Monitoring data analysis method based on cloud monitoring system

Abstract

The invention discloses a monitoring data analysis method based on a cloud monitoring system, and particularly relates to the technical field of monitoring data, and the method specifically comprises the following steps: collecting monitoring data in an examination room through a monitoring camera, a sensor and other devices, building a data analysis model for examinee behavior information parameters and examination room environment parameters, generating a monitoring evaluation coefficient, and carrying out the analysis of the monitoring evaluation coefficient. Comparing the monitoring evaluation coefficients with a threshold value, generating a high monitoring signal and a low monitoring signal, sending out an early warning prompt for the high monitoring signal, not sending out an early warning prompt for the low monitoring signal, establishing a data set for the monitoring evaluation coefficients corresponding to the low monitoring signal, analyzing the monitoring evaluation coefficients in the data set, generating a risk monitoring signal, and performing risk monitoring on the risk monitoring signal. According to the method, the examination room with problems can be determined through cloud monitoring analysis, manual invigilation is reduced, and the examination room patrol route of examination room patrol personnel can be optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring data. More specifically, the present invention relates to a method for analyzing monitoring data based on a cloud monitoring system. Background Art

[0002] In the field of universities, a cloud monitoring system is a software system used to monitor, manage, and optimize IT resources and services based on a cloud computing architecture within a campus. These systems are usually managed and maintained by the information technology department or a specialized technical team of the university, aiming to ensure the stability, availability, and security of the cloud infrastructure and services within the campus;

[0003] Currently, the cloud monitoring system for unmanned examination rooms can usually only monitor activities within the examination room, but cannot monitor activities inside the examinees' bodies or other non-visual ranges. Therefore, some cheating behaviors may not be fully detected. The main reason is that the cloud monitoring system for examination rooms may not be sensitive enough in anomaly detection and cannot effectively identify the cheating behaviors or other abnormal situations of the examinees, resulting in missed detections. And mainly, it still requires manual judgment of the examinees' cheating behaviors during monitoring.

[0004] To solve the above defects, a technical solution is provided now. Summary of the Invention

[0005] In order to overcome the above defects of the prior art, an embodiment of the present invention provides a method for analyzing monitoring data based on a cloud monitoring system to solve the problems raised in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for analyzing monitoring data based on a cloud monitoring system specifically includes the following steps:

[0008] S1: Collect monitoring data in the examination room through devices such as monitoring cameras and sensors. The monitoring data includes examinee behavior information parameters and examination room environment parameters;

[0009] S2: Establish a data analysis model for the examinee behavior information parameters and the examination room environment parameters to generate a monitoring evaluation coefficient;

[0010] S3: Compare the monitoring evaluation coefficient with a threshold value to generate a high monitoring signal and a low monitoring signal, and issue a warning prompt for the high monitoring signal and do not issue a warning prompt for the low monitoring signal;

[0011] S4: Establish a data set for the monitoring evaluation coefficients corresponding to the low monitoring signals, analyze the monitoring evaluation coefficients within the data set to generate a risk monitoring signal, and generate a monitoring coefficient through the risk monitoring signal.

[0012] In a preferred embodiment, the candidate behavior information parameters include a head offset coefficient and a position stability distribution coefficient, and the examination room environment parameters include a decibel anomaly deviation coefficient. After collection, the head offset coefficient, the position stability distribution coefficient, and the decibel anomaly deviation coefficient are respectively marked as TB pc , WZ wd , and FB yc .

[0013] In a preferred embodiment, the acquisition logic of the head offset coefficient is as follows:

[0014] Identify the candidate's head features through a high-definition camera, record the candidate's head features in a rectangular coordinate system, and mark the coordinates of the candidate's head features in the rectangular coordinate system as: (x n , y n ); where n = 1, 2, 3... N, N is a positive integer, and n is the numbering of the times of collecting the candidate's head features;

[0015] Set the standard values of the coordinates of the candidate's head features in the rectangular coordinate system, and mark the standard values of the coordinates of the candidate's head features in the rectangular coordinate system as: (x bz , y bz );

[0016] Calculate the head offset coefficient, and the calculation formula is: where TB pc is the head offset coefficient.

