Information Record Management System Based on Face Recognition Technology

Abstract

The present invention discloses an information record management system based on face recognition technology. The present invention relates to the field of information management technology and solves the problem that the information record of the existing face recognition technology is complicated and cumbersome in management. The present invention provides a structured face information entry and storage module. By refining the division and storage of face information, the system can efficiently process and manage the information record of face recognition technology. By setting the acquisition and analysis of all matching information within a preset time T, the system can timely adjust the storage location and access rights of face information, thereby improving the efficiency and security of the entire authorization process. By classifying face information in detail and storing data in different quadrants according to recognition frequency, it is convenient to reasonably manage the information record of face recognition technology.

Description

Technical Field

[0001] The present invention relates to the technical field of information management, and in particular to an information record management system based on face recognition technology. Background Art

[0002] In today's digital age, face recognition technology has been widely used in various information record management systems, especially in the fields of security, surveillance, and personal identity verification. Modern face recognition information record management systems use advanced cameras and image processing software to confirm personal identity by analyzing the structural features of the face. These systems can quickly capture facial images from video streams or photos and compare them with facial data stored in the database to achieve fast and automatic identity verification.

[0003] For example, the Chinese invention patent with publication number CN110378093A discloses a management system based on face recognition technology. The system includes multiple modules. The continuation information entry module allows managers to enter information about continuation personnel and their permissions so that work handover can be carried out when necessary. The abnormal detection module sends an abnormal signal when the pre-stored facial information does not match the current real-time captured facial information. This serves as a layer of detection for system security. When the work continuation module receives an abnormal signal, if the currently logged-in manager has entered the continuation information, this information will be used to determine whether there is authority to continue the work, thereby loading the management authority content into the currently logged-in management interface, etc. Its purpose is to provide a management system based on face recognition technology to assist in improving the efficiency of managers in continuing work.

[0004] Although face recognition information record management provides significant improvements in many aspects, these systems also have some technical and operational challenges. First, the information records of existing face recognition technologies are complex and cumbersome to manage, and there are certain limitations in managing and storing face recognition information records. Summary of the invention

[0005] In view of the deficiencies of the prior art, the present invention provides an information record management system based on face recognition technology, which solves the problem that the information record of the prior art face recognition technology is complicated and cumbersome in management.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an information record management system based on face recognition technology, comprising:

[0007] The face information input module is used to collect and input new face information, record the new face information as newly added face information, and send it to the subsequent face information storage module;

[0008] A face information storage module is used to store authorized face information and newly added face information, and mark them as Rj, where j=1, 2, ..., m, m represents the total face information, and j represents the jth face recognition information in the total face information;

[0009] The face information classification module is used to obtain all matching information within a preset time T, including successful matching information and failed matching information, and determine high-frequency face information, medium-frequency face information and low-frequency face information by analyzing the matching information. The preset time T is a time period of n days from the current moment, where n=1, 2, ..., 30;

[0010] A storage division module is used to reasonably store high-frequency face information, medium-frequency face information and low-frequency face information by constructing a plane rectangular coordinate system and drawing circles to determine different quadrants;

[0011] A face information adjustment module, used to further adjust the storage of high-frequency face information, medium-frequency face information and low-frequency face information in different quadrants of face information by constructing circle 2;

[0012] The rapid identification module is used to quickly compare and identify the collected identification information with the face information to determine whether a successful match is generated or a failed match is generated.

[0013] Preferably, in the face information division module, the successful matching information includes the identification information and the face information Rj, and the failed matching information only includes the identification information.

[0014] Preferably, the specific method of determining high-frequency face information, medium-frequency face information and low-frequency face information is:

[0015] S1: Obtain the matching information of each day in the previous n days, then determine the successful matching information from the matching information, and then obtain the face information Rj from all the successful matching information;

[0016] S2: Count the authorization records of facial information Rj every day in n days, obtain the number of days x that each Rj appears in n days, and mark it as R j-x , where x = 1, 2, ..., n, when R j-x / n>Q1, it is marked as high-frequency face information, when Q2 <R j-x / n≤Q1, it is marked as medium-frequency face information. j-x / n≤Q2 is marked as low-frequency face information, where Q1 and Q2 are preset values.

