阅读说明：本技术 基于人脸识别的在线自动监考系统及方法 (Online automatic invigilation system and method based on face recognition ) 是由 彭峻国 于 2021-08-31 设计创作，主要内容包括：本发明涉及监考系统技术领域,公开了一种基于人脸识别的在线自动监考系统,包括处理器模块,以及分别与处理模块连接的人脸识别模块、动作采集模块、提醒模块；人脸识别模块,用来提取考生的面部信息特征得到特征集合,并将特征集合发送至处理器模块进行考生身份认证；动作采集模块采集考生在考试过程中的动作并发送至处理器模块进行作弊行为的认定,并得到判断结果,随后将判断结果发送至提醒模块；提醒模块根据判断结果,按照预设的提醒方式作出提醒。本发明通过人脸识别技术确认考生身份,同时考试过程中全程自动监考,具有防止考生作弊的有益效果。(The invention relates to the technical field of invigilation systems, and discloses an online automatic invigilation system based on face recognition, which comprises a processor module, a face recognition module, an action acquisition module and a reminding module, wherein the face recognition module, the action acquisition module and the reminding module are respectively connected with the processor module; the face recognition module is used for extracting facial information features of the examinees to obtain a feature set and sending the feature set to the processor module for identity authentication of the examinees; the action acquisition module acquires the action of the examinee in the examination process and sends the action to the processor module for identifying cheating behaviors, a judgment result is obtained, and then the judgment result is sent to the reminding module; and the reminding module is used for reminding according to the judgment result and a preset reminding mode. The invention confirms the identity of the examinee through the face recognition technology, and meanwhile, the examinee is automatically invigilated in the whole examination process, thereby having the beneficial effect of preventing the examinee from cheating.)

1. On-line automatic invigilation system based on face identification, its characterized in that: the face recognition system comprises a processor module, a face recognition module, an action acquisition module and a reminding module, wherein the face recognition module, the action acquisition module and the reminding module are respectively connected with the processing module;

the face recognition module is used for extracting facial information features of the examinees to obtain a feature set and sending the feature set to the processor module for identity authentication of the examinees;

the action acquisition module comprises a plurality of high-definition cameras arranged in the examination room and is used for acquiring actions of the examinee in the examination process, forming an action set and then sending the action set to the processor module for identifying cheating behaviors;

the processor module comprises an analysis and judgment unit and a data storage unit;

the analysis and judgment unit is used for receiving the feature set, comparing the feature set with the pre-stored facial features of the examinee and confirming the identity of the examinee; after receiving the action set, the analysis and judgment unit judges the actions of the examinees according to a preset judgment method, and if the preset judgment condition is met, the corresponding examinees are judged to have cheating behaviors; otherwise, judging that the corresponding examinee is not abnormal;

the data storage unit is used for storing data of the face recognition module, data of the action acquisition module, facial feature data of an examinee and judgment method data;

the processor module is used for analyzing and judging the action set made by the examinee in the examination process, obtaining a judgment result and then sending the judgment result to the reminding module;

the reminding module comprises a display screen and a loudspeaker; the reminding module is used for receiving the judgment result and reminding according to a preset reminding mode.

2. The face recognition based on-line automatic invigilation system of claim 1, characterized in that: the pre-stored facial features of the examinees are the facial features of all the examinees collected and stored in the data storage unit before examination.

3. The face recognition based on-line automatic invigilation system of claim 1, characterized in that: the preset judging method comprises the steps of comparing the collected action set with pre-stored cheating behaviors, judging that the examinee cheats if the contrast is larger than a threshold value, and judging that the examinee is not abnormal if the contrast is smaller than the threshold value.

4. The face recognition based on-line automatic invigilation system of claim 3, characterized in that: the pre-stored cheating behaviors include a head turning behavior, an upper body deviation behavior, a little-copy behavior and a seat-off behavior.

