Security attendance linkage system based on video surveillance cloud platform

Abstract

The invention discloses a security attendance linkage system based on a video surveillance cloud platform, comprising an external unit, a video analysis unit, a storage self-decision unit, a forward monitoring unit, a check-in monitoring unit, a data interception unit, a preliminary calculation unit and an attendance unit; the forward monitoring unit automatically obtains a clock-in video when a user clocks in; then the check-in video is obtained by using the check-in monitoring unit, and the check-in monitoring unit keeps the installation place secret during the acquisition to obtain the check-in video; then the forward monitoring unit transmits the clock-in video to the video analysis unit, and the check-in monitoring unit is used to transmit the check-in video to the video analysis unit; the video analysis unit is used to perform preliminary interception and statistics on the clock-in video and the check-in video, obtain a real face group and a false face, and automatically obtain a person with deceptive behavior and a person who has not clocked in; the application is simple, effective, and easy to use.

Description

Technical Field

[0001] The present invention belongs to the field of linkage attendance monitoring, relates to video monitoring technology, and specifically is a security attendance linkage system based on a video monitoring cloud platform. Background Art

[0002] The patent with the publication number CN109284983A discloses a 2.4G long-distance video capture attendance and safety reporting system, including a communication service center, a base station, a video monitoring system, a 2.4G long-distance attendance terminal, a router, a WeChat enterprise account server, a mobile phone, a 2.4G card reader, a video device and an electronic student ID card; first, the 2.4G long-distance attendance terminal automatically senses and reads the electronic student ID card, generates a student attendance record and uploads it to the communication service center, the communication service center parses and processes the attendance record and sends instructions to the video monitoring system to extract the video of 15 seconds before and after the student's attendance time; then it is pushed to the student's parents through the WeChat enterprise account. It can effectively improve school safety management and promote campus informatization construction; parents can view the video of their children entering and leaving the school in real time through the WeChat enterprise account on their mobile phones, which makes parents feel more at ease; for schools, attendance records provide a strong basis for the school to clarify the handling of responsible accidents.

[0003] The patent with publication number CN205582024U also provides a student attendance machine for video monitoring, including a real-time video acquisition module, a central processing unit, a card reader, a clock module, a display module and a communication module; the output ends of the real-time video acquisition module and the card reader are connected to the central processing unit; the output end of the central processing unit is connected to the communication module, and the communication module is connected to an external background server; the central processing unit is also connected to the clock module and the display module respectively. The real-time video in front of the attendance machine is collected, and the attendance machine is provided with a communication module, which can communicate with the external background server in real time.

[0004] However, the above two patents provide a video recognition clock-in machine and also provide the existing technology of related video capture attendance. However, for both, firstly, video recognition can be deceived in a certain way, and there is no linkage monitoring of the actual video. When automatic recognition and clock-in are achieved, it can self-service detect whether there are some attendance situations of the corresponding personnel that are inconsistent with the actual work records, and make statistics on the actual attendance of the personnel; based on this, a solution is provided. Summary of the invention

[0005] The purpose of the present invention is to provide a security attendance linkage system based on a video surveillance cloud platform.

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

[0007] The security attendance linkage system based on the video surveillance cloud platform includes an external unit, a video analysis unit, a storage self-decision unit, a positive monitoring unit, a negative monitoring unit, a data interception unit, a preliminary calculation unit and an attendance unit;

[0008] The forward monitoring unit is a video clocking device installed at a designated location in the target area, which is used to automatically obtain the clocking-in video when the user clocks in; the reverse monitoring unit is a monitoring camera device installed in the target area, which is used to obtain the real-time video of the target area and mark the real-time video as the reverse video;

[0009] The forward monitoring unit is used to transmit the punch-in video to the video parsing unit, and the reverse monitoring unit is used to transmit the reverse video to the video parsing unit;

[0010] The video parsing unit stores the standard number of people on duty in the target area in advance. The standard number of people on duty refers to the number of all people on duty under normal circumstances. The video parsing unit is used to perform preliminary interception and statistics on the clock-in video and the verification video to obtain the real face group and the false face group.