[0017] In a preferred embodiment, the acquisition logic of the position stability distribution coefficient is as follows:

[0018] Identify the candidate's position features through a high-definition camera, record the candidate's position features in a rectangular coordinate system, and mark the coordinates of the candidate's position features in the rectangular coordinate system as: (a i , b i ); where i = 1, 2, 3... I, I is a positive integer, and i is the numbering of the times of collecting the candidate's position features;

[0019] Calculate the average value and standard deviation of the candidate's position features at the abscissa and ordinate, and mark the average value and standard deviation of the candidate's position features at the abscissa as: a avg and a bzc , and mark the average value and standard deviation of the candidate's position features at the ordinate as: b avg and b bzc , where

[0020] Calculate the position stability distribution coefficient, and the calculation formula is: where WZwd is the position stability distribution coefficient.

[0021] In a preferred embodiment, the acquisition logic of the decibel anomaly deviation coefficient is as follows:

[0022] Set a gradient range fb for the decibel min ~fb max , and obtain the decibel value in the examination room in real time. Mark the decibel value in the examination room as: fb. If fb is within the gradient range fb min ~fb max , it indicates that the decibel value in the examination room is normal. If fb is not within the gradient range fb min ~fb max x, it indicates that the decibel value in the examination room is abnormal. Calibrate the deviation value of the decibel in the examination room as fb(t). The way to obtain fb(t) is:

[0023] If fb is less than fb min , then fb(t) is the absolute value of the difference between b and fb min . If fb is greater than fb max , then fb(t) is the absolute value of the difference between fb and fb max ;

[0024] Calculate the decibel anomaly deviation coefficient through the formula. The expression is: where fb(t) is the deviation value of the decibel in the examination room, and t 1 ~t 2 is the time period when the decibel in the examination room is not within the gradient range fb min ~fb max .

[0025] In a preferred embodiment, establish a data analysis model for the candidate behavior information parameters and the examination room environment parameters to generate a monitoring and evaluation coefficient, including:

[0026] Obtain the head offset coefficient, position stability distribution coefficient, and decibel anomaly deviation coefficient according to the candidate behavior information parameters and the examination room environment parameters;

[0027] Establish a data analysis model for the head offset coefficient, position stability distribution coefficient, and decibel anomaly deviation coefficient to generate a monitoring and evaluation coefficient;

[0028] The head offset coefficient, position stability distribution coefficient, and decibel anomaly deviation coefficient are positively correlated with the monitoring and evaluation coefficient.

[0029] In a preferred embodiment, compare the generated monitoring and evaluation coefficient with the threshold YZ 1 . If the monitoring and evaluation coefficient is greater than or equal to the threshold YZ 1, indicating that the higher the possibility of an abnormality occurring in the examination room, a high monitoring signal is generated to prompt the examination room patrol personnel to go to the examination room in time to determine the problem with the abnormality. If the monitoring evaluation coefficient is less than the threshold YZ 1 , indicating that the possibility of an abnormal situation occurring in the examination room is relatively low, a low monitoring signal is generated.

[0030] In a preferred embodiment, the low monitoring signals are collected, and a data set is established for the monitoring evaluation coefficients generated in the examination rooms with low monitoring signals. The data set is marked as M, then M = {JK k} = {JK 1 、JK 2 、JK 3 ……JK K}, where k is the number of monitoring evaluation coefficients corresponding to the low monitoring signals, k = 1, 2, 3...K, and K is a positive integer.

[0031] In a preferred embodiment, the monitoring evaluation coefficients in the data set are respectively compared with the thresholds YZ 1 、YZ 2 、YZ 3 . Among them, YZ 1 >YZ 2 >YZ 3 , including:

[0032] If the monitoring evaluation coefficient is less than the threshold YZ 1 and greater than or equal to the threshold YZ 2 , a high-risk supervision signal is generated;

[0033] If the monitoring evaluation coefficient is less than the threshold YZ 2 and greater than or equal to the threshold YZ 3 , a medium-risk supervision signal is generated;

[0034] If the monitoring evaluation coefficient is less than the threshold YZ 3 , a low-risk supervision signal is generated;

[0035] The risk supervision signals after comparison with the thresholds YZ 1 、YZ 2 、YZ 3 in the data set are used to establish a data set Q, and the quantities of the high-risk supervision signals, medium-risk supervision signals, and low-risk supervision signals in the data set Q are counted;