[0017] Preferably, in the storage partition module, the specific storage method is:

[0018] P1: Construct a plane rectangular coordinate system, that is, there are the first quadrant, the second quadrant, the third quadrant and the fourth quadrant. Then, with the origin as the center, construct the circle equation r1²=x²+y² to draw a circle, marked as circle 1. Circle 1 is combined with the horizontal and vertical coordinates to divide circle 1 into four areas, where r1 is the preset value:

[0019] That is, the first quadrant area, the second quadrant area, the third quadrant area and the fourth quadrant area;

[0020] P2: Store high-frequency face information in the first quadrant, medium-frequency face information in the second quadrant, low-frequency face information in the third quadrant, and newly added face information in the fourth quadrant.

[0021] Preferably, in the face information adjustment module, the specific adjustment method is:

[0022] AS1: Then, with the origin as the center, construct the circle equation r2²=x²+y² again to draw a circle, marked as circle 2, where r2>r1, and r2 is the preset value;

[0023] Divided into a first quadrant sector, a second quadrant sector, a third quadrant sector and a fourth quadrant sector;

[0024] AS2: In the high-frequency face information in the first quadrant, randomly select a high-frequency face information as the target information Rj, and obtain the number of days R that the target information Rj appears in n days. j-x , and in R j-x The number of authorized records per day in a day Hc, 1≤c≤R j-x , through the formula , and get R j-x The average number of authorization records in a day is Hp, and then the formula is used , get the discrete value Wb;

[0025] AS3: Compare the discrete value Wb with the preset value Q3 to determine the evaluation value Pj of the target information Rj:

[0026] AS4: Repeat steps 2 to 3 to determine that all high-frequency face information is used as the evaluation value Pj of the target information;

[0027] AS5: The high-frequency face information with an evaluation value Pj greater than the preset threshold Q4 is retained in the first quadrant, and the high-frequency face information with an evaluation value Pj less than or equal to the preset threshold Q4 is transferred to the first quadrant fan-shaped area;

[0028] The intermediate frequency face information or the low frequency face information is synchronously transferred to the corresponding second quadrant sector area or the third quadrant sector area according to the same principle.

[0029] Preferably, in step AS3, it further includes:

[0030] When Wb>Q3, Hc is sorted in descending order of |Hc-Hp|, and the first Hc value is deleted each time, and the discrete value Wb of the remaining Hc is recalculated until Wb≤Q3, and the number g1 of deleted Hc is obtained, and g1 is judged. When g1>gs, the average value of g1 deleted Hc is calculated as the evaluation value Pj of the target information Rj;

[0031] When g1≤gs, the average value of the remaining Hc is calculated as the evaluation value Pj of the target information Rj, where gs is a preset value, which is specifically set by professionals;

[0032] When Wb≤Q3, the average value Hp is used as the evaluation value Pj of the target information Rj.

[0033] Preferably, in the quick identification module, the identification method is:

[0034] The collected recognition information is first compared with the face information in the first quadrant. If the comparison fails, it is then compared with the fan-shaped area in the first quadrant.

[0035] When the comparison and judgment of the first quadrant sector area is not successful, the comparison and judgment is then carried out with the second quadrant area. If the comparison and judgment is not successful, the comparison and judgment is then carried out with the second quadrant sector area;

[0036] And so on, until the fourth quadrant sector is matched, if no match is successful, it is determined that the recognition fails and a failed match information is generated. If the above situation is matched successfully, a successful match information is generated.

[0037] Preferably, it also includes:

[0038] The matching information management module includes successful matching information and failed matching information, and the generated successful matching information and failed matching information are classified and stored.

[0039] Preferably, in the matching information management module, the specific classification storage method is:

[0040] Each time classification is performed, the matching information generated on the previous day is obtained, and the matching information is divided into successful matching information and failed matching information according to the matching results. The successful matching information and failed matching information are stored in the fourth quadrant sector area in sequence according to the order of time series, that is, the fourth quadrant sector area is divided into storing successful matching information and storing failed matching information.