[0011] The video analysis unit is used to transmit the fake face and the real face group to the data interception unit, and the data interception unit stores the regular working hours, which refers to the actual working hours of the personnel in the target area; the data interception unit receives the real face group transmitted by the video analysis unit and performs follow-up monitoring on it. The follow-up monitoring behavior is performed once every day after get off work to obtain the disappearance time and disappearance frequency of all the real faces;

[0012] The data interception unit is used to transmit the disappearance time and disappearance frequency of the false face and the real face to the preliminary calculation unit, and the preliminary calculation unit is used to perform preliminary calculation processing on the disappearance time and disappearance frequency of the real face to obtain the negative face, the regular face and the above-standard face.

[0013] Furthermore, the specific method of preliminary interception statistics is as follows:

[0014] Step 1: Get the real-time number of people in the clock-in video and the standard number of people on duty;

[0015] Step 2: When the real-time number of people is consistent with the standard number of people on duty, no processing is done;

[0016] Step 3: Otherwise, a detailed inspection signal is generated; at this time, all facial information in the punch-in video is automatically obtained, the facial information is matched with the standard facial information, the non-existent standard facial information is marked as an absent face, and the non-existent facial information in the standard facial information is marked as a redundant face;

[0017] Step 4: After the start of working hours at T1 time, where T1 is the preset time, all facial information in the verification video is automatically acquired again and marked as the verification facial information group;

[0018] Step 5: Obtain all facial information in the check-in video, and mark the facial information as a check-in face group; compare the check-in face group with the verification face information group, and mark the inconsistent check-in face groups as false faces;

[0019] Step 6: Mark the faces in the check-in face group that are consistent with the verified face information as the actual face group;

[0020] Step 7: Get the fake face and real face groups.

[0021] Furthermore, the specific method of the follow-up monitoring behavior is:

[0022] S1: Get all the actual face groups;

[0023] S2: Obtain the disappearance time and disappearance count of all the real faces in the real face group; the disappearance time refers to the time when the real face is not detected; the disappearance count refers to the time when a single face disappears for more than T1 time and does not return to the target area, which is marked as one time;

[0024] S3: Get the disappearance time and disappearance times of all the real faces.

[0025] Furthermore, the time T1 in step S2 is specifically a value preset by the management personnel.

[0026] Furthermore, the specific value of the T1 time in step S2 is defined in the following manner:

[0027] S201: Collect all office workers in the target area;

[0028] S202: Select any office worker and obtain the time during the day when the employee's real face is not obtained without taking leave for business trip;

[0029] S203: Mark all times as disappearance times, and all disappearance times constitute a disappearance time group Pi, i=1, ..., n;

[0030] S204: Obtain the mean of the disappearance time group, then calculate the absolute value of the difference between all disappearance times and the mean, sum all absolute values, and obtain the difference sum value;

[0031] S205: When the difference and value are greater than X1, the disappearance time is automatically deleted, otherwise no processing is performed; in order from large to small absolute values, the corresponding disappearance time with the largest absolute value is first deleted;

[0032] S206: Repeat steps S204-S206 for the remaining disappearance time until the difference sum is less than or equal to X1; X1 is a preset value;

[0033] S207: At this time, the mean of the disappearance time after deletion is obtained, and the mean is marked as a reasonable time;

[0034] S208: Obtain the reasonable working time of all office workers, calculate the average, obtain the median between the maximum and average of the reasonable time, and mark the median as the upper limit; obtain the median between the minimum and average of the reasonable time, and mark the median as the lower limit;

[0035] S209: The value of T1 is within the range from the lower limit to the upper limit of the target.

[0036] Furthermore, the specific steps of the initial calculation are as follows:

[0037] SS1: Calculate the wear attribute value according to the formula. The specific calculation formula is:

[0038] Disappearance attribute value = 0.41*disappearance time + 0.59*disappearance times;

[0039] Here, 0.41 and 0.59 are the weights preset by managers;

[0040] SS2: When the attrition attribute value is greater than X3, it is considered a negative face;

[0041] When X2≤attack attribute value≤X3, it is considered as a regular face;

[0042] The remaining ones are marked as above-standard faces.

[0043] Furthermore, the preliminary calculation unit is used to transmit the false faces, negative faces, regular faces and above-standard faces to the video analysis unit with the help of the attendance unit, and the video analysis unit is used to transmit the false faces, negative faces, regular faces and above-standard faces with timestamps to the storage self-decision-making unit.