5. The online automatic invigilation method based on the face recognition is characterized by comprising the following steps of:

step S1, starting an online automatic invigilation system before a preset time period from the beginning of an examination, confirming the identity of an examinee by a face recognition module before the examinee enters an examination room, and reminding the module to send out an examination seat number corresponding to the examinee when the identity of the examinee is confirmed to be correct;

step S2, after meeting the requirement of examination starting time and the examinee starts moving a pen to answer questions, the on-line automatic invigilating system starts invigilating work, the action collection module collects the actions made by the examinee in the examination process in real time and forms an action set, and then the action set is sent to the processor module;

step S3, after receiving the action set, the processor module judges the action of the examinee according to a preset judgment method, and if the preset judgment condition is met and the contrast ratio of the collected action set and the pre-stored cheating behavior characteristic picture is above a threshold value, the corresponding examinee is judged to have cheating behavior; otherwise, judging that the corresponding examinee is not abnormal; after the judgment result is obtained, the processor module sends the judgment result to the reminding module;

and step S4, the reminding module receives the judgment result and then reminds according to a preset reminding mode.

6. The online automatic invigilation method based on face recognition according to claim 5, characterized in that: the preset reminding mode is that if the examinee is judged to have cheating behaviors, voice reminding is sent out in the examination room through a loudspeaker; and if the examinee is judged to be abnormal, sending remark reminding to a background.

7. The online automatic invigilation method based on face recognition according to claim 6, characterized in that: the voice prompt is to play voice containing prompt information; the reminding information is a sentence of 'please pay attention to you have cheating'.

8. The online automatic invigilation method based on face recognition according to claim 6, characterized in that: the cheating act includes the acts of,

head steering behavior, wherein the left-right steering angle of the examinee's head exceeds the seat center line by 45 degrees;

the upper half body deviating behavior is that the left-right deviation angle of the upper half body of the examinee exceeds the seat center line by 30 degrees;

small reading behavior, the hand of the examinee leaves the seat, and the head droop angle exceeds 60 degrees;

and (4) leaving the seat, and leaving the current examination seat by the body of the examinee.

9. The online automatic invigilation method based on face recognition according to claim 5, characterized in that: in step S3, when the motion is captured, if the high-definition camera captures a hand-lifting motion within the capture time period, the capturing motions corresponding to all the images within the capture time period are compliant motions.

10. The online automatic invigilation method based on face recognition according to claim 5, characterized in that: the reminding module sends out the examination seat number corresponding to the examinee, and the display screen displays the seat coordinate of the examinee and a walking route from the current position of the examinee to the position of the seat coordinate to the examinee passing through identity authentication.

Technical Field

The invention relates to the technical field of invigilation systems, in particular to an online automatic invigilation system and method based on face recognition.

Background

Along with the development of science and technology, more and more traditional trades are perfectly combined with science and technology, and especially in the education industry, the one-to-many teaching from the opposite side below the original line is developed to the one-to-one teaching on the line by utilizing a computer, so that the whole teaching mode is greatly changed. In the traditional examination, in order to avoid cheating by extremely individual examinees and create a fair examination environment for vast examinees, a teacher is required to supervise the examination manually in the whole examination process, manpower resources are wasted to a great extent, and the examination monitoring system is also a work with great workload for supervising the teacher. Therefore, in order to reduce the extra work of the proctor and ensure the fairness of the examination, the research of the on-line automatic proctor system is very necessary.

In the prior art, an intelligent invigilation anti-cheating system relates to the technical field of invigilation systems. The invention comprises an invigilation system, wherein the invigilation system is connected with a monitoring camera, and the invigilation camera is arranged right in front of an examinee; the invigilation system comprises an examinee information confirmation module, a test paper management module, an examinee behavior analysis module, an acquisition module and a storage module; the examinee information confirming module comprises a face recognition unit, the collecting module is used for collecting images and sounds of cheating actions of the examinees, and the invigilating system is connected with a server. According to the invention, the identity information of the examinee is confirmed through the face recognition unit, and the examinee behavior analysis module captures and analyzes the head action, the voice and the limb action of the examinee so as to judge whether the examinee has cheating behaviors or not, so that the burden of invigilators is reduced, and the cheating phenomenon in invigilators is effectively avoided.

