Intelligent track supervision system based on big data

Abstract

The invention discloses an intelligent track supervision system based on big data, and relates to the technical field of behavior detection, and the system comprises a management and control center which is in communication connection with an access record generation module, an analysis module, a personnel track prediction module and an abnormal behavior judgment module. The in-out record generation module is used for collecting photos and action tracks of in-out personnel and generating corresponding personnel in-out records; the analysis module is used for analyzing personnel access records and sending an analysis result to the personnel trajectory prediction module; the personnel track prediction module is used for establishing a normal behavior model of the in-out personnel according to the analysis result of the personnel in-out record; and the abnormal behavior judgment module is used for judging whether the in-out personnel have abnormal behaviors according to the in-out personnel normal behavior model.

Description

Technical Field

[0001] The present invention relates to the technical field of behavior detection, and in particular to a trajectory intelligent supervision system based on big data. Background Art

[0002] As the scale of the domestic video surveillance market continues to expand, the accuracy of video detection systems is becoming increasingly higher. However, the current video detection systems on the market have single functions and lack intelligent prediction of personnel behavior and real-time warning. They can only collect real-time video data through cameras and require monitoring personnel to determine whether there are abnormal events. This is inefficient and labor-intensive. The video data collected by conventional cameras is prone to poor video quality and difficulty in distinguishing due to various environmental factors. For example, sunlight and leaves can easily block the objects to be collected. For this reason, a trajectory intelligent supervision system based on big data is provided. Summary of the invention

[0003] In order to solve the above technical problems, the purpose of the present invention is to provide a trajectory intelligent supervision system based on big data; A trajectory intelligent supervision system based on big data, comprising a control center, wherein the control center is communicatively connected to an entry and exit record database, an entry and exit record generation module, an analysis module, a personnel trajectory prediction module, and an abnormal behavior judgment module; The database is used to store data from various modules, and the database also stores the identity information of all people entering and leaving; The entry and exit record generation module is used to manage the entry and exit of personnel through the checkpoint and collect the movement trajectory of the entry and exit personnel, and generate corresponding entry and exit records of personnel; The analysis module is used to analyze the personnel entry and exit records and send the analysis results to the personnel trajectory prediction module; The personnel trajectory prediction module is used to establish a normal behavior model of personnel entering and leaving according to the analysis results of personnel entry and exit records; The abnormal behavior judgment module is used to judge whether the person entering or leaving has abnormal behavior according to the normal behavior model of the person entering or leaving.

[0004] Furthermore, the process of managing the entry and exit of personnel through the checkpoint by the entry and exit record generation module includes: The entry and exit record generation module is equipped with an identity card scanning device, a gate, multiple cameras and sensors. People entering and exiting can scan their identity cards through the identity card scanning device, and the identity card scanning device reads the identity information in the identity card; The corresponding identity information in the identity information is searched in the database for comparison to determine whether the identity information is correct. If it is judged to be wrong, entry is denied and an alarm is issued. If it is correct, real-time photos of people entering and leaving are taken through the camera, and compared with the photos in the identity information for a second time. If the comparison is consistent, the gate is opened and a record of people entering and exiting is generated; if it is inconsistent, the gate is not opened and an alarm is issued.

[0005] Furthermore, the process of generating the personnel entry and exit records includes: The entry and exit record generation module opens the gate. After the sensor senses that a person is entering or leaving the gate, the camera automatically records the video data of the person passing through the gate, and then automatically generates a record of the person entering or leaving the gate and sends it to the database.

[0006] Furthermore, the analysis process of the personnel entry and exit records includes: Divide the video data of the entry and exit checkpoint into several image data, set numbers for them, and mark the entry and exit personnel in the image data based on the photos of the entry and exit personnel in the identity information of the entry and exit personnel; Based on the region of the image data where the in-and-out persons are marked, other regions in the image data are blurred, thereby generating image data that only includes the marked in-and-out persons; Generate background image data, and merge all blurred image data with the background image data to form new entry and exit video data; The video data of the entry and exit checkpoint in the personnel entry and exit record is replaced with the new video data of the entry and exit checkpoint, and a new personnel entry and exit record is obtained and sent to the database.

