CN105959624B - Examination hall monitoring data processing method and its automatic invigilator's system of realization - Google Patents
Examination hall monitoring data processing method and its automatic invigilator's system of realization Download PDFInfo
- Publication number
- CN105959624B CN105959624B CN201610286408.9A CN201610286408A CN105959624B CN 105959624 B CN105959624 B CN 105959624B CN 201610286408 A CN201610286408 A CN 201610286408A CN 105959624 B CN105959624 B CN 105959624B
- Authority
- CN
- China
- Prior art keywords
- image
- block
- violation
- area
- seating area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/182—Level alarms, e.g. alarms responsive to variables exceeding a threshold
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Emergency Management (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of examination hall monitoring data processing method and its automatic invigilator's systems of realization, the system that this method is realized includes monitoring end module, server module, video data preprocessing module, the video monitoring picture for obtaining to monitoring end module samples, and the image after sampling is pre-processed, and divide seating area；Identification and alarm module, violation movement in seating area for identification, when identifying the movement more than preset threshold, the rank that calibration acts in violation of rules and regulations, and the rank and violation are acted the associated data packet and be sent to server, while being sent to monitor supervision platform for the seating area information being related to is acted in violation of rules and regulations；Monitor supervision platform shows monitored picture, and the information sent according to identification and alarm module, highlights the seating area of violation.This method is fast to the recognition speed of violation movement, and discrimination is high, and automatic invigilator's system facilitates monitor supervision platform personnel to observe, and the quick communication with Field Force.
Description
Technical field
The present invention relates to figure action, behavior monitoring method and system more particularly to one kind in a kind of fixed scene to answer
Automatic invigilator's system of examination hall monitoring data processing method and its realization in examination hall.
Background technique
With the continuous development of video identification field technology, now, by video surveillance applications in invigilator's environment, to reduce
The manpower consumption of invigilator, comprehensive, the accuracy for improving invigilator, become an important application direction in video identification field.
But in the prior art, that there are no a kind of structures is simple, resource cost is few, while again can be accurately to invigilator
Invigilator's system of object progress behavior judgement.The prior art is toward contact using one-to-one formula video common in nonpaperlized examination
Monitor mode, and this mode had not only needed a large amount of terminal device, but cannot overall monitor examinee well movement, while
Biggish psychological burden is brought to examinee, is unfavorable for the normal performance of examinee.
In addition, in the patent application for example application No. is CN201110135345.4, by the way that invigilator will be divided on examination hall
Region, the fixation in region of taking an examination and zone of action, and the region according to the situation that is abnormal, decide whether those exceptions
Situation carries out alarm or prompting processing, and this technical solution is only the improvement in terms of region division, and for whether can
It is enough that the methods of accurate judgement, reduction erroneous judgement are carried out to abnormal behaviour, then without proposing any substantive scheme.For another example applying
Number for CN201010537314.7 patent application in, then be by judging whether to occur to the overall view monitoring in monitored picture
Unusual condition, and different grade and corresponding processing mode are set to unusual condition in advance, thus monitoring unusual condition
When, it can be performed corresponding processing in server end, still, not provide how to judge different unusual conditions in this application
Grade, and how foundation video image carries out situation judgement automatically, these core links are not provided with any technological improvement
Scheme.
The above-mentioned prior art haves the defects that the method not judged well examination hall deliberate action, and
The problems such as resource consumption is big fails to propose that one kind preferably realizes the method invigilated automatically.
Summary of the invention
In view of this, the embodiment of the present invention provides a kind of examination hall monitoring data processing method, method includes the following steps:
S1: when reaching the test time, start the video monitoring picture for obtaining monitoring system in real time；
S2: the video monitoring picture data stream is sampled；
S3: pre-processing the video data after sampling, and examinee seat is carried out region according to preset regional scope
It divides, obtains each seating area；
S4: action recognition is carried out to the image behind division region, when identifying the movement more than preset threshold, is determined as
It acts in violation of rules and regulations；
S5: when identify act in violation of rules and regulations when, the video pictures of one preset duration of start recording and the video pictures
Time started information, and server is sent to using the video pictures and the time started information as data packet；
S6: the seating area where acting in violation of rules and regulations is focused on display in monitor supervision platform, and issues prompting.