Although this system can perform the function of intelligent invigilation, the invigilation process is too complicated and low in accuracy, and is not suitable for all examinations, such as traditional paper examinations. Therefore, an on-line invigilation system suitable for all examinations and having high efficiency and accuracy in invigilation process is urgently needed.

Disclosure of Invention

The invention aims to provide an online automatic invigilation system based on face recognition, and solves the technical problems of insufficient adaptability, low efficiency and low accuracy of the existing invigilation system.

In order to achieve the purpose, the invention adopts the following technical scheme: the online automatic invigilation system based on the face recognition comprises a processor module, a face recognition module, an action acquisition module and a reminding module, wherein the face recognition module, the action acquisition module and the reminding module are respectively connected with the processor module;

the face recognition module is used for extracting facial information features of the examinees to obtain a feature set and sending the feature set to the processor module for identity authentication of the examinees;

the action acquisition module comprises a plurality of high-definition cameras arranged in the examination room and is used for acquiring actions of the examinee in the examination process, forming an action set and then sending the action set to the processor module for identifying cheating actions;

the processor module comprises an analysis and judgment unit and a data storage unit;

the analysis and judgment unit is used for receiving the feature set, comparing the feature set with the pre-stored facial features of the examinee and confirming the identity of the examinee; after receiving the action set, the analysis and judgment unit judges the actions of the examinees according to a preset judgment method, and if the preset judgment condition is met, the corresponding examinees are judged to have cheating behaviors; otherwise, judging that the corresponding examinee is not abnormal;

the data storage unit is used for storing data of the face recognition module, data of the action acquisition module, facial feature data of the examinee and judgment method data;

the processor module is used for analyzing and judging the action set made by the examinee in the examination process, obtaining a judgment result and then sending the judgment result to the reminding module;

the reminding module comprises a display screen and a loudspeaker; the reminding module is used for receiving the judgment result and reminding according to a preset reminding mode.

The principle and the advantages of the scheme are as follows: during practical application, firstly, the identity of an examinee is confirmed by utilizing a face recognition technology, so that cheating behaviors of the examinee are avoided, preliminary cheating prevention is carried out, then, a plurality of cameras arranged in the examination room are utilized, all-around monitoring is carried out on all the behaviors of the examinee in the examination process of the examinee, the situation that monitoring dead angles occur due to unreasonable camera setting is avoided, and therefore the opportunity of cheating of individual examinees is given, and the fairness of the examination is damaged. After the camera collects the actions of the examinees, the actions form an action set and are sent to the processor module, the actions of the examinees are judged according to a judgment method preset in the processor module, and the collected action set is compared with the pre-stored cheating behaviors; if the contrast is above the threshold value, judging that the examinee cheats; if the contrast is below the threshold value, the examinee is judged to be abnormal, and finally the judgment result is sent to the reminding module to make corresponding reminding, so that the fairness and the justness of the examination are ensured as much as possible, and a healthy and normal examination environment is created.

Preferably, as an improvement, the pre-stored facial features of the examinees are facial features of all examinees collected and stored in the data storage unit before examination.

The facial features of all examinees are collected and stored before the examination in consideration of the change of the facial features of people in the growth process, so that identity comparison original documents are provided during the examination, the real-time property of the facial features of the examination can be guaranteed, the comparison speed in the matching process can be guaranteed, the identity authentication time is shortened, the phenomena of congestion and the influence on the examination process caused by the fact that the examinees still authenticate the identities after the examination starts are avoided.

Preferably, as an improvement, the preset determination method is to compare the collected action set with a pre-stored cheating behavior, determine that the examinee cheats if the contrast is greater than a threshold, and determine that the examinee is not abnormal if the contrast is less than the threshold.