[0007] Furthermore, the process of generating background image data includes: Before blurring the image data, a number of image data are randomly selected from the image data and divided into a number of regions of equal size. A codebook is established for the pixels in each region, and each codebook consists of multiple code words. Set a codeword number threshold and compare it with the number of codewords in the pixel, and mark it as a background pixel or a target pixel based on the judgment result; A background pixel threshold is set and compared with the number of background pixels in the area. The pixels are marked as background pixels or target pixels based on the judgment results. The areas marked as background areas in all selected image data are integrated to form background image data that does not include people entering and leaving.

[0008] Furthermore, the personnel trajectory prediction module analyzes the personnel entry and exit records in the following process: The personnel trajectory prediction module obtains new personnel entry and exit records from the database, and sets the coordinate axis, initial position and end position according to the movement route of the personnel in and out of the new entry and exit video data in the personnel entry and exit records, and then obtains the movement speed according to the video time and the distance from the initial position to the end position; The head of the person entering or leaving is taken as the first feature point, the two hands are taken as the second and third feature points, and two vector sets are set according to the length of the trajectory of the lateral movement of the person entering or leaving. The average weight and distance weight of the vector coordinates from the first feature point to the second feature point or the third feature point in the vector set are obtained; Based on the personnel entry and exit records corresponding to the personnel entering and exiting, the moving speed, average weight and distance weight of all corresponding personnel entry and exit records are obtained, and then the corresponding normal moving speed, normal average weight and normal distance weight are obtained.

[0009] Furthermore, the process of establishing a normal behavior model for people entering and leaving includes: Select the maximum and minimum values ​​from all moving speeds, average weights and distance weights, set a mark, subtract the normal moving speed, normal average weight and normal distance weight from them, and obtain two moving speed range thresholds, two average weight range thresholds, and two distance weight range thresholds; According to the six range thresholds, the normal movement range of people entering and leaving is set, and then the normal behavior model of people entering and leaving corresponding to each person entering and leaving is established and uploaded to the database.

[0010] Furthermore, the process of establishing a normal behavior model for people entering and leaving includes: Select the maximum and minimum values ​​from all moving speeds, average weights and distance weights, set a mark, subtract the normal moving speed, normal average weight and normal distance weight from them, and obtain two moving speed range thresholds, two average weight range thresholds, and two distance weight range thresholds; According to the six range thresholds, the normal movement range of people entering and leaving is set, and then the normal behavior model of people entering and leaving corresponding to each person entering and leaving is established and uploaded to the database.

[0011] Furthermore, the process of the abnormal behavior judgment module judging the behavior of people entering and leaving includes: According to the latest personnel entry and exit records, the corresponding movement speed, average weight and distance weight are calculated, and then the corresponding movement model is established; Compare the movement model with the normal behavior model of the in-and-out personnel corresponding to the in-and-out personnel, and then determine whether its movement speed, average weight and distance weight are within the normal movement range of the in-and-out personnel; If the moving speed, average weight, and distance weight are all within the normal range of human motion, then the movement trajectory of the in-and-out personnel is judged to be normal; If two of the moving speed, average weight, and distance weight are within the normal range of human motion, the movement trajectory of the person entering or leaving is judged to be slightly abnormal, and a behavior abnormality label is set for their identity information. If the behavior abnormality label is set for the corresponding entry and exit records three times in a row, the movement trajectory of the person entering or leaving is judged to be abnormal; If only one or none of the moving speed, average weight and distance weight is within the normal range of human motion, the movement trajectory of the in-and-out personnel is judged to be abnormal.

[0012] Furthermore, if it is determined that the movement trajectory of the person entering or leaving is abnormal, the abnormal behavior judgment module sends a command to the control center, and the control center then suspends the authority of the person's identity card to enter the checkpoint; If the abnormal behavior judgment module determines that the movement trajectory of the in-and-out personnel is slightly abnormal, the personnel trajectory prediction module will update the corresponding personnel normal behavior model according to the corresponding personnel entry and exit records.