[0041] The present invention provides an information record management system based on face recognition technology, which has the following beneficial effects compared with the prior art:

[0042] (1) The present invention provides a structured face information entry and storage module. By refining the division and storage of face information, the system can efficiently process and manage the information records of face recognition technology. By setting the acquisition and analysis of all matching information within a preset time T, the system can timely adjust the storage location and access rights of face information, thereby improving the efficiency and security of the entire authorization process. By classifying face information in detail (high-frequency, medium-frequency and low-frequency face information) and storing the data in different quadrants according to the recognition frequency, it is convenient to reasonably manage the information records of face recognition technology.

[0043] (2) The present invention can optimize the recognition process and improve the response speed and accuracy of the system through a fast recognition module. The module first compares the face information in the first quadrant. If a match is not successful, it then compares with other quadrants in turn until matching information is found or confirmation fails. This hierarchical and priority comparison strategy significantly improves the speed and efficiency of recognition, thereby reducing the overall processing time.

[0044] (3) The present invention classifies and stores successful matching and failed matching information through a matching information management module. The realization of this function makes information tracking and management more efficient. In addition, this classification storage method can also optimize the data storage structure, improve the speed and accuracy of data retrieval, and thus improve the operating efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The present invention will be further described below in conjunction with the accompanying drawings.

[0046] Figure 1 This is a system framework diagram of the information record management system based on face recognition technology of the present invention;

[0047] Figure 2 This is a storage method diagram of the information record management system based on face recognition technology of the present invention. DETAILED DESCRIPTION

[0048] 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.

[0049] Embodiment 1

[0050] See also Figure 1-Figure 2, the present invention provides an information record management system based on face recognition technology, including a face information entry module, a face information storage module, a face information division module, a storage division module, a face information adjustment module, a fast recognition module and a matching information management module;

[0051] The face information input module is used to collect and input new face information, record the new face information as newly added face information, and send it to the subsequent face information storage module;

[0052] It should be noted that the newly added facial information represents the addition of a new member. When authorization through facial recognition is required, the new member can be authorized through facial recognition;

[0053] A face information storage module is used to store authorized face information and newly added face information, and mark them as Rj, where j=1, 2, ..., m, m represents the total face information, and j represents the jth face recognition information in the total face information;

[0054] The face information classification module is used to obtain all matching information within a preset time T, where the matching information includes successful matching information and failed matching information, and determine high-frequency face information, medium-frequency face information and low-frequency face information by analyzing the matching information. The successful matching information includes recognition information and face information Rj, and the failed matching information only includes recognition information. The preset time T is a time period of n days from the current moment, where n=1, 2, ..., 30;

[0055] It should be noted that, in the face recognition process, it is necessary to first collect face images or face features of the entrant, and judge the collected face images or features with the face information pre-stored in the system through a set method (face recognition method based on algebraic features and face recognition method based on template matching, etc.) to judge whether the entrant can be authorized, that is, the matching information includes successful matching information and failed matching information. The successful matching information includes the unique match between the identification information (face image or face feature) collected during the face recognition process and the face information (pre-stored face information), and the failed matching information only includes the identification information collected during the face recognition process;

[0056] The specific method for determining high-frequency face information, medium-frequency face information and low-frequency face information is as follows:

[0057] S1: Obtain the matching information of each day in the previous n days, then determine the successful matching information from the matching information, and then obtain the face information Rj from all the successful matching information;

[0058] S2: Count the authorization records of facial information Rj every day in n days, obtain the number of days x that each Rj appears in n days, and mark it as R j-x , where x = 1, 2, ..., n, when R j-x / n>Q1, it is marked as high-frequency face information, when Q2 <R j-x / n≤Q1, it is marked as medium-frequency face information. j-x / n≤Q2 is marked as low-frequency face information, where Q1 and Q2 are preset values, and the specific parameters are set by the staff;