The cheating behaviors used as comparison are prestored in the system in advance, so that the system can directly call data for comparison, meanwhile, the difference of the cheating behaviors is considered, a threshold value of contrast is set, a reasonable error range is provided, and the accuracy of judgment on the cheating behaviors is guaranteed to the maximum extent.

Preferably, as an improvement, the pre-stored cheating actions include head turning actions, upper body deviation actions, little landing actions, and out-of-seat actions.

Through the experience and big data analysis of the previous examination, the cheating behaviors are divided into a head turning behavior, an upper body deviating behavior, a small copying behavior and a seat leaving behavior, so that the cheating behaviors are wider in range, the comprehensiveness of the judgment of the cheating behaviors is guaranteed, meanwhile, the judgment process is accelerated through comparison of the prestored cheating behaviors, the working efficiency of the system is guaranteed, the treatment is timely carried out when the cheating behaviors appear, and the fairness of the examination is guaranteed.

The invention also provides an online automatic invigilation method based on face recognition, which comprises the following steps:

step S1, starting an online automatic invigilation system before a preset time period from the beginning of an examination, confirming the identity of an examinee by a face recognition module before the examinee enters an examination room, and reminding the module to send out an examination seat number corresponding to the examinee when the identity of the examinee is confirmed to be correct;

step S2, after meeting the requirement of examination starting time and the examinee starts moving a pen to answer questions, the on-line automatic invigilating system starts invigilating work, the action collection module collects the actions made by the examinee in the examination process in real time and forms an action set, and then the action set is sent to the processor module;

step S3, after receiving the action set, the processor module judges the action of the examinee according to a preset judgment method, and if the preset judgment condition is met and the contrast ratio of the collected action set and the pre-stored cheating behavior characteristic picture is above a threshold value, the corresponding examinee is judged to have cheating behavior; otherwise, judging that the corresponding examinee is not abnormal; after the judgment result is obtained, the processor module sends the judgment result to the reminding module;

and step S4, the reminding module receives the judgment result and then reminds according to a preset reminding mode.

The method has the advantages that the overall invigilation method is simple and clear in logic, after the identity of the examinee is checked through the face recognition technology, the camera of the motion acquisition module is used for monitoring in the whole process of the examination, the acquired motion is sent to the processor module to judge the motion of the examinee according to the preset judgment method, whether the motion made by the examinee in the examination process is cheating or self habitual behavior can be distinguished in a key mode, psychological pressure brought to the examinee by misjudgment is avoided from influencing examination results, meanwhile, it is guaranteed that no cheating behavior is omitted, and fairness of the examination is guaranteed.

Preferably, as an improvement, the preset reminding mode is that if the examinee is judged to have cheating behaviors, a voice reminding is sent out in the examination room through a loudspeaker; and if the examinee is judged to be abnormal, sending remark reminding to a background.

Through the judgment of the system, if the examinee is determined to have cheating behaviors, the voice prompt is sent out in the examination room through the loudspeaker, so that the effects of preventing the cheating behaviors and warning are achieved; if the examinee is judged to have no abnormity, the examinee makes other action behaviors in the examination process, though the action behaviors are not determined as cheating behaviors by the system, for the sake of insurance, remark reminding is sent to the invigilator through the background, the examinee is focused in the subsequent examination process, the cheating behaviors are avoided, and the examination fairness is ensured.

Preferably, as an improvement, the voice prompt is to play a voice containing the prompt information; the reminding information is a sentence of 'please pay attention to you have cheating'.

The examinee cheating behavior is stopped by voice 'please pay attention to and you have cheating behavior', and meanwhile the voice warning device plays a role in warning other examinees to eliminate cheating ideas of other examinees and avoids the cheating behavior appearing again in the examination process.