[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention analyzes multiple entry and exit records of people entering and leaving, and then analyzes the corresponding action habits of the people entering and leaving, and generates a corresponding normal behavior model. Based on the new entry and exit records of the people entering and leaving, a corresponding action model is generated and matched with the normal behavior model, so as to judge whether the current action of the people entering and leaving is abnormal. The normal behavior model will be updated in real time according to the latest entry and exit records of the people entering and leaving, thereby reducing the possibility of misjudgment and ensuring the accuracy and timeliness of the normal behavior model; 2. In the process of analyzing the entry and exit records of personnel, the entry and exit video data in the personnel entry and exit records are divided into several image data, and the image data is divided into several small areas, and then the pixel points of only background elements in the small areas are extracted, and image data that does not include the entry and exit personnel are drawn. The background image data obtained by this method can to a certain extent avoid environmental problems caused by lighting changes and cluttered backgrounds, thereby making the subsequent extraction of labeled target personnel areas more accurate. In addition, this method has low computational complexity, can dynamically adapt to various environments, and has good stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0015] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0016] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0017] like Figure 1 As shown, a trajectory intelligent supervision system based on big data includes a control center, and the control center is communicatively connected with an entry and exit record generation module, an analysis module, a personnel trajectory prediction module, and an abnormal behavior judgment module; It should be noted that the control center is also connected to a database, which is used to store data from various modules, and the database also stores the identity information of all people entering and leaving; The entry and exit record generation module is used to collect photos and movement trajectories of people entering and leaving, and generate corresponding entry and exit records of people, specifically including: The entry and exit record generation module is equipped with an identity card scanning device, a gate, multiple cameras and sensors. The person entering and exiting can scan the identity card through the identity card scanning device, and the identity card scanning device reads the identity information in the identity card. It should be noted that the identity information includes the person number, name, age and photo; The entry and exit record generation module searches for the corresponding identity information from the database according to the personnel number in the identity information for comparison. If the two are consistent, the camera takes a real-time photo of the person entering and exiting, and compares it with the photo in the identity information. If the comparison is consistent, the gate is opened; if not, the gate is not opened and an alarm is issued; if the identity information is inconsistent with the corresponding identity information in the database, the gate is not opened and an alarm is issued; Furthermore, the entry and exit record generation module opens the gate, and after the sensor senses that the person entering or leaving the gate passes through the gate, the camera automatically records the video data of the person entering or leaving the gate, and then the entry and exit record generation module automatically generates the person entry and exit record and sends it to the database; It should be noted that the personnel entry and exit records include the identity information of the personnel entering and exiting, the entry and exit time, and the video data of the personnel entering and exiting the checkpoint.

[0018] The analysis module is used to analyze the personnel entry and exit records and send the analysis results to the personnel trajectory prediction module, specifically including: The analysis module obtains the personnel entry and exit records from the database, and performs denoising on the video data of the personnel entering and exiting the checkpoints in the personnel entry and exit records; The video data of the entrance and exit is divided into several image data and numbered P1, P2, ..., P n, where n is an integer greater than 1. Based on the photos of the in-and-out persons in the identity information of the in-and-out persons, the in-and-out persons in the image data are marked. After the marking is completed, based on the area of ​​the image data where the marked in-and-out persons are located, other areas in the image data are blurred, thereby generating image data that only includes the marked in-and-out persons; It should be noted that before blurring the image data, N image data are randomly selected from the image data and divided into several regions of equal size. A codebook is established for the pixels in each region. Each codebook consists of multiple code words. The more code words a pixel's codebook includes, the more content the pixel contains. Set a codeword number threshold and compare it with the codeword number in the pixel point, and mark it as a background pixel or a target pixel according to the judgment result, where the background pixel is a pixel that does not include people entering or leaving, and the target pixel is a pixel that includes people entering or leaving; If the number of code words in a pixel is greater than or equal to the code word number threshold, the pixel is determined to be a target pixel; if the number of code words in a pixel is less than the code word number threshold, the pixel is determined to be a background pixel; Set the background pixel threshold. If the number of background pixels in the area is greater than or equal to the background pixel threshold, the marked area is the background area. If the number of background pixels in the area is less than the background pixel threshold, the marked area is the target area. Integrate all the areas marked as background areas in the N image data to form background image data that does not include people entering or leaving, and merge all the blurred image data with the background image data to form new video data of entering or leaving the checkpoint; In the above processing method, firstly, the video data of people entering and leaving the checkpoint is divided into several picture data, and several image data are randomly selected therefrom, and the selected image data is divided into several small areas, and the code point quantity of the pixel points in each small area is evaluated to determine whether it is a background area, and then the background image data is integrated and drawn. The background image data obtained by this method can avoid the environmental problems caused by lighting changes and cluttered backgrounds to a certain extent, so that the subsequent extraction of the labeled target personnel area is more accurate, and the method has low computational complexity, can dynamically adapt to various environments, and has good stability; The video data of the entry and exit checkpoint in the personnel entry and exit record is replaced with the new video data of the entry and exit checkpoint, and a new personnel entry and exit record is obtained and sent to the database.