[0036] Label the number of high - risk monitoring signals, the number of medium - risk monitoring signals, and the number of low - risk monitoring signals as GD, ZD, and DD respectively. Establish a data analysis model for the number of high - risk monitoring signals GD, the number of medium - risk monitoring signals ZD, and the number of low - risk monitoring signals DD to generate a monitoring coefficient. The formula is: JC=(GD + ZD + DD)e β 1 GD+β 2 ZD+β 3 DD+1 ; where JC is the monitoring coefficient, β 1 、β 2 、β 3 are the proportionality coefficients of the number of high - risk monitoring signals GD, the number of medium - risk monitoring signals ZD, and the number of low - risk monitoring signals DD. β 1 、β 2 、β 3 are greater than 0.

[0037] Technical effects and advantages of the present invention:

[0038] The present invention collects the candidate behavior information parameters and examination room environment parameters in the unmanned examination room, establishes a data analysis model for the candidate behavior information parameters and examination room environment parameters to generate a monitoring and evaluation coefficient, compares the monitoring and evaluation coefficient with a threshold value to generate high - monitoring signals and low - monitoring signals. If the invigilation status of the examination room is poor, high - monitoring signals are generated to prompt the examination room patrol personnel to patrol the examination room and determine the problems in the examination room, which helps to identify problematic examination rooms through cloud monitoring analysis, reduce manual invigilation, and can optimize the patrol routes of the examination room patrol personnel. Brief Description of the Drawings

[0039] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the accompanying drawings;

[0040] Figure 1 is a schematic flow chart of a monitoring data analysis method based on a cloud monitoring system of the present invention. Detailed Embodiments

[0041] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0042] Embodiment 1

[0043] Figure 1 The flowchart of a method for analyzing monitoring data based on a cloud monitoring system according to the present invention is given, which specifically includes the following steps:

[0044] S1: Collect monitoring data in the examination room through devices such as monitoring cameras and sensors. The monitoring data includes candidate behavior information parameters and examination room environment parameters;

[0045] S2: Establish a data analysis model for the candidate behavior information parameters and examination room environment parameters to generate a monitoring evaluation coefficient;

[0046] S3: Compare the monitoring evaluation coefficient with a threshold value to generate a high monitoring signal and a low monitoring signal, and issue a warning prompt for the high monitoring signal, and do not issue a warning prompt for the low monitoring signal;

[0047] S4: Establish a data set for the monitoring evaluation coefficients corresponding to the low monitoring signals, analyze the monitoring evaluation coefficients in the data set to generate a risk monitoring signal, and generate a monitoring coefficient through the risk monitoring signal.

[0048] In step 1, the candidate behavior information parameters include a head offset coefficient and a position stability distribution coefficient, and the examination room environment parameters include a decibel anomaly deviation coefficient. After collection, the head offset coefficient, the position stability distribution coefficient, and the decibel anomaly deviation coefficient are respectively marked as TB pc 、WZ wd and FB yc ;

[0049] If, after the behavior of a candidate in the examination room is collected by the monitoring device, it is found that the candidate's behavior is abnormal, it may indicate that the candidate has cheating behavior, examination anxiety, physical discomfort, etc.;

[0050] Cheating behavior: Abnormal behaviors such as head offset and position offset may imply that the candidate is trying to cheat, such as trying to peek at others' test papers, communicate with others, or obtain external materials, etc.;

[0051] Examination anxiety: Some candidates may exhibit behaviors such as head offset and position offset due to nervousness or anxiety, which may indicate that the candidate's mental state is unstable and requires special attention;

[0052] Physical discomfort: Some candidates may exhibit behaviors such as head offset and position offset due to physical discomfort, such as dizziness and vertigo, etc., which may need to be dealt with in a timely manner to ensure the health and safety of the candidate.

[0053] It should be noted that a high-definition camera and a video surveillance system are used to monitor the examination room in real time. This includes technologies such as the installation location and angle adjustment of the camera, the transmission and storage of video streams, etc. The resources provided by the cloud computing platform, such as storage, computing, and network, are utilized to process, store, and analyze the monitoring data. This includes technologies such as the erection, configuration, and management of cloud servers. Artificial intelligence technologies, such as machine learning and deep learning, are used to perform intelligent analysis and recognition on the monitoring data. For example, the behavior characteristics of candidates are identified through image recognition technology, or the voice data in the examination room is analyzed through natural language processing technology.