Preferably, as an improvement, the cheating act comprises,

head steering behavior, wherein the left-right steering angle of the examinee's head exceeds the seat center line by 45 degrees;

the upper half body deviating behavior is that the left-right deviation angle of the upper half body of the examinee exceeds the seat center line by 30 degrees;

small reading behavior, the hand of the examinee leaves the seat, and the head droop angle exceeds 60 degrees;

and (4) leaving the seat, and leaving the current examination seat by the body of the examinee.

Through practical tests, the test paper of the test taker beside the test taker can be seen when the left-right steering angle of the head exceeds 45 degrees serving as a center line and the left-right deviation angle of the upper half body exceeds 30 degrees serving as a seat center line, so that cheating is realized; meanwhile, according to the characteristics of small reading behaviors, the examinees need to take out the small reading with hands and look at the small reading with heads lowered, the examinees are set to leave the seats with hands, the head droop angle exceeds 60 degrees, the cheating is determined, and finally, if the examinees leave the current examination seats, the cheating is uniformly determined no matter what conditions, all possible cheating behaviors are avoided through the careful classification, and the fairness of the examination is guaranteed.

Preferably, as a modification, in step S3, when the high-definition camera captures a hand-lifting motion within the capture time period during motion capture, the capturing motions corresponding to all images within the capture time period are compliant motions.

Considering that an emergency exists in the examination process, if actions which are indicated by the hands of the examinee in advance are collected in the collection time period, a series of actions which are allowed are not in the consideration range of cheating actions, so that the humanization of the examination rules can be embodied, and the fairness of the examination can be ensured in the rules.

Preferably, as an improvement, the prompting module sends out an examination seat number corresponding to the examinee, and the display screen displays the seat coordinates of the examinee and a walking route from the current position of the examinee to the position of the seat coordinates to the examinee passing through identity authentication.

The seat coordinates of the examinees passing through identity verification are displayed through the display screen, so that the examinees can know the seat coordinates in advance, time waste is avoided, the examinees can find the positions one by one, meanwhile, the walking routes of the examinees to the seats are displayed in a simulation mode on the display screen, the examinees can be in place according to the work, the intellectualization of the system is reflected, the opportunities of cheating of the examinees can be reduced, and the fairness of the examination is guaranteed.

Drawings

Fig. 1 is a system diagram of an embodiment of an online automatic invigilation system and method based on face recognition.

Fig. 2 is a schematic flow chart of an embodiment of the online automatic invigilation system and method based on face recognition of the invention.

Detailed Description

The following is further detailed by way of specific embodiments:

the reference numbers in the drawings of the specification include: the device comprises a face recognition module 1, a processor module 2, an action acquisition module 3, a reminding module 4, an analysis and judgment unit 5, a data storage unit 6, a display screen 7 and a loudspeaker 8.

The first embodiment is as follows:

this embodiment is substantially as shown in figure 1: the online automatic invigilation system based on the face recognition comprises a processor module 2, a face recognition module 1, an action acquisition module 3 and a reminding module 4, wherein the face recognition module 1, the action acquisition module 3 and the reminding module 4 are respectively connected with the processing module;

the face recognition module 1 is used for extracting facial information features of the examinees to obtain a feature set and sending the feature set to the processor module 2 for identity authentication of the examinees;

the action acquisition module 3 comprises a plurality of high-definition cameras arranged in the examination room and is used for acquiring actions of the examinee in the examination process, forming an action set and then sending the action set to the processor module 2 for identifying cheating behaviors;

the processor module 2 comprises an analysis and judgment unit 5 and a data storage unit 6;

the analysis and judgment unit 5 is used for receiving the feature set, comparing the feature set with the facial features of all examinees collected before examination and stored in the data storage unit 6, and confirming the identities of the examinees; after receiving the action set, the analysis and judgment unit 5 compares the collected action set picture with the pre-stored cheating action pictures, such as head turning action, upper body deviation action, small copying action and leaving seat; if the contrast similarity of the pictures is more than 90%, determining that the examinee cheats; if the contrast similarity of the pictures is below 90%, judging that the examinee is not abnormal;

the data storage unit 6 is used for storing the data of the face recognition module 1, the data of the action acquisition module 3, the facial feature data of the examinee and the judgment method data;

the processor module 2 is used for analyzing and judging the action set made by the examinee in the examination process, obtaining a judgment result and then sending the judgment result to the reminding module 4;

the reminding module 4 comprises a display screen 7 and a loudspeaker; the reminding module 4 is used for receiving the judgment result and reminding according to a preset reminding mode.