[0059] For example, if facial information R1 has authorization records for 19 days within 30 days, the ratio 19 / 30 is compared with Q1 and Q2 to determine the corresponding high-frequency facial information, medium-frequency facial information, and low-frequency facial information;

[0060] A storage division module is used to reasonably store high-frequency face information, medium-frequency face information and low-frequency face information;

[0061] The specific storage method is:

[0062] P1: Construct a plane rectangular coordinate system, that is, there are the first quadrant, the second quadrant, the third quadrant and the fourth quadrant. Then, with the origin as the center, construct the circle equation r1²=x²+y² to draw a circle, marked as circle 1. Circle 1 is combined with the horizontal and vertical coordinates to divide circle 1 into four areas, where r1 is the preset value. The specific parameters are set by the staff according to the size of the storage capacity, such as Figure 2 As shown:

[0063] That is, the first quadrant area, the second quadrant area, the third quadrant area and the fourth quadrant area;

[0064] P2: Store high-frequency face information in the first quadrant, medium-frequency face information in the second quadrant, low-frequency face information in the third quadrant, and newly added face information in the fourth quadrant;

[0065] A face information adjustment module, used to further adjust the face information in different quadrants by constructing circle 2;

[0066] The specific adjustment methods are:

[0067] AS1: Then, with the origin as the center, construct the circle equation r2²=x²+y² again to draw a circle, marked as circle 2, where r2>r1, r2 is the preset value, and the specific parameters are set by the staff according to the size of the storage capacity. Combined with circle 1, Figure 2 As shown;

[0068] Divided into the first quadrant sector, the second quadrant sector, the third quadrant sector and the fourth quadrant sector:

[0069] AS2: In the high-frequency face information in the first quadrant, randomly select a high-frequency face information as the target information Rj, and obtain the number of days R that the target information Rj appears in n days. j-x , and in R j-x The number of authorized records per day in a day Hc, 1≤c≤R j-x , through the formula , and get R j-x The average number of authorization records in a day is Hp, and then the formula is used , get the discrete value Wb;

[0070] AS3: Compare the discrete value Wb with the preset value Q3:

[0071] When Wb>Q3, Hc is sorted in descending order of |Hc-Hp|, and the first Hc value is deleted each time, and the discrete value Wb of the remaining Hc is recalculated until Wb≤Q3, and the number g1 of deleted Hc is obtained, and g1 is judged. When g1>gs, the average value of g1 deleted Hc is calculated as the evaluation value Pj of the target information Rj;

[0072] When g1≤gs, the average value of the remaining Hc is calculated as the evaluation value Pj of the target information Rj, where gs is a preset value, which is specifically set by professionals;

[0073] When Wb≤Q3, the average value Hp is used as the evaluation value Pj of the target information Rj;

[0074] AS4: Repeat steps 2 to 3 to determine that all high-frequency face information is used as the evaluation value Pj of the target information;

[0075] AS5: The high-frequency face information with an evaluation value Pj greater than the preset threshold Q4 is retained in the first quadrant, and the high-frequency face information with an evaluation value Pj less than or equal to the preset threshold Q4 is transferred to the first quadrant fan-shaped area;

[0076] Similarly, the medium-frequency face information and the low-frequency face information are transferred to the corresponding second quadrant sector area or the third quadrant sector area respectively in the same manner (steps AS2 to AS5);

[0077] It should be noted that the authorization record mentioned above indicates that the identification information collected during the face recognition process successfully matches a certain face information in the face information, that is, one authorization record represents the successfully matched face information Rj;

[0078] The above provides a structured face information entry and storage module. By refining the division and storage of face information, the system can efficiently process and manage the information records of face recognition technology. By setting the acquisition and analysis of all matching information within a preset time T, the system can timely adjust the storage location and access rights of face information, thereby improving the efficiency and security of the entire authorization process. By classifying face information in detail (high-frequency, medium-frequency and low-frequency face information) and storing data in different quadrants according to recognition frequency, it is convenient to reasonably manage the information records of face recognition technology.