[0019] The personnel trajectory prediction module is used to establish a normal behavior model of personnel entering and leaving according to the analysis results of personnel entry and exit records, which specifically includes: The personnel trajectory prediction module obtains new personnel entry and exit records from the database and numbers the personnel entry and exit records according to the numbers of the personnel entering and exiting in the personnel entry and exit records, for example, H 1，1 , H 2，1 , H 2,2 ,……,H i，j It should be noted that for any person entering or leaving the room, there are several entry and exit records, so the number H i，j Indicates the jth entry and exit record of the person numbered i, where i, j = 1, 2, 3, ...; Furthermore, based on the movement route of the person in and out of the new video data of the person in and out of the checkpoint in the personnel entry and exit record, a coordinate axis is set, the initial position of the person in and out of the video data is set to p (0, 0), the end position is q (E, 0), and based on the video time t and the distance from the initial position to the end position, the moving speed M is calculated i，j =E / t; The head of the person entering or leaving is taken as the first feature point, the two hands are taken as the second and third feature points, and the vector sets S1 and S2 are set according to the length E of the trajectory of the lateral movement of the person entering or leaving, where S1={(s1, L1) (s2, L2)…(s n , L n )}, S2={((s1,R1) (s2,R2)......(s n , R n ))}, s1, s2, ... E represent the position of the first feature point under the current vector coordinates, L n and R n Indicates the positions of the second and third feature points under the current vector coordinates, s n =E; Calculate the average weight X and distance weight D of the vector coordinates from the first feature point to the second feature point or the third feature point in the vector sets S1 and S2. The calculation formula is:

[0020]

[0021] Among them, X i，j Number H i，j The average weight of the vector coordinates from the first feature point to the second feature point or the third feature point in the vector set S1 and S2 corresponding to the personnel entry and exit records, D i，j Number H i，j The distance weight from the first feature point to the second feature point or the third feature point in the vector set S1 and S2 corresponding to the personnel entry and exit records; Furthermore, based on all the entry and exit records of personnel, all corresponding moving speeds M, average weights X and distance weights D are calculated, and a normal behavior model of the personnel entering and leaving is established according to the numbers of the personnel entering and leaving; Specifically, based on the entry and exit records corresponding to the person numbered i, the moving speed M, average weight X and distance weight D of all the corresponding entry and exit records are calculated, and then the corresponding normal moving speed m is calculated. i , normal average weight x i and the normal distance weight d i The calculation process is , , , select the maximum and minimum values ​​from all moving speeds M, average weights X, and distance weights D, and mark them as M max 、M min , X max , X min , D max , D min , change the normal moving speed to m i , normal average weight x i and the normal distance weight d i Subtract it from the moving speed range threshold m i，1 =m i -M min , m i，2 =M max -m i Similarly, we get the average weight range threshold x i，1 、x i，2 , distance weight range threshold d i，1 d i，2 ; Set the normal movement range of people entering and leaving (m i，1 , m i )∪(m i , m i，2 )、(x i，1 , x i )∪(x i , x i，2 ), (d i，1 , d i )∪(d i , d i，2 ), and then establish a normal behavior model for each person entering and leaving, and upload it to the database.