[0044] Furthermore, the storage unit receives the negative faces, regular faces and above-standard faces with timestamps transmitted by the video parsing unit, and performs statistical analysis on its own, specifically:

[0045] Those who have been marked as negative faces for more than 5 days in a single month are redefined as recommended personnel;

[0046] Those whose faces are marked as above the standard on more than 85% of the days in a single month are redefined as recommended personnel;

[0047] Mark the person corresponding to the fake face as untrustworthy;

[0048] The storage self-decision-making unit is used to transmit the non-trusted personnel, the recommended personnel for poor performance and the recommended personnel for excellent performance to the external unit for real-time display.

[0049] Beneficial effects of the present invention:

[0050] The present invention automatically obtains the punch-in video through the forward monitoring unit when the user punches in; then the real-time video of the target area is obtained by the reverse monitoring unit, and the reverse monitoring unit keeps the installation location secret during the acquisition, and obtains the reverse video; then the forward monitoring unit transmits the punch-in video to the video analysis unit, and the reverse monitoring unit is used to transmit the reverse video to the video analysis unit;

[0051] The video analysis unit is used to perform preliminary interception and statistics on the clock-in video and the verification video, obtain the real face group and the false face group, and automatically obtain the people who deceived the clock-in and the people who did not clock in; this application is simple, effective, and easy to use. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0053] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0054] like Figure 1 As shown, the security attendance linkage system based on the video surveillance cloud platform provided by the present application includes an external unit, a video analysis unit, a storage self-decision unit, a forward monitoring unit, a negative monitoring unit, a data interception unit, a preliminary calculation unit and an attendance unit;

[0055] The forward monitoring unit is a video clocking device installed at a designated location in the target area, which is used to automatically obtain the clocking-in video when the user clocks in; the reverse monitoring unit is a monitoring camera device installed in the target area, which is used to obtain the real-time video of the target area and mark the real-time video as the reverse video;

[0056] The forward monitoring unit is used to transmit the punch-in video to the video parsing unit, and the reverse monitoring unit is used to transmit the reverse video to the video parsing unit;

[0057] The video parsing unit stores the standard number of people on duty in the target area in advance. The standard number of people on duty refers to the number of all personnel on duty under normal circumstances. The video parsing unit is used to perform preliminary interception and statistics on the clock-in video and the direction verification video, specifically:

[0058] Step 1: Get the real-time number of people in the clock-in video and the standard number of people on duty;

[0059] Step 2: When the real-time number of people is consistent with the standard number of people on duty, no processing is done;

[0060] Step 3: Otherwise, a detailed inspection signal is generated; at this time, all facial information in the punch-in video is automatically obtained, the facial information is matched with the standard facial information, the non-existent standard facial information is marked as an absent face, and the non-existent facial information in the standard facial information is marked as a redundant face;

[0061] Step 4: After the start of working hours at T1 time, where T1 is the preset time, all facial information in the verification video is automatically acquired again and marked as the verification facial information group;

[0062] Step 5: Obtain all facial information in the check-in video, and mark the facial information as a check-in face group; compare the check-in face group with the verification face information group, and mark the inconsistent check-in face groups as false faces;

[0063] Step 6: Mark the faces in the check-in face group that are consistent with the verified face information as the actual face group;

[0064] Step 7: Obtain fake face and real face group;

[0065] The video analysis unit is used to transmit the fake face and the real face group to the data interception unit. The data interception unit stores the regular working hours, which refers to the actual working hours of the personnel in the target area. The data interception unit receives the real face group transmitted by the video analysis unit and performs follow-up monitoring on it. The follow-up monitoring behavior is performed once every day after get off work, specifically:

[0066] S1: Get all the actual face groups;

[0067] S2: Obtain the disappearance time and disappearance count of all the real faces in the real face group; the disappearance time refers to the time when the real face is not detected; the disappearance count refers to the time when a single face disappears for more than T1 time and does not return to the target area, which is marked as one time;

[0068] The T1 time here is a preset value by the administrator, and the value range can also be defined in the following ways:

[0069] S201: Collect all office workers in the target area;

[0070] S202: Select any office worker and obtain the time during the day when the employee's real face is not obtained without taking leave for business trip;

[0071] S203: Mark all times as disappearance times, and all disappearance times constitute a disappearance time group Pi, i=1, ..., n;