[0054] The acquisition logic of the head offset coefficient is as follows: The head characteristics of the candidate are recognized through a high-definition camera, and the head characteristics of the candidate are recorded in a rectangular coordinate system, and the coordinates of the head characteristics of the candidate in the rectangular coordinate system are marked as: (x n , y n ); where n = 1, 2, 3... N, N is a positive integer, and n is the numbering of the times of collecting the head characteristics of the candidate;

[0055] Set the standard value of the coordinates of the head characteristics of the candidate in the rectangular coordinate system, and mark the standard value of the coordinates of the head characteristics of the candidate in the rectangular coordinate system as: (x bz , y bz );

[0056] Calculate the head offset coefficient, and the calculation formula is: where TB pc is the head offset coefficient.

[0057] It can be seen from the formula that the larger the head offset coefficient, the more times the candidate shows abnormal head offset behaviors such as tilting the head or lowering the head in the picture captured by the high-definition camera, indicating that the candidate's behavior is more abnormal. On the contrary, the smaller the head offset coefficient, the fewer times the candidate shows abnormal head offset behaviors such as tilting the head or lowering the head in the picture captured by the high-definition camera, indicating that the candidate's behavior is more normal.

[0058] The acquisition logic of the position stable distribution coefficient is as follows: The position characteristics of the candidate are recognized through a high-definition camera, and the position characteristics of the candidate are recorded in a rectangular coordinate system, and the coordinates of the position characteristics of the candidate in the rectangular coordinate system are marked as: (a i , b i ); where i = 1, 2, 3... I, I is a positive integer, and i is the numbering of the times of collecting the position characteristics of the candidate;

[0059] Calculate the average value and standard deviation of the candidate's position characteristics at the abscissa and ordinate, and mark the average value and standard deviation of the candidate's position characteristics at the abscissa as: a avg and a bzc, mark the mean and standard deviation of the candidate's position characteristics on the vertical axis as: b avg and b bzc ,in,

[0060] Calculate the position stability distribution coefficient, the calculation formula is: Among them, WZ wd is the position stability distribution coefficient.

[0061] It can be seen from the formula that the larger the position stability distribution coefficient is, the more it means that the position of the examinee is constantly changing in the picture captured by the high-definition camera, and the amplitude of each position change is larger, indicating that the examinee's behavior is more abnormal. Conversely, the smaller the position stability distribution coefficient is, the more it means that the amplitude of each position change of the examinee is larger in the picture captured by the high-definition camera, indicating that the examinee's behavior is more normal.

[0062] When the decibel level in the examination room is abnormal, the following situations may occur:

[0063] External noise interference: There are noise sources around the examination room, such as construction, traffic, human voices, etc., which affect the quiet environment in the examination room. Solutions include strengthening sound insulation measures, adjusting the examination room settings to be away from noise sources, and installing noise absorption devices in the examination room.

[0064] Internal sound interference: There are unnecessary sound interferences in the examination room, such as the sound of students talking to each other and the sound of chairs rubbing against each other. Management regulations and monitoring systems should be used to control the behavior in the examination room and remind students to keep quiet;

[0065] Equipment failure: Monitoring equipment or other electronic equipment in the examination room may fail, causing abnormal sounds. The faulty equipment should be repaired or replaced in a timely manner to ensure normal operation of the equipment;

[0066] Malicious sabotage: It is possible that someone will deliberately create noise disturbances in the examination room to affect the examination order. Malicious sabotage should be discovered and dealt with in a timely manner through enhanced security measures and monitoring systems.

[0067] For monitoring of decibels in the examination room, it can be obtained through the sound collection device built into the camera, or by reasonably arranging sound sensors or audio collection equipment according to the layout and needs of the examination room.