Preferably, the difference in areas of the specific region of the continuous interframe of monitoring image, and preset area difference threshold value are sought, institute is based on
Difference in areas and difference in areas threshold value are stated, violation movement is preset as multiple alert levels；It is each alarm at monitor supervision platform end
Rank setting focuses on display mode accordingly.
Preferably, the action recognition in the step S4 includes at least one kind below: standing, movement of squatting down；It turns one's head, turn
Body movement；Frequent movement of bowing；Arm substantially wobbling action etc..This system of recognition accuracy in to(for) such movement is high.
It preferably, further comprise examining a block partiting step, which includes: before the step S1
S01: one time threshold k 1 of setting, before obtaining examinee's admission, monitor video data of the duration within K1；
S02: sampling the monitor video data, obtains N number of sampled images therein；
S03: each sampled images are filtered, and obtain the gray level image of filtered sampled images；
S04: edge processing is carried out to each above-mentioned gray level image, obtains edge image, and the horizontal line in detection image
With vertical line；
S05: image is divided into W1 sub-block, the sum of horizontal line quantity and vertical line quantity in every piece of statistics Z, when mutual
When the Z of the adjacent sub-block is greater than a threshold value Z1, same block is arranged in the sub-block and is marked；The face of the sub-block
Product is less than or equal to each to examine the 1/4 of the practical shared area in the picture in position；
S06: merging the sub-block with same block label, obtains in each sampled images and examines a block
Image；
S07: comparing in each sampled images and examine a block, when the division for examining a block has differences, obtains
The two most width of the identical quantity of a block of examining divided examine a block image；Same in a block image position is examined when two width is examined
When examining the division of block and having differences, compare the Z for examining a block having differences in two images, retain that Z value is biggish to be examined
Position block image, to examine a block in the image, as seating area.
Moreover it is preferred that in this method, the seating area before step S1 is determined, can also by user input equipment,
Regional assignment of the user in a certain examination hall monitored picture is received, as the seating area paid close attention to, which can be with
Selected in picture by the way of region using such as user by mouse drag, can also by set a fixed size window,
The window is extracted in the different zones in picture by user, to determine whole seating areas etc..
Preferably, in step S04, the grey level histogram of gray level image is obtained first, and obtains the maximum peak in histogram
Value and minimum peak filter out gray value in image and are more than or equal to peak-peak and the pixel less than or equal to minimum peak, obtain
To edge-detected image；
Secondly, carrying out edge detection to edge-detected image to described, edge image is obtained.
Preferably, Face datection and character contour detection are carried out to the video data after the sampling in examination process first；
Secondly, comparing in the video data after sampling, the face area change of two continuous frames image and character contour face
Product variation；Above-mentioned area change is two field pictures area absolute value of the difference；
An at least first threshold L1, and an at least second threshold S1 are set, when face area change L >=L1 and character contour
When area change S >=S1, it is set as highest alarm level；As L < L1 and S >=S1, it is set as the second alarm level；When L >=
When L1 and S < S1, it is set as third alarm level.
Preferably, when triggering alarm, the corresponding alarm level of each data packet is recorded, as when time datagram of alarm
Text is sent to server.
Preferably, the data message further includes the video clips index in the data packet；The data message can answer
For in the real time monitoring of low bandwidth or the real time monitoring of cell phone；The data message is also used to drive monitor in real time
Scene shows switching action.
Preferably, the Face datection and character contour detect method particularly includes:
Firstly, the image to the video data after sampling carries out gray processing processing, gray level image is obtained, and to gray level image
Enhanced, specifically in the following way:
Wherein, f (x, y) indicates that the gray value of original image, g (x, y) indicate that the gray value of enhancing image, h1 indicate to meetWhen maximum gradation value, h2 indicate meetWhen minimum gradation value, hist [] indicate present image gray value
Histogram, N indicate the total pixel number of present image；
Secondly, carrying out region detection, specific detection mode to enhanced image are as follows:
Wherein, f (xi,yj) grey scale pixel value of the expression in the M1*M2 neighborhood centered on (x, y), M1, M2 indicate neighborhood
Length and width, C indicates offset, and when carrying out Face datection, C takes C1, and when carrying out character contour detection, C takes C2, the C,
The constant that C1, C2 are positive.