An online automatic invigilation method based on face recognition is shown in the attached figure 2 and comprises the following steps:

step S1, starting an online automatic invigilation system before a preset time period from the beginning of an examination, confirming the identity of an examinee by a face recognition module before the examinee enters an examination room, and reminding the module to send out an examination seat number corresponding to the examinee when the identity of the examinee is confirmed to be correct;

step S2, after meeting the requirement of examination starting time and the examinee starts moving a pen to answer questions, the on-line automatic invigilating system starts invigilating work, the action collection module collects the actions made by the examinee in the examination process in real time and forms an action set, and then the action set is sent to the processor module;

step S3, after receiving the action set, the processor module judges the action of the examinee according to a preset judgment method, if the preset judgment condition is met, the contrast between the collected action set and the pre-stored cheating behavior characteristic picture, namely the contrast similarity is more than a threshold value, the corresponding examinee is judged to have cheating behavior; otherwise, judging that the corresponding examinee is not abnormal; after the judgment result is obtained, the processor module sends the judgment result to the reminding module;

and step S4, the reminding module receives the judgment result and then reminds according to a preset reminding mode.

The face recognition technology is utilized to identify the examinees entering the examination room, so that the examinees can be ensured to participate in the examinees for the first cheating prevention; utilize a plurality of cameras of setting in the examination room afterwards, carry out 360 degrees no dead angles control to all examinee's action in examinee's examination in-process, do not give any examinee's chance of practising fraud, guarantee the fairness of examination. After the camera of the action acquisition module 3 acquires the actions of the examinee in the examination process, the actions form an action set and are sent to the processor module 2, the actions of the examinee are analyzed and judged according to a judgment method preset in the processor module 2, and the acquired action set is compared with the pre-stored cheating actions; if the contrast is above the threshold value, judging that the examinee cheats; if the contrast is below the threshold, the examinee is judged to be abnormal, finally, the judgment result is sent to the reminding module 4 to make corresponding reminding, the fairness and the justness of the examination are ensured as much as possible, a healthy and normal examination environment is created, and the learning achievement of each examinee who tries to learn is respected.

The specific implementation process of this embodiment is as follows:

firstly, open the automatic system of invigilating on line 40 minutes before the beginning of the examination, the face identification module 1 passes through high definition digtal camera and gathers examinee's facial image before the examinee gets into the examination room, utilize face identification technique, extract examinee's facial information characteristic, with gather before the examination and store into the facial characteristic of all examinees of data storage unit 6 and carry out the contrast, thereby confirm the identity of examinee, after the identity of examinee confirms errorlessly, remind module 4 to send the pronunciation of the examination seat number that the examinee corresponds through the speaker.

And in the second step, after meeting the requirement of the time of starting the examinee, the loudspeaker sends out a question answering starting instruction, the automatic invigilating system starts the on-line automatic invigilating work, the action collection module 3 collects the actions of the examinee in the examination process in real time through a plurality of high-definition cameras arranged in the examination field and forms an action set, and then the action set is sent to the processor module 2.