[0068] The acquisition logic of the decibel abnormal deviation coefficient is: set the gradient range fb for decibel min ~fb max , obtain the decibel value in the examination room in real time, and mark the decibel value in the examination room as: fb. If fb is within the gradient range fb min ~fb maxInside, it indicates that the decibel value in the examination room is normal. If fb is not within the gradient range fb min ~fb max x, it indicates that the decibel value in the examination room is abnormal. The deviation value of the decibel in the examination room is calibrated as fb(t). The way to obtain fb(t) is as follows:

[0069] If fb is less than fb min , then fb(t) is the absolute value of the difference between b and fb min If fb is greater than fb max , then fb(t) is the absolute value of the difference between fb and fb max ;

[0070] It should be noted that the larger the deviation value fb(t) of the decibel, the greater the possible abnormal situation in the examination room. On the contrary, the smaller the possible abnormal situation in the examination room;

[0071] The decibel abnormal deviation coefficient is calculated by the formula. The expression is: Among them, fb(t) is the deviation value of the decibel in the examination room, and t 1 ~t 2 is the time period when the decibel in the examination room is not within the gradient range fb min ~fb max ;

[0072] It can be seen from the formula that the larger the decibel abnormal deviation coefficient, the longer the time when the examination room environment is different from the normal examination room state, and the greater the possibility of abnormal situations in the examination room. On the contrary, the smaller the decibel abnormal deviation coefficient, the shorter the time when the examination room environment is different from the normal examination room state, and the smaller the possibility of abnormal situations in the examination room.

[0073] After obtaining the head offset coefficient TB pc , the position stability distribution coefficient WZ wd and the decibel abnormal deviation coefficient FB yc , a data analysis model is established to generate a monitoring and evaluation coefficient. The calculation formula of the monitoring and evaluation coefficient is: Among them, JK is the monitoring and evaluation coefficient, and α 1 , α 2 , α 3 are the proportionality coefficients of the head offset coefficient, the position stability distribution coefficient and the decibel abnormal deviation coefficient, and α 1 , α 2 , α 3 are greater than 0.

[0074] As can be seen from the formula, the larger the head offset coefficient, the position stability distribution coefficient, and the decibel anomaly deviation coefficient, the larger the monitoring evaluation coefficient, indicating a higher possibility of anomalies in the examination room. Conversely, the smaller the head offset coefficient, the position stability distribution coefficient, and the decibel anomaly deviation coefficient, the smaller the monitoring evaluation coefficient, indicating a lower possibility of anomalies in the examination room.

[0075] Generate the monitoring evaluation coefficient and compare it with the threshold YZ 1 If the monitoring evaluation coefficient is greater than or equal to the threshold YZ 1 , it indicates a higher possibility of anomalies in the examination room, then generate a high monitoring signal to prompt the examination room patrol personnel to go to the examination room in time to determine the problems with anomalies. If the monitoring evaluation coefficient is less than the threshold YZ 1 , it indicates a lower possibility of anomalies in the examination room, then generate a low monitoring signal.

[0076] Collect the low monitoring signals, establish a data set for the monitoring evaluation coefficients generated in the examination rooms with low monitoring signals, and mark the data set as M. Then M = {JK k} = {JK 1 , JK 2 , JK 3 ... JK K}), where k is the number of monitoring evaluation coefficients corresponding to the low monitoring signals, k = 1, 2, 3... K, and K is a positive integer;

[0077] Compare the monitoring evaluation coefficients in the data set with the thresholds YZ 1 , YZ 2 , YZ 3 respectively. Among them, YZ 1 > YZ 2 > YZ 3 . If the monitoring evaluation coefficient is less than the threshold YZ 1 and greater than or equal to the threshold YZ 2 , then generate a high-risk monitoring signal. If the monitoring evaluation coefficient is less than the threshold YZ 2 and greater than or equal to the threshold YZ 3 , then generate a medium-risk monitoring signal. If the monitoring evaluation coefficient is less than the threshold YZ 3 , then generate a low-risk monitoring signal. For the thresholds YZ 1 , YZ 2 , YZ 3After comparison, a data set Q of risk monitoring signals is established. The quantities of high-risk monitoring signals, medium-risk monitoring signals, and low-risk monitoring signals within the data set Q are counted. The quantities of high-risk monitoring signals, medium-risk monitoring signals, and low-risk monitoring signals are respectively labeled as GD, ZD, and DD. A data analysis model is established for the quantities GD of high-risk monitoring signals, ZD of medium-risk monitoring signals, and DD of low-risk monitoring signals to generate a monitoring coefficient. The formula is as follows: where JC is the monitoring coefficient, and β 1 , β 2 , β 3 are the proportionality coefficients for the quantities GD of high-risk monitoring signals, ZD of medium-risk monitoring signals, and DD of low-risk monitoring signals. β 1 , β 2 , β 3 are greater than 0.