In addition, the present invention also provides one kind, invigilator's system, applicable method as described above, the system include: automatically
End module, including at least one monitoring camera are monitored, for obtaining the video monitoring picture in examination hall in real time；
Server module, for receiving the number by acting associated video pictures in violation of rules and regulations and the video pictures time started is constituted
According to packet；
Video data preprocessing module, the video monitoring picture for obtaining to monitoring end module sample, and
Image after sampling is pre-processed, and divides seating area；
Identification and alarm module, the violation movement in seating area, dynamic more than preset threshold when identifying for identification
When making, the rank that calibration acts in violation of rules and regulations, and the rank and violation are acted the associated data packet and be sent to server,
The seating area information being related to will be acted in violation of rules and regulations is sent to monitor supervision platform simultaneously；
Monitor supervision platform, the video monitoring picture sent for showing monitoring end module, and according to identification and alarm module
The information of transmission highlights the seating area for acting be related in violation of rules and regulations in monitored picture.
Preferably, the system also includes seating area demarcating modules, for the input according to user, in monitoring image
The seating area of calibration concern in advance.
Preferably, the system also includes automatic seating area demarcating modules, for the preset duration before examinee's admission
Video monitoring data, sampled, extract the monitored picture of a preset quantity, according to the seating areas of those monitored pictures
Identification, obtains the seating area in monitored picture.
Preferably, the specific method for obtaining the seating area in monitored picture includes:
S01: one time threshold k 1 of setting, before obtaining examinee's admission, monitor video data of the duration within K1；
S02: sampling the monitor video data, obtains N number of sampled images therein；
S03: each sampled images are filtered, and obtain the gray level image of filtered sampled images；
S04: edge processing is carried out to each above-mentioned gray level image, obtains edge image, and the horizontal line in detection image
With vertical line；
S05: image is divided into W1 sub-block, the sum of horizontal line quantity and vertical line quantity in every piece of statistics Z, when mutual
When the Z of the adjacent sub-block is greater than a threshold value Z1, same block is arranged in the sub-block and is marked；The face of the sub-block
Product is less than or equal to each to examine the 1/4 of the practical shared area in the picture in position；
S06: merging the sub-block with same block label, obtains in each sampled images and examines a block
Image；
S07: comparing in each sampled images and examine a block, when the division for examining a block has differences, obtains
The two most width of the identical quantity of a block of examining divided examine a block image；Same in a block image position is examined when two width is examined
When examining the division of block and having differences, compare the Z for examining a block having differences in two images, retain that Z value is biggish to be examined
Position block image, to examine a block in the image, as seating area.
Preferably, the video data preprocessing module further comprises profile detection module, for detecting seating area
In facial contour and character contour；And
The human face region area and character contour area in image after seeking detection, and by Face datection area and people
Object contour area is sent to the identification and alarm module.
Preferably, the system also includes radio receiving transmitting modules, and the data for that will identify and alarm module is sent are sent
To wireless terminal；And receive the control information from wireless terminal.
Preferably, the system also includes feedback module, receives the control information of wireless terminal, when the control information table
Show current alerts belong to false alarm or it is processed when, eliminate monitor supervision platform on corresponding seating area is highlighted.
Compared with prior art, technical solution of the present invention structure is simple, and calculates the image recognition of monitored picture and processing
Method speed is fast, can meet the needs of real time monitoring well；And by way of image sampling, reduce at later image
Data volume during reason；The present invention has also taken into account the prompting of monitor supervision platform side, and real-time to the information of field personnel
Sharing functionality and information feedback, improve the operational efficiency of monitoring system.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the examination hall monitoring system of the embodiment of the present invention；
Fig. 2 is the examination hall monitoring data processing method flow chart of the embodiment of the present invention.
Specific embodiment
A kind of application program recommended method of the embodiment of the present invention and device are described in detail with reference to the accompanying drawing.It should
Clear, described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based in the present invention
Embodiment, all other embodiment obtained by those of ordinary skill in the art without making creative efforts, all
Belong to the scope of protection of the invention.
Those skilled in the art should know it is further that following specific embodiments or specific embodiment, which are the present invention,
The set-up mode of series of optimum explaining specific summary of the invention and enumerating, and being between these set-up modes can be mutual
In conjunction with or it is interrelated use, unless clearly proposing some of them or a certain specific embodiment or embodiment party in the present invention
Formula can not be associated setting or is used in conjunction with other embodiments or embodiment.Meanwhile following specific embodiment or
Embodiment is only as the set-up mode optimized, and not as the understanding limited the scope of protection of the present invention.