Thirdly, after the processor module 2 receives the action set, judging the action of the examinee according to a preset judging method, and comparing the collected action set picture with a pre-stored cheating action picture, such as head turning action, wherein the left and right turning angles of the examinee's head exceed the seat center line by 45 degrees; the upper half body deviating behavior is that the left-right deviation angle of the upper half body of the examinee exceeds the seat center line by 30 degrees; small reading behavior, the hand of the examinee leaves the seat, and the head droop angle exceeds 60 degrees; an out-of-seat behavior in which the examinee's body leaves the current examination seat; if the contrast similarity of the pictures is more than 90%, determining that the examinee cheats; if the contrast similarity of the pictures is below 90%, judging that the examinee is not abnormal; and obtaining a judgment result after the judgment is finished, and sending the judgment result to the reminding module 4.

Fourthly, after receiving the judgment result, the reminding module 4 sends out a voice reminding of 'please pay attention to you have cheating behaviors' in the examination room through a loudspeaker if the examinee is judged to have cheating behaviors; and if the examinee is judged to be abnormal, sending a reminding message of 'suspected cheating' of the examinee to a background.

And fifthly, sending a ringing prompt by the loudspeaker after the time reaches the examination ending time to remind all examinees to stop pens immediately to stand up and to go out of the examination room in order, and after all examinees in the examination room leave the examination room, monitoring and stopping running by the on-line automatic invigilation system.

Example two:

this embodiment is substantially the same as the first embodiment, except that: the threshold value of the contrast similarity of the cheating behavior pictures is set to 80%.

The threshold value of the contrast similarity of the pictures is reduced to 80%, the influence on the judgment standard due to the individual difference of the examinees is considered, after the threshold value is reduced, the included action surface is wider, the monitoring strength on the examinees is increased, the condition that the examinees do not cheat and the cheating opportunity are ensured, and the fairness of the examinees is further ensured.

The specific implementation process of this embodiment is the same as that of the first embodiment, except that:

thirdly, after the processor module 2 receives the action set, judging the action of the examinee according to a preset judging method, and comparing the collected action set picture with a pre-stored cheating action picture, such as head turning action, wherein the left and right turning angles of the examinee's head exceed the seat center line by 45 degrees; the upper half body deviating behavior is that the left-right deviation angle of the upper half body of the examinee exceeds the seat center line by 30 degrees; small reading behavior, the hand of the examinee leaves the seat, and the head droop angle exceeds 60 degrees; an out-of-seat behavior in which the examinee's body leaves the current examination seat; if the contrast similarity of the pictures is more than 80%, determining that the examinee cheats; if the contrast similarity of the pictures is below 80%, judging that the examinee is not abnormal; and obtaining a judgment result after the judgment is finished, and sending the judgment result to the reminding module 4.

Example three:

this embodiment is substantially the same as the first embodiment, except that: in the examination process, if an unexpected accident occurs, the examinee can raise a hand to indicate a request to make an action, after the invigilator allows the action, the requested action can be performed, and the system judges that the action is a compliance action.

In combination with the actual examination situation, some emergency situations, such as test paper problems, stomachache of an examinee, pen refill depletion of the examinee and other accidents, are inevitable in the examination process, and at the moment, the examinee can only hold hands for help, so that a series of actions made after the examinee holds the hand and gets the permission of the invigilator are all regarded as compliant actions and are not included in the judgment range of cheating. The intelligent and humanized invigilating system is embodied through the setting, and the intelligent and humanized invigilating system is more suitable for actual examination scenes and has a wider application range.

The specific implementation process of this embodiment is the same as that of the first embodiment, except that:

thirdly, after the processor module 2 receives the action set, judging the action of the examinee according to a preset judging method, and comparing the collected action set picture with a pre-stored cheating action picture, such as head turning action, wherein the left and right turning angles of the examinee's head exceed the seat center line by 45 degrees; the upper half body deviating behavior is that the left-right deviation angle of the upper half body of the examinee exceeds the seat center line by 30 degrees; small reading behavior, the hand of the examinee leaves the seat, and the head droop angle exceeds 60 degrees; an out-of-seat behavior in which the examinee's body leaves the current examination seat; if the contrast similarity of the pictures is more than 90%, determining that the examinee cheats; if the contrast similarity of the pictures is below 90%, judging that the examinee is not abnormal; if an unexpected accident situation occurs in the examination process, the examinee can raise a hand to indicate a request to make an action, the high-definition camera shoots three minutes after the hand raising action, and the shooting actions corresponding to all the collected images are compliance actions; and obtaining a judgment result after the judgment is finished, and sending the judgment result to the reminding module 4.