[0022] The abnormal behavior judgment module is used to judge whether the person entering or leaving has abnormal behavior according to the normal behavior model of the person entering or leaving, and specifically includes: The abnormal behavior judgment module obtains the corresponding normal behavior model of the in-and-out personnel from the database based on the latest in-and-out records of the personnel in the database and the identity information of the in-and-out personnel; According to the latest personnel entry and exit records, the corresponding movement speed M, average weight X and distance weight D are calculated, and then the corresponding movement model is established; Compare the movement model with the normal behavior model of the in-and-out personnel corresponding to the in-and-out personnel, and then determine whether its movement speed M, average weight X and distance weight D are within the normal movement range of the in-and-out personnel; If the moving speed M, average weight X, and distance weight D are all within the normal range of human motion, then the trajectory of the in-and-out person's motion is considered normal; If two of the moving speed M, average weight X, and distance weight D are within the normal range of human motion, the movement trajectory of the person entering or leaving is judged to be slightly abnormal, and a behavior abnormality label is set for their identity information. If the behavior abnormality label is set for three consecutive corresponding entry and exit records, the movement trajectory of the person entering or leaving is judged to be abnormal; If only one or none of the moving speed M, average weight X, and distance weight D is within the normal range of human motion, then the movement trajectory of the in-and-out personnel is judged to be abnormal; It should be noted that if the movement trajectory of the person entering or leaving is judged to be abnormal, the abnormal behavior judgment module will send an instruction to the control center, and the control center will suspend the identity card's authority to enter and exit the checkpoint; if the movement trajectory of the person entering or leaving is judged to be slightly abnormal, the person trajectory prediction module will update the corresponding person's normal behavior model based on the corresponding person entry and exit records.

[0023] 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. A trajectory intelligent supervision system based on big data, characterized in that: It includes a control center, which is communicatively connected to a database, an entry and exit record generation module, an analysis module, a personnel trajectory prediction module, and an abnormal behavior judgment module; The database is used to store data from various modules, and the database also stores the identity information of all people entering and leaving; The entry and exit record generation module is used to manage the entry and exit of personnel through the checkpoint and collect the movement trajectory of the entry and exit personnel, and generate corresponding entry and exit records of personnel; The analysis module is used to analyze the personnel entry and exit records and send the analysis results to the personnel trajectory prediction module; The personnel trajectory prediction module is used to establish a normal behavior model of personnel entering and leaving according to the analysis results of personnel entry and exit records; The abnormal behavior judgment module is used to judge whether the person entering or leaving has abnormal behavior according to the normal behavior model of the person entering or leaving.

2. According to the big data-based intelligent trajectory monitoring system of claim 1, it is characterized in that: The process of managing the entry and exit personnel passing through the checkpoint by the entry and exit record generation module includes: The entry and exit record generation module is equipped with an identity card scanning device, a gate, multiple cameras and sensors. People entering and exiting can scan their identity cards through the identity card scanning device, and the identity card scanning device reads the identity information in the identity card; The corresponding identity information in the identity information is searched in the database for comparison to determine whether the identity information is correct. If it is judged to be wrong, entry is denied and an alarm is issued. If it is correct, real-time photos of people entering and leaving are taken through the camera, and compared with the photos in the identity information for a second time. If the comparison is consistent, the gate is opened and a record of people entering and exiting is generated; if it is inconsistent, the gate is not opened and an alarm is issued.

3. According to the big data-based intelligent trajectory monitoring system of claim 2, it is characterized in that: The generation process of the personnel entry and exit records includes: The entry and exit record generation module opens the gate. After the sensor senses that a person is entering or leaving the gate, the camera automatically records the video data of the person passing through the gate, and then automatically generates a record of the person entering or leaving the gate and sends it to the database.

4. According to claim 3, a trajectory intelligent supervision system based on big data is characterized in that: The analysis process of the personnel entry and exit records includes: Divide the video data of the entry and exit checkpoint into several image data, set numbers for them, and mark the entry and exit personnel in the image data based on the photos of the entry and exit personnel in the identity information of the entry and exit personnel; Based on the region of the image data where the in-and-out persons are marked, other regions in the image data are blurred, thereby generating image data that only includes the marked in-and-out persons; Generate background image data, and merge all blurred image data with the background image data to form new entry and exit video data; The video data of the entry and exit checkpoint in the personnel entry and exit record is replaced with the new video data of the entry and exit checkpoint, and a new personnel entry and exit record is obtained and sent to the database.