Embodiment 1:
As shown in Fig. 2, providing a kind of examination hall monitoring data processing method, this method in one embodiment of the present of invention
The following steps are included:
S1: when reaching the test time, start the video monitoring picture for obtaining monitoring system in real time；
S2: the video monitoring picture data stream is sampled；
S3: pre-processing the video data after sampling, and examinee seat is carried out region according to preset regional scope
It divides, obtains each seating area；
S4: action recognition is carried out to the image behind division region, when identifying the movement more than preset threshold, is determined as
It acts in violation of rules and regulations；
S5: when identify act in violation of rules and regulations when, the video pictures of one preset duration of start recording and the video pictures
Time started information, and server is sent to using the video pictures and the time started information as data packet；
S6: the seating area where acting in violation of rules and regulations is focused on display in monitor supervision platform, and issues prompting.
It, can be using the specific region difference in areas between continuous sampling image, as prison in a specific embodiment
The judgment basis of control movement, specifically, seeking the difference in areas of the specific region of the continuous interframe of monitoring image, and preset area is poor
Threshold value is based on the difference in areas and difference in areas threshold value, and violation movement is preset as multiple alert levels；At monitor supervision platform end
Mode is focused on display accordingly for the setting of each alert levels.
In a specific embodiment, the action recognition in the step S4 includes at least one kind below: standing, squats
Lower movement；It turns one's head, touch turn；Frequent movement of bowing；Arm substantially wobbling action etc..For the knowledge of such movement in this system
Other accuracy rate is high.
It, can be in advance automatically in monitoring area before being monitored after being set the exam in a specific embodiment
Seating area delimited, i.e., further comprises examining a block partiting step, which includes: before the step S1
S01: one time threshold k 1 of setting, before obtaining examinee's admission, monitor video data of the duration within K1；
S02: sampling the monitor video data, obtains N number of sampled images therein；
S03: each sampled images are filtered, and obtain the gray level image of filtered sampled images；
S04: edge processing is carried out to each above-mentioned gray level image, obtains edge image, and the horizontal line in detection image
With vertical line；
S05: image is divided into W1 sub-block, the sum of horizontal line quantity and vertical line quantity in every piece of statistics Z, when mutual
When the Z of the adjacent sub-block is greater than a threshold value Z1, same block is arranged in the sub-block and is marked；The face of the sub-block
Product is less than or equal to each to examine the 1/4 of the practical shared area in the picture in position；
S06: merging the sub-block with same block label, obtains in each sampled images and examines a block
Image；
S07: comparing in each sampled images and examine a block, when the division for examining a block has differences, obtains
The two most width of the identical quantity of a block of examining divided examine a block image；Same in a block image position is examined when two width is examined
When examining the division of block and having differences, compare the Z for examining a block having differences in two images, retain that Z value is biggish to be examined
Position block image, to examine a block in the image, as seating area.
In addition, the seating area before step S1 determines in another specific embodiment, it can also be defeated by user
Enter equipment, regional assignment of the user in a certain examination hall monitored picture is received, as the seating area paid close attention to, the delimitation side
Formula can be selected in picture by the way of region using such as user by mouse drag, and one fixed size of setting can also be passed through
Window is extracted the window in the different zones in picture by user, to determine whole seating areas etc..
In a specific embodiment, in step S04, the grey level histogram of gray level image is obtained first, and obtain straight
Peak-peak and minimum peak in square figure filter out gray value in image and are more than or equal to peak-peak and are less than or equal to minimum peak
Pixel, obtain to edge-detected image；
Secondly, carrying out edge detection to edge-detected image to described, edge image is obtained.
In a specific embodiment, first to the video data after the sampling in examination process, Face datection is carried out
And character contour detection；
Secondly, comparing in the video data after sampling, the face area change of two continuous frames image and character contour face
Product variation；Above-mentioned area change is two field pictures area absolute value of the difference；
An at least first threshold L1, and an at least second threshold S1 are set, when face area change L >=L1 and character contour
When area change S >=S1, it is set as highest alarm level；As L < L1 and S >=S1, it is set as the second alarm level；When L >=
When L1 and S < S1, it is set as third alarm level.The setting of the threshold value is only used as a preferred mode, also can be set more
More threshold value, so that alarm level is arranged to more grades.