[0078] As can be seen from the formula, the larger the monitoring coefficient, the greater the abnormal situation in the examination room, indicating that there are more small actions by candidates in the examination room or the examination room environment is worse. On the contrary, the smaller the monitoring coefficient, the smaller the abnormal situation in the examination room, indicating that there are fewer small actions by candidates in the examination room or the examination room environment is better.

[0079] According to the monitoring coefficients of the examination rooms, they are sorted from large to small to obtain the rankings of different examination rooms. For the examination rooms with large monitoring coefficients, examination room patrol personnel are arranged to prioritize the investigation of the examination rooms.

[0080] The present invention collects the candidate behavior information parameters and examination room environment parameters in an unmanned examination room, establishes a data analysis model for the candidate behavior information parameters and examination room environment parameters to generate a monitoring and evaluation coefficient, compares the monitoring and evaluation coefficient with a threshold value to generate high monitoring signals and low monitoring signals. If the invigilation status of the examination room is poor, high monitoring signals are generated to prompt the examination room patrol personnel to patrol the examination room and determine the problems in the examination room, which helps to identify the problematic examination rooms through cloud monitoring analysis, reduce manual invigilation, and can optimize the patrol routes of the examination room patrol personnel.

[0081] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data and performing software simulation to obtain a formula that is closest to the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0082] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, or magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0083] It should be understood that in various embodiments of the present application, the order numbers of the above processes do not indicate the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0084] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0085] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0086] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the devices or units can be in electrical, mechanical, or other forms.

[0087] If the above functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs and other various media that can store program codes.

[0088] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A monitoring data analysis method based on a cloud monitoring system, characterized in that: The specific steps include: S1: Collect monitoring data in the examination room through monitoring cameras, sensors and other equipment. The monitoring data includes the parameters of the examinee's behavior information and the examination room environment parameters; S2: Establish a data analysis model based on the examinee behavior information parameters and examination room environment parameters to generate monitoring evaluation coefficients; S3: Compare the monitoring evaluation coefficient with the threshold value, generate a high monitoring signal and a low monitoring signal, issue an early warning for the high monitoring signal, and do not issue an early warning for the low monitoring signal; S4: Establish a data set for the monitoring evaluation coefficients corresponding to the low monitoring signal, analyze the monitoring evaluation coefficients in the data set, generate a risk monitoring signal, and generate a monitoring coefficient through the risk monitoring signal.

2. The monitoring data analysis method based on the cloud monitoring system according to claim 1, characterized in that: The examinee behavior information parameters include the head deviation coefficient and the position stability distribution coefficient, and the examination room environment parameters include the decibel abnormal deviation coefficient. After collection, the head deviation coefficient, position stability distribution coefficient and decibel abnormal deviation coefficient are marked as TB respectively. pc , WZ wd and Facebook yc .

3. The monitoring data analysis method based on the cloud monitoring system according to claim 2 is characterized in that: The logic for obtaining the head offset coefficient is: The head features of the examinee are identified by a high-definition camera, the head features of the examinee are recorded in a rectangular coordinate system, and the coordinates of the examinee's head features in the rectangular coordinate system are marked as: (x n ,y n );wherein, n=1, 2, 3...N, N is a positive integer, and n is the number of times the head features of the examinee are collected; Set the standard value of the candidate's head feature in the rectangular coordinate system, and mark the standard value of the candidate's head feature in the rectangular coordinate system as: (x bz ,y bz ); Calculate the head offset coefficient using the following formula: Among them, TB pc is the head offset coefficient.

4. The monitoring data analysis method based on the cloud monitoring system according to claim 2 is characterized in that: The acquisition logic of the position stability distribution coefficient is: The position features of the examinee are identified by a high-definition camera, the position features of the examinee are recorded in a rectangular coordinate system, and the coordinates of the examinee's position features in the rectangular coordinate system are marked as: (a i , b i );wherein, i=1, 2, 3...I, I is a positive integer, and i is the number of times the position characteristics of the examinee are collected; Calculate the mean and standard deviation of the candidate's position characteristics at the horizontal and vertical coordinates, and mark the mean and standard deviation of the candidate's position characteristics at the horizontal coordinate as: a avg and a bzc , mark the mean and standard deviation of the candidate's position characteristics on the vertical axis as: b avg and b bzc ,in, Calculate the position stability distribution coefficient, the calculation formula is: Among them, WZ wd is the position stability distribution coefficient.