[0072] S204: Obtain the mean of the disappearance time group, then calculate the absolute value of the difference between all disappearance times and the mean, sum all absolute values, and obtain the difference sum value;

[0073] S205: When the difference and value are greater than X1, the disappearance time is automatically deleted, otherwise no processing is performed; in order from large to small absolute values, the corresponding disappearance time with the largest absolute value is first deleted;

[0074] S206: Repeat steps S204-S206 for the remaining disappearance time until the difference sum is less than or equal to X1; X1 is a preset value;

[0075] S207: At this time, the mean of the disappearance time after deletion is obtained, and the mean is marked as a reasonable time;

[0076] S208: Obtain the reasonable working time of all office workers, calculate the average, obtain the median between the maximum and average of the reasonable time, and mark the median as the upper limit; obtain the median between the minimum and average of the reasonable time, and mark the median as the lower limit;

[0077] S209: The value of T1 is within the range from the lower limit to the upper limit of the target;

[0078] S3: Get the disappearance time and disappearance count of all real faces;

[0079] The data interception unit is used to transmit the disappearance time and disappearance frequency of the fake face and the real face to the preliminary calculation unit, and the preliminary calculation unit is used to perform preliminary calculation processing on the disappearance time and disappearance frequency of the real face. The specific steps of the preliminary calculation processing are:

[0080] SS1: Calculate the wear attribute value according to the formula. The specific calculation formula is:

[0081] Disappearance attribute value = 0.41*disappearance time + 0.59*disappearance times;

[0082] Here, 0.41 and 0.59 are the weights preset by managers;

[0083] SS2: When the attrition attribute value is greater than X3, it is considered a negative face;

[0084] When X2≤attack attribute value≤X3, it is considered as a regular face;

[0085] Mark the remaining faces as above-standard faces;

[0086] The preliminary calculation unit is used to transmit the false faces, negative faces, regular faces and above-standard faces to the video analysis unit with the help of the attendance unit, and the video analysis unit is used to transmit the false faces, negative faces, regular faces and above-standard faces with timestamps to the storage self-decision unit;

[0087] The storage unit receives the negative faces, regular faces and above-standard faces with timestamps transmitted by the video parsing unit, and performs statistical analysis on its own, specifically:

[0088] Those who have been marked as negative faces for more than 5 days in a single month are redefined as recommended personnel;

[0089] Those whose faces are marked as above the standard on more than 85% of the days in a single month are redefined as recommended personnel;

[0090] Mark the person corresponding to the fake face as untrustworthy;

[0091] The storage self-decision-making unit is used to transmit the non-trusted personnel, the recommended personnel for poor performance and the recommended personnel for excellent performance to the external unit for real-time display.

[0092] The present invention automatically obtains the punch-in video through the forward monitoring unit when the user punches in; then the real-time video of the target area is obtained by the reverse monitoring unit, and the reverse monitoring unit keeps the installation location secret during the acquisition, and obtains the reverse video; then the forward monitoring unit transmits the punch-in video to the video analysis unit, and the reverse monitoring unit is used to transmit the reverse video to the video analysis unit;

[0093] The video analysis unit is used to perform preliminary interception and statistics on the clock-in video and the verification video, obtain the real face group and the false face group, and automatically obtain the people who deceived the clock-in and the people who did not clock in; this application is simple, effective, and easy to use.

