when the face recognition machine camera recognizes the person whose height value is within the target interval for the preset b times, and take the recognition height with the shortest recognition time as the optimal recognition height corresponding to the target interval, and mark it as H1, where b is a preset value, and the specific value is formulated by relevant personnel according to actual needs, b≥1;

[0057] S3: Repeat the above steps S1-S2 to obtain the optimal recognition height He corresponding to each recognition height interval, where e is the number of recognition height intervals, e=a-1;

[0058] The height adjustment acquisition module acquires the people to be identified and acquires and analyzes their height data to obtain the height to be adjusted. The specific method is as follows:

[0059] When multiple persons need to be recognized, the height values ​​corresponding to the persons to be recognized are obtained, and the recognition height intervals corresponding to the persons to be recognized are obtained according to the height values ​​corresponding to the persons to be recognized, and the intervals are used as adjustment intervals, and the optimal recognition heights corresponding to the adjustment intervals are obtained, and the adjustment heights are marked as adjustment heights Hc, where c is the number of adjustment intervals, and e≥c≥1;

[0060] The adjustment sequence table acquisition module is used to obtain the adjustment sequence table of the face recognition camera according to the adjustment heights corresponding to each adjustment interval and the number of people to be recognized corresponding to each adjustment interval. The specific method is as follows:

[0061] The number of persons to be identified corresponding to each adjustment interval is obtained, and is used as Rc to obtain the real-time height E of the face recognition camera;

[0062] The priority coefficient Fc corresponding to each adjustment height is obtained by the formula Rc×β1+(1 / |Hc-E|)×β2=Fc, where β1 and β2 are both preset fixed coefficients, and the specific values ​​are formulated by relevant personnel according to actual needs;

[0063] According to the values ​​of the priority coefficients Fc corresponding to the respective adjustment heights, the respective adjustment heights Hc are sorted in descending order, thereby obtaining an adjustment sequence table of the face recognition machine camera;

[0064] The adjustment control module is used to adjust the height of the face recognition camera in sequence according to the order of each adjustment height in the adjustment sequence table of the face recognition camera, which is conducive to face recognition of people in a height range with a large number of people in the crowd to be recognized;

[0065] By analyzing historical recognition data, the optimal recognition height can be automatically determined for each recognition height range, which helps the camera capture facial images at the optimal position and angle, thereby improving the efficiency and accuracy of facial recognition; the adjustment control module adjusts the camera height in sequence according to the adjustment sequence table, giving priority to height ranges with a larger number of people, which further improves recognition efficiency, reduces waiting time, ensures the accuracy and speed of the facial recognition system, and improves overall security management efficiency.

[0066] Embodiment 2

[0067] As the second embodiment of the present invention, when the present application is implemented, compared with the first embodiment, the technical solution of the present embodiment is different from that of the first embodiment only in the present embodiment;

[0068] The module for obtaining the recommended height during a period is used to obtain the historical recognition data of the face recognition camera within a preset time period T. At the same time, a day is divided into multiple periods, and the historical recognition data of the face recognition camera in each period is analyzed to obtain the recommended height corresponding to the face recognition camera in each period. The specific method is as follows:

[0069] It should be noted that T is the preset time length, which is 60 days here, indicating that the time period of 60 days is counted from the moment the data is obtained, and the moment the data is obtained is not included;

[0070] Divide a day into multiple time periods, which can be equal or unequal. Here, the 24 hours of a day are equally divided according to the preset time interval Q, and then multiple time periods are obtained. Q is a preset value, and the specific value is formulated by relevant personnel according to actual needs;

[0071] S01: randomly select a time period from multiple time periods as the target time period;

[0072] S02: Obtain the height values ​​of the identified persons corresponding to the face recognition camera in the target time period every day within the preset time period T, and obtain the number of the height values ​​of the identified persons in each identification height interval according to the height values ​​of the identified persons corresponding to the target time period every day, and mark them as LTe respectively;

[0073] S001: Select a height interval as the analysis interval;

[0074] S002: Obtain the number MT of height values ​​of people recognized by the face recognition camera every day within the preset time T that are within the analysis interval,

[0075] By formula Calculate the discrete value U of MT. When the discrete value U is less than the preset value Y1, the mean value of MT is used as the flow value F1 corresponding to the analysis interval in the target period. When the discrete value is greater than or equal to the preset value Y1, the mean of the maximum and minimum values ​​in MT is used as the flow value F1 corresponding to the analysis interval in the target period, where Mp is the mean value of MT, and T≥r≥1.

[0076] S003: Repeat steps S001-S002 in this way, obtain the crowd flow values ​​Fe corresponding to each identified height interval in the target time period, obtain the identified height interval corresponding to the maximum value, and obtain the adjustment height corresponding to the identified height interval, and use it as the recommended height J1 corresponding to the face recognition camera in the target time period, and bind it to the corresponding target time period;

[0077] S03: Repeat steps S01-S02 to obtain the recommended height Jv corresponding to the face recognition camera in each time period, and bind it to the corresponding time period one by one, v is the number of corresponding time periods, v≥1;

[0078] When no one passes by the facial recognition machine, the time period recommended height acquisition module can automatically adjust the facial recognition machine camera to the recommended height of the corresponding time period according to the height distribution of people in different time periods and the recommended height corresponding to each time period. This facilitates the facial recognition machine to promptly identify the corresponding person of the height most likely to appear in the corresponding time period, helps the camera capture facial images at the best position, thereby improving the accuracy of facial recognition, and can more intelligently predict and adapt to the height distribution of people in different time periods, thereby improving the accuracy and efficiency of facial recognition and further improving the overall security management effect.