5. The big data-based trajectory intelligent supervision system according to claim 4 is characterized in that: The process of generating background image data includes: Before blurring the image data, a number of image data are randomly selected from the image data and divided into a number of regions of equal size. A codebook is established for the pixels in each region, and each codebook consists of multiple code words. Set a codeword number threshold and compare it with the number of codewords in the pixel, and mark it as a background pixel or a target pixel based on the judgment result; A background pixel threshold is set and compared with the number of background pixels in the area. The pixels are marked as background pixels or target pixels based on the judgment results. The areas marked as background areas in all selected image data are integrated to form background image data that does not include people entering and leaving.

6. The big data-based trajectory intelligent monitoring system according to claim 4 is characterized in that: The personnel trajectory prediction module analyzes personnel entry and exit records in the following process: The personnel trajectory prediction module obtains new personnel entry and exit records from the database, and sets the coordinate axis, initial position and end position according to the movement route of the personnel in and out of the new entry and exit video data in the personnel entry and exit records, and then obtains the movement speed according to the video time and the distance from the initial position to the end position; The head of the person entering or leaving is taken as the first feature point, the two hands are taken as the second and third feature points, and two vector sets are set according to the length of the trajectory of the lateral movement of the person entering or leaving. The average weight and distance weight of the vector coordinates from the first feature point to the second feature point or the third feature point in the vector set are obtained; Based on the personnel entry and exit records corresponding to the personnel entering and exiting, the moving speed, average weight and distance weight of all corresponding personnel entry and exit records are obtained, and then the corresponding normal moving speed, normal average weight and normal distance weight are obtained.

7. The big data-based trajectory intelligent supervision system according to claim 6 is characterized in that: The process of establishing a normal behavior model for people entering and leaving includes: Select the maximum and minimum values ​​from all moving speeds, average weights and distance weights, set a mark, subtract the normal moving speed, normal average weight and normal distance weight from them, and obtain two moving speed range thresholds, two average weight range thresholds, and two distance weight range thresholds; According to the six range thresholds, the normal movement range of people entering and leaving is set, and then the normal behavior model of people entering and leaving corresponding to each person entering and leaving is established and uploaded to the database.

8. The big data-based trajectory intelligent monitoring system according to claim 7 is characterized in that: The process of the abnormal behavior judgment module judging the behavior of people entering and leaving includes: According to the latest personnel entry and exit records, the corresponding movement speed, average weight and distance weight are calculated, and then the corresponding movement model is established; Compare the movement model with the normal behavior model of the in-and-out personnel corresponding to the in-and-out personnel, and then determine whether its movement speed, average weight and distance weight are within the normal movement range of the in-and-out personnel; If the moving speed, average weight, and distance weight are all within the normal range of human motion, then the movement trajectory of the in-and-out personnel is judged to be normal; If two of the moving speed, average weight, and distance weight are within the normal range of human motion, the movement trajectory of the person entering or leaving is judged to be slightly abnormal, and a behavior abnormality label is set for their identity information. If the behavior abnormality label is set for the corresponding entry and exit records three times in a row, the movement trajectory of the person entering or leaving is judged to be abnormal; If only one or none of the moving speed, average weight and distance weight is within the normal range of human motion, the movement trajectory of the in-and-out personnel is judged to be abnormal.

9. The big data-based trajectory intelligent monitoring system according to claim 8 is characterized in that: If the movement trajectory of the person entering or leaving is judged to be abnormal, the abnormal behavior judgment module sends a command to the control center, and then the control center suspends the permission of the person's ID card to enter the checkpoint; If the abnormal behavior judgment module determines that the movement trajectory of the in-and-out personnel is slightly abnormal, the personnel trajectory prediction module will update the corresponding personnel normal behavior model according to the corresponding personnel entry and exit records.

Images