[0094] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. The security attendance linkage system based on the video surveillance cloud platform is characterized by: include: Positive monitoring unit: It is a video clocking device installed at a designated location in the target area, used to automatically obtain the clocking-in video of the user when clocking in and transmit it to the video analysis unit; Nuclear direction monitoring unit: It is a monitoring camera installed in the target area, used to obtain real-time video of the target area, mark the real-time video as nuclear direction video and transmit it to the video analysis unit; The video analysis unit is used to perform preliminary interception and statistics on the check-in video and the verification video, obtain the real face group and the false face group and transmit them to the data interception unit; The data interception unit receives the real face group transmitted by the video analysis unit and performs follow-up monitoring on it. The follow-up monitoring is performed once every day after get off work to obtain all single disappearance times and disappearance times of all the real faces in one day; Preliminary calculation unit: It performs preliminary calculation on all single disappearance times and disappearance times of the false faces and real faces transmitted by the data interception unit within one day, and obtains negative faces, regular faces and above-standard faces; The video analysis unit pre-stores the standard number of people on duty in the target area. The standard number of people on duty refers to the number of all personnel who should be on duty under normal circumstances. The specific method of preliminary interception and statistics is as follows: Step 1: Get the real-time number of people in the clock-in video and the standard number of people on duty; Step 2: When the real-time number of people is consistent with the standard number of people on duty, no processing is done; Step 3: Otherwise, a detailed inspection signal is generated; at this time, all facial information in the punch-in video is automatically obtained, the facial information is matched with the standard facial information, the non-existent standard facial information is marked as an absent face, and the non-existent facial information in the standard facial information is marked as a redundant face; Step 4: After the working hours start at time T2, where T2 is the preset time, all the face information in the video is automatically acquired again and marked as the face of the face information group; Step 5: Obtain all facial information in the check-in video, and mark the facial information as faces in the check-in face group; compare the check-in face group with the verification face information group, and mark the faces in the inconsistent check-in face group as false faces; Step 6: Mark the faces in the check-in face group that are consistent with the verified face information group as faces in the actual face group; Step 7: Obtain fake face and real face group; The data interception unit stores regular working hours, which refer to the actual working hours of personnel in the target area. The specific methods of follow-up monitoring behavior are: S1: Obtain the actual face group; S2: Obtain all single disappearance times and disappearance times of all the real faces in the real face group within one day; the disappearance time refers to the time when the real face is not detected once; the disappearance time of a single face exceeds T1 time and does not return to the target area is recorded as one disappearance; The specific value of the T1 time in step S2 is defined in the following manner: S201: Collect all office workers in the target area; S202: Select any office worker and obtain the time during the day when the employee's real face is not obtained without taking leave for business trip; S203: Mark each time in a day when the actual face is not obtained as a disappearance time, and all the disappearance times in a day constitute a disappearance time group P={P1, P2..., Pi}, i=1,...,n, and n is the number of times the actual face is not obtained in a day; S204: Obtain the mean of the disappearance time group, then calculate the absolute value of the difference between all disappearance times and the mean, sum all absolute values, and obtain the difference sum value; S205: When the difference and value are greater than X1, the disappearance time is automatically deleted, otherwise no processing is performed; in order from large to small absolute values, the corresponding disappearance time with the largest absolute value is first deleted; S206: Repeat steps S204-S205 for the remaining disappearance time until the difference sum is less than or equal to X1; X1 is a preset value; S207: At this time, the mean of the disappearance time group after deletion is obtained, and the mean is marked as a reasonable time; S208: Obtain the reasonable time of all office workers, calculate the average, obtain the median between the maximum value of the reasonable time and the average, and mark the median as the upper limit of the target; The median between the minimum and mean of the reasonable time is marked as the lower limit of the target; S209: The value of T1 satisfies the range from the lower limit to the upper limit of the target; The specific steps of initial calculation are as follows: SS1: Calculate the wear attribute value according to the formula. The specific calculation formula is: The disappearance attribute value = 0.41*the sum of all elements in the disappearance time group + 0.59*the disappearance times; Here, 0.41 and 0.59 are the weights preset by managers; SS2: When the attrition attribute value is greater than X3, it is considered a negative face; When X2≤attack attribute value≤X3, it is considered as a regular face; The remaining ones are marked as above-standard faces, and X2 and X3 are preset values.

2. The security attendance linkage system based on the video surveillance cloud platform according to claim 1 is characterized in that: The preliminary calculation unit is used to transmit the false faces, negative faces, regular faces and above-standard faces to the video analysis unit with the help of the attendance unit, and the video analysis unit is used to transmit the false faces, negative faces, regular faces and above-standard faces with timestamps to the storage self-decision-making unit.

3. The security attendance linkage system based on the video surveillance cloud platform according to claim 2 is characterized in that: The storage unit receives the negative faces, regular faces and above-standard faces with timestamps transmitted by the video parsing unit, and performs statistical analysis on its own, specifically: Those who have been marked as negative faces for more than 5 days in a single month are redefined as recommended personnel; Those whose faces are marked as above the standard on more than 85% of the days in a single month are redefined as recommended personnel; Mark the person corresponding to the fake face as untrustworthy; The storage self-decision-making unit is used to transmit the non-trusted personnel, the recommended personnel for poor performance and the recommended personnel for excellent performance to the external unit for real-time display.

Images