[0079] Embodiment 3

[0080] As the third embodiment of the present invention, when the present application is specifically implemented, compared with the first and second embodiments, the technical solution of this embodiment is to combine the solutions of the first and second embodiments mentioned above for implementation.

[0081] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0082] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. The face recognition security management system based on data analysis is characterized by: include: The data analysis module is used to analyze the historical recognition data of the face recognition camera, thereby obtaining the recognition height interval corresponding to the camera, and at the same time obtaining the optimal recognition height corresponding to each recognition height interval; The height adjustment acquisition module acquires the people to be identified and analyzes their height data to obtain the height to be adjusted, and controls the height of the face recognition camera according to the height to be adjusted; An adjustment sequence table acquisition module is used to obtain an adjustment sequence table of the face recognition machine camera according to the adjustment heights corresponding to each adjustment interval and the number of persons to be recognized corresponding to each adjustment interval; The module for obtaining the recommended height during a period is used to obtain the historical recognition data of the face recognition camera within a preset time period T, and to divide a day into multiple periods, analyze the historical recognition data of the face recognition camera in each period, and then obtain the recommended height corresponding to the face recognition camera in each period; The specific method of obtaining the optimal recognition height corresponding to each recognition height interval is: The height range acquisition unit in the data analysis module analyzes the height values ​​of people in the historical recognition data of the face recognition camera, and then obtains the corresponding height range of the camera; The optimal recognition height acquisition unit in the data analysis module analyzes the height of the face recognition machine camera adjusted each time in the preset n recognitions in each recognition height interval, thereby obtaining the optimal recognition height corresponding to each recognition height interval, where n is a preset value, and n≥1; The specific method for obtaining the identified height range is: Obtain the height values ​​of the identified persons corresponding to the preset n identifications by the face recognition camera, and mark them as Gn respectively. Here, the same values ​​in the person's height values ​​Gn are marked as the same height state value, and then obtain the corresponding height state value Ga in the person's height value Gn, where a refers to the number corresponding to the height state value Ga, and n≥a≥1; According to the numerical values ​​corresponding to the height status values ​​Ga, the height status values ​​Ga are sorted in ascending order, and every two adjacent height status values ​​are used as an identification height interval, thereby obtaining a-1 identification height intervals; The optimal recognition height corresponding to each recognition height interval is obtained in the following way: S1: randomly select a height interval from each identified height interval as the target interval; S2: Obtain the recognition height and recognition time corresponding to the face recognition machine camera when the face recognition machine camera recognizes the person whose height value is within the target interval for a preset b times, and take the recognition height with the shortest recognition time as the best recognition height corresponding to the target interval, and mark it as H1, where b is a preset value, b≥1; S3: Repeat the above steps S1-S2 to obtain the optimal recognition height He corresponding to each recognition height interval, where e is the number of recognition height intervals, e=a-1; The specific method of obtaining the height to be adjusted is: When multiple persons need to be recognized, the height values ​​corresponding to the persons to be recognized are obtained, and the recognition height intervals corresponding to the persons to be recognized are obtained according to the height values ​​corresponding to the persons to be recognized, and the intervals are used as adjustment intervals, and the optimal recognition heights corresponding to the adjustment intervals are obtained, and the adjustment heights are marked as adjustment heights Hc, where c is the number of adjustment intervals, and e≥c≥1; The specific method of obtaining the adjustment sequence table of the face recognition camera is: The number of persons to be identified corresponding to each adjustment interval is obtained, and is used as Rc to obtain the real-time height E of the face recognition camera; The priority coefficient Fc corresponding to each adjustment height is obtained by the formula Rc×β1+(1 / |Hc-E|)×β2=Fc, where β1 and β2 are both preset fixed coefficients; According to the values ​​of the priority coefficients Fc corresponding to the respective adjustment heights, the respective adjustment heights Hc are sorted in descending order, thereby obtaining an adjustment sequence table of the face recognition machine camera; The specific method for obtaining the recommended heights of the face recognition camera in different time periods is as follows: S01: randomly select a time period from multiple time periods as the target time period; S02: Obtain the height values ​​of the identified persons corresponding to the face recognition camera in the target time period every day within the preset time period T, and obtain the number of the height values ​​of the identified persons in each identification height interval according to the height values ​​of the identified persons corresponding to the target time period every day, and mark them as LTe respectively, where T is the preset time period, and T≥1; S001: Select a height interval as the analysis interval; S002: Obtain the number MT of height values ​​of people recognized by the face recognition camera every day within the preset time period T and falling within the analysis interval, perform discrete analysis on MT, and obtain the human flow value F1 corresponding to the analysis interval within the target time period; S003: Repeat steps S001-S002 to obtain the crowd flow values ​​Fe corresponding to each identified height interval in the target time period, obtain the identified height interval corresponding to the maximum value, and use the corresponding adjustment height as the recommended height J1 of the face recognition camera in the target time period; S03: Repeat steps S01-S02 to obtain the recommended height Jv corresponding to the face recognition camera in each time period, and bind it to the corresponding time period one by one, where v is the number of time periods, v≥1.

2. The face recognition security management system based on data analysis according to claim 1 is characterized in that: The specific method of discrete analysis of MT is: By formula The discrete value U of MT is calculated. When the discrete value U is less than the preset value Y1, the mean value of MT is used as the passenger flow value F1 corresponding to the analysis interval within the target time period. When the discrete value is greater than or equal to the preset value Y1, the mean of the maximum and minimum values ​​in MT is used as the passenger flow value F1 corresponding to the analysis interval within the target time period, where Mp is the mean value of MT, and T≥r≥1.

3. The face recognition security management system based on data analysis according to claim 2 is characterized in that: The specific way to divide a day into multiple periods is as follows: According to a preset time interval Q, 24 hours of a day are equally divided to obtain multiple time periods, where Q is a preset value.

Images