Example four:

this embodiment is substantially the same as the first embodiment, except that: when an examinee enters an examination room and after the face recognition is passed, the display screen 7 displays the seat coordinates of the examinee and a walking route from the current position of the examinee to the position of the seat coordinates to the examinee who passes the identity authentication.

Because the examination needs to queue up one by one for face recognition before entering the examination room to check the identity information of the examinees, and after the identity is confirmed to be correct through the face recognition, the examinees need to find their seats quickly and sit well, so that the cheating content is prevented from being marked at other positions of the examination room. At the moment, in order to reduce the time for the examinee to find the seat and prevent the examinee from cheating, all seats of the examinee are simulated on the display screen 7, the coordinate of the seat of the current examinee is highlighted, the optimal route for the examinee to walk to the seat of the examinee from the entrance of the examinee is displayed in a simulation mode, the examinee is guided to walk to the seat of the examinee, the efficiency for the examinee to find the seat is improved, the phenomenon of congestion at the entrance of the examinee is avoided, the actions of the examinee are all within the monitoring range, cheating of the examinee is prevented, and fairness of the examinee is guaranteed.

The specific implementation process of this embodiment is the same as that of the first embodiment, except that:

firstly, an online automatic invigilation system is started 40 minutes before an examination begins, a face recognition module 1 collects facial images of an examinee through a high-definition camera before the examinee enters an examination room, a face recognition technology is utilized, facial information characteristics of the examinee are extracted, and the facial images are compared with facial characteristics of all the examinees collected and stored in a data storage unit 6 before the examination, so that the identity of the examinee is confirmed, after the identity of the examinee is confirmed to be correct, a reminding module 4 reminds the examinee to quickly find the own examination seat through simulating and displaying seat coordinates and a walking route of the examinee on a display screen 7.

Example five:

this example is substantially the same as example four, except that: when the system detects that the same examinee has been subjected to two cheating reminders, the system informs the examinee through a loudspeaker that the examinee is disqualified from the examination.

In order to ensure the fairness of the examination and avoid the confirmation of cheating behaviors of the examinees due to system misjudgment, when the system judges the cheating of the examinees for the first time, only a warning is given through a loudspeaker; if the examinee is judged to cheat for the second time by the system in the process of the following examinee, the fact that the examinee cheats is directly identified, the examination score of the examinee is cancelled, and the examinee immediately stops the pen to quickly leave the examination room. The purpose of making no unnecessary use for each examinee and putting no cheating behaviors is achieved to the maximum extent, and the fairness of the examination is guaranteed.

The specific implementation process of this embodiment is the same as that of the fourth embodiment, except that:

fourthly, after receiving the judgment result, the reminding module 4 sends out a voice reminding of 'please pay attention to you have cheating behaviors' in the examination room through a loudspeaker if the examinee is judged to have cheating behaviors; if the examinee is judged to be abnormal, sending remark reminding of 'suspected cheating' of the examinee to a background; if the same examinee is reminded of cheating twice, the examinee is informed of canceling the examination qualification through the loudspeaker, and the examinee immediately stops answering questions and leaves the examination room quickly.

The foregoing is merely an example of the present invention and common general knowledge in the art of designing and/or characterizing particular aspects and/or features is not described in any greater detail herein. It should be noted that, for those skilled in the art, without departing from the technical solution of the present invention, several variations and modifications can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicability of the patent. The scope of the claims of the present application shall be determined by the contents of the claims, and the description of the embodiments and the like in the specification shall be used to explain the contents of the claims.