5. The monitoring data analysis method based on the cloud monitoring system according to claim 2, characterized in that: The acquisition logic of the decibel abnormal deviation coefficient is: Set the gradient range fb for decibels min ~fb max , obtain the decibel value in the examination room in real time, and mark the decibel value in the examination room as: fb. If fb is within the gradient range fb min ~fb max If fb is not within the gradient range, it indicates that the decibel value in the examination room is normal. min ~fb max x, indicating that the decibel value in the examination room is abnormal. The deviation value of the decibel in the examination room is calibrated as fb(t). The method of obtaining fb(t) is: If fb is less than fb min , then fb(t) is b and fb min The absolute value of the difference, if fb is greater than fb max , then fb(t) is fb and fb max The absolute value of the difference; The decibel abnormal deviation coefficient is calculated by the formula, and the expression is: Among them, fb(t) is the deviation value of decibel in the examination room, t1~t2 is the decibel in the examination room that is not within the gradient range fb min ~fb max The time period between.

6. The monitoring data analysis method based on a cloud monitoring system according to claim 1, characterized in that: The data analysis model is established based on the examinee behavior information parameters and the examination room environment parameters to generate monitoring evaluation coefficients, including: According to the examinee's behavior information parameters and the examination room environment parameters, the head displacement coefficient, position stability distribution coefficient, and decibel abnormal deviation coefficient are obtained; The head deviation coefficient, position stability distribution coefficient and decibel abnormal deviation coefficient are used to establish a data analysis model to generate a monitoring evaluation coefficient; The head excursion coefficient, position stability distribution coefficient, decibel abnormal deviation coefficient and monitoring evaluation coefficient are positively correlated.

7. A monitoring data analysis method based on a cloud monitoring system according to claim 6, characterized in that: The generated monitoring evaluation coefficient is compared with the threshold value YZ1. If the monitoring evaluation coefficient is greater than or equal to the threshold value YZ1, it indicates that the possibility of abnormality in the examination room is higher, and a high monitoring signal is generated to prompt the examination room patrol personnel to go to the examination room in time to determine the abnormal problem. If the monitoring evaluation coefficient is less than the threshold value YZ1, it indicates that the possibility of abnormality in the examination room is low, and a low monitoring signal is generated.

8. The monitoring data analysis method based on the cloud monitoring system according to claim 7, characterized in that: Collect low monitoring signals, establish a data set of monitoring evaluation coefficients generated in the examination room with low monitoring signals, and mark the data set as M, then M={JK k }={JK1, JK2, JK3...JK K }, where k is the number of monitoring evaluation coefficients corresponding to the low monitoring signal, k = 1, 2, 3...K, and K is a positive integer.

9. The monitoring data analysis method based on a cloud monitoring system according to claim 8, characterized in that: The monitoring evaluation coefficients in the data set are compared with the thresholds YZ1, YZ2, and YZ3 respectively, where YZ1>YZ2>YZ3, including: If the monitoring evaluation coefficient is less than the threshold value YZ1 and greater than or equal to the threshold value YZ2, a high-risk monitoring signal is generated; If the monitoring evaluation coefficient is less than the threshold value YZ2 and greater than or equal to the threshold value YZ3, a medium risk monitoring signal is generated; If the monitoring evaluation coefficient is less than the threshold value YZ3, a low-risk monitoring signal is generated; The risk monitoring signals in the data set that are compared with the thresholds YZ1, YZ2, and YZ3 are used to establish a data set Q, and the number of high-risk monitoring signals, medium-risk monitoring signals, and low-risk monitoring signals in the data set Q is counted; The number of high-risk monitoring signals, the number of medium-risk monitoring signals, and the number of low-risk monitoring signals are calibrated as GD, ZD, and DD, respectively. A data analysis model is established based on the number of high-risk monitoring signals GD, the number of medium-risk monitoring signals ZD, and the number of low-risk monitoring signals DD to generate a monitoring coefficient based on the following formula: Among them, JC is the monitoring coefficient, β1, β2, and β3 are the proportional coefficients of the number of high-risk monitoring signals GD, the number of medium-risk monitoring signals ZD, and the number of low-risk monitoring signals DD, and β1, β2, and β3 are greater than 0.

Images