[0079] Embodiment 2

[0080] In the specific implementation process, this embodiment is based on the first embodiment, and the difference from the first embodiment is that this embodiment is for the fast identification module:

[0081] A fast identification module is used to quickly compare and identify the collected identification information. The identification method is to compare and judge the collected identification information with the face information in the first quadrant first. If the comparison is unsuccessful, then compare and judge with the fan-shaped area of ​​the first quadrant;

[0082] When the comparison and judgment of the first quadrant sector area is not successful, the comparison and judgment is then carried out with the second quadrant area. If the comparison and judgment is not successful, the comparison and judgment is then carried out with the second quadrant sector area;

[0083] And so on, until the fourth quadrant sector is matched, if no match is successful, it is judged that the recognition fails, and a failed match information is generated. When the above situation is matched successfully, a successful match information is generated;

[0084] The fast recognition module can optimize the recognition process and improve the system's response speed and accuracy. The module first compares the face information in the first quadrant. If no match is found, it will then compare with other quadrants in turn until a match is found or confirmation fails. This hierarchical and priority comparison strategy significantly improves the speed and efficiency of recognition.

[0085] Embodiment 3

[0086] In the specific implementation process, this embodiment is based on the first and second embodiments, and the difference from the first and second embodiments is that this embodiment also includes:

[0087] A matching information management module, the matching information includes successful matching information and failed matching information, and the generated successful matching information and failed matching information are classified and stored;

[0088] The specific classification storage method is:

[0089] Each time classification is performed, the matching information generated on the previous day is obtained, and the matching information is divided into successful matching information and failed matching information according to the matching results, and the successful matching information and failed matching information are sequentially stored in the fourth quadrant sector area in the order of time series, that is, the fourth quadrant sector area is divided into storing successful matching information and storing failed matching information;

[0090] Embodiment 4

[0091] The specific implementation process of this embodiment includes all the implementation processes of the above three groups of embodiments.

[0092] Some of the data in the above formulas are numerically calculated by removing their dimensions. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0093] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The information record management system based on face recognition technology is characterized by: include: The face information input module is used to collect and input new face information, record the new face information as newly added face information, and send it to the subsequent face information storage module; A face information storage module is used to store authorized face information and newly added face information, and mark them as Rj, where j=1, 2, ..., m, m represents the total face information, and j represents the jth face recognition information in the total face information; The face information classification module is used to obtain all matching information within a preset time T, including successful matching information and failed matching information, and determine high-frequency face information, medium-frequency face information and low-frequency face information by analyzing the matching information. The preset time T is a time period of n days from the current moment, where n=1, 2, ..., 30; A storage division module is used to reasonably store high-frequency face information, medium-frequency face information and low-frequency face information by constructing a plane rectangular coordinate system and drawing circles to determine different quadrants; A face information adjustment module, used to further adjust the storage of high-frequency face information, medium-frequency face information and low-frequency face information in different quadrants of face information by constructing circle 2; A fast recognition module is used to quickly compare and identify the collected recognition information with the face information to determine whether a successful match or a failed match is generated; In the face information adjustment module, the specific adjustment method is as follows: AS1: Then, with the origin as the center, construct the circle equation r2²=x²+y² again to draw a circle, marked as circle 2, where r2>r1, and r2 is the preset value; Divided into a first quadrant sector, a second quadrant sector, a third quadrant sector and a fourth quadrant sector; AS2: In the high-frequency face information in the first quadrant, randomly select a high-frequency face information as the target information Rj, and obtain the number of days R that the target information Rj appears in n days. j-x , and in R j-x The number of authorized records per day in a day Hc, 1≤c≤R j-x , through the formula , and get R j-x The average number of authorization records in a day is Hp, and then the formula is used , get the discrete value Wb; AS3: Compare the discrete value Wb with the preset value Q3 to determine the evaluation value Pj of the target information Rj: AS4: Repeat steps 2 to 3 to determine that all high-frequency face information is used as the evaluation value Pj of the target information; AS5: The high-frequency face information with an evaluation value Pj greater than the preset threshold Q4 is retained in the first quadrant, and the high-frequency face information with an evaluation value Pj less than or equal to the preset threshold Q4 is transferred to the first quadrant fan-shaped area; The intermediate frequency face information or the low frequency face information is synchronously transferred to the corresponding second quadrant sector area or the third quadrant sector area according to the same principle.

2. The information record management system based on face recognition technology according to claim 1 is characterized in that: In the face information division module, the successful matching information includes the identification information and the face information Rj, and the failed matching information only includes the identification information.

3. The information record management system based on face recognition technology according to claim 2 is characterized in that: The specific method for determining high-frequency face information, medium-frequency face information and low-frequency face information is as follows: S1: Obtain the matching information of each day in the previous n days, then determine the successful matching information from the matching information, and then obtain the face information Rj from all the successful matching information; S2: Count the authorization records of facial information Rj every day in n days, obtain the number of days x that each Rj appears in n days, and mark it as R j-x , where x = 1, 2, ..., n, when R j-x / n>Q1, it is marked as high-frequency face information, when Q2 <R j-x / n≤Q1, it is marked as medium-frequency face information. j-x / n≤Q2 is marked as low-frequency face information, where Q1 and Q2 are preset values.

4. The information record management system based on face recognition technology according to claim 1 is characterized in that: In the storage partition module, the specific storage method is: P1: Construct a plane rectangular coordinate system, that is, there are the first quadrant, the second quadrant, the third quadrant and the fourth quadrant. Then, with the origin as the center, construct the circle equation r1²=x²+y² to draw a circle, marked as circle 1. Circle 1 is combined with the horizontal and vertical coordinates to divide circle 1 into four areas, where r1 is the preset value: That is, the first quadrant area, the second quadrant area, the third quadrant area and the fourth quadrant area; P2: Store high-frequency face information in the first quadrant, medium-frequency face information in the second quadrant, low-frequency face information in the third quadrant, and newly added face information in the fourth quadrant.

5. The information record management system based on face recognition technology according to claim 1 is characterized in that: In step AS3, it also includes: When Wb>Q3, Hc is sorted in descending order of |Hc-Hp|, and the first Hc value is deleted each time, and the discrete value Wb of the remaining Hc is recalculated until Wb≤Q3, and the number g1 of deleted Hc is obtained, and g1 is judged. When g1>gs, the average value of g1 deleted Hc is calculated as the evaluation value Pj of the target information Rj; When g1≤gs, the average value of the remaining Hc is calculated as the evaluation value Pj of the target information Rj, where gs is a preset value, which is specifically set by professionals; When Wb≤Q3, the average value Hp is used as the evaluation value Pj of the target information Rj.

6. The information record management system based on face recognition technology according to claim 1, characterized in that: In the quick identification module, the identification method is: The collected recognition information is first compared with the face information in the first quadrant. If the comparison fails, it is then compared with the fan-shaped area in the first quadrant. When the comparison and judgment of the first quadrant sector area is not successful, the comparison and judgment is then carried out with the second quadrant area. If the comparison and judgment is not successful, the comparison and judgment is then carried out with the second quadrant sector area; And so on, until the fourth quadrant sector is matched, if no match is successful, it is determined that the recognition fails and a failed match information is generated. If the above situation is matched successfully, a successful match information is generated.

7. The information record management system based on face recognition technology according to claim 1 is characterized in that: Also includes: The matching information management module includes successful matching information and failed matching information, and the generated successful matching information and failed matching information are classified and stored.

8. The information record management system based on face recognition technology according to claim 7 is characterized in that: In the matching information management module, the specific classification storage method is as follows: Each time classification is performed, the matching information generated on the previous day is obtained, and the matching information is divided into successful matching information and failed matching information according to the matching results. The successful matching information and failed matching information are stored in the fourth quadrant sector area in sequence according to the order of time series, that is, the fourth quadrant sector area is divided into storing successful matching information and storing failed matching information.

Images