In a specific embodiment, when triggering alarm, the corresponding alarm level of each data packet is recorded, as working as
The data message of secondary alarm, is sent to server.
In a specific embodiment, the data message further includes the video clips index in the data packet；Institute
Stating data message can be applied in the real time monitoring of low bandwidth or the real time monitoring of cell phone；The data message is also used to reality
When driving monitor scene show switching action.
In a specific embodiment, the Face datection and character contour detection method particularly includes:
Firstly, the image to the video data after sampling carries out gray processing processing, gray level image is obtained, and to gray level image
Enhanced, specifically in the following way:
Wherein, f (x, y) indicates that the gray value of original image, g (x, y) indicate that the gray value of enhancing image, h1 indicate to meetWhen maximum gradation value, h2 indicate meetWhen minimum gradation value, hist [] indicate present image gray scale
Value histogram, N indicate the total pixel number of present image；
Secondly, carrying out region detection, specific detection mode to enhanced image are as follows:
Wherein, f (xi,yj) grey scale pixel value of the expression in the M1*M2 neighborhood centered on (x, y), M1, M2 indicate neighborhood
Length and width, C indicates offset, and when carrying out Face datection, C takes C1, and when carrying out character contour detection, C takes C2, the C,
The constant that C1, C2 are positive.For example, when being detected to human face region, C1 value can be arranged it is lower, by face
Region is distinguished with background and environmental area as wide as possible, similarly, C2 can be arranged to higher, with maximum magnitude capsule
Include human body contour outline.
Embodiment 2:
In yet another embodiment, applicable strictly according to the facts as shown in Figure 1, the present invention also provides a kind of automatic invigilator's system
Method described in example 1 is applied, which includes:
End module, including at least one monitoring camera are monitored, for obtaining the video monitoring picture in examination hall in real time；
Server module, for receiving the number by acting associated video pictures in violation of rules and regulations and the video pictures time started is constituted
According to packet；
Video data preprocessing module, the video monitoring picture for obtaining to monitoring end module sample, and
Image after sampling is pre-processed, and divides seating area；
Identification and alarm module, the violation movement in seating area, dynamic more than preset threshold when identifying for identification
When making, the rank that calibration acts in violation of rules and regulations, and the rank and violation are acted the associated data packet and be sent to server,
The seating area information being related to will be acted in violation of rules and regulations is sent to monitor supervision platform simultaneously；
Monitor supervision platform, the video monitoring picture sent for showing monitoring end module, and according to identification and alarm module
The information of transmission highlights the seating area for acting be related in violation of rules and regulations in monitored picture.
In a specific embodiment, the system also includes seating area demarcating modules, for according to the defeated of user
Enter, demarcates the seating area of concern in advance in monitoring image.At this point, system may include user's input module, to receive use
The input at family, the input can be using conventional mechanical input devices or touch input device etc..
In a specific embodiment, the system also includes automatic seating area demarcating modules, are used for examinee's admission
The video monitoring data of a preceding preset duration, is sampled, and the monitored picture of a preset quantity is extracted, according to these monitoring
The seating area of picture identifies, obtains the seating area in monitored picture.
In a specific embodiment, the specific method for obtaining the seating area in monitored picture includes:
S01: one time threshold k 1 of setting, before obtaining examinee's admission, monitor video data of the duration within K1；
S02: sampling the monitor video data, obtains N number of sampled images therein；
S03: each sampled images are filtered, and obtain the gray level image of filtered sampled images；
S04: edge processing is carried out to each above-mentioned gray level image, obtains edge image, and the horizontal line in detection image
With vertical line；
S05: image is divided into W1 sub-block, the sum of horizontal line quantity and vertical line quantity in every piece of statistics Z, when mutual
When the Z of the adjacent sub-block is greater than a threshold value Z1, same block is arranged in the sub-block and is marked；The face of the sub-block
Product is less than or equal to each to examine the 1/4 of the practical shared area in the picture in position；
S06: merging the sub-block with same block label, obtains in each sampled images and examines a block
Image；
S07: comparing in each sampled images and examine a block, when the division for examining a block has differences, obtains
The two most width of the identical quantity of a block of examining divided examine a block image；Same in a block image position is examined when two width is examined
When examining the division of block and having differences, compare the Z for examining a block having differences in two images, retain that Z value is biggish to be examined
Position block image, to examine a block in the image, as seating area.
In a specific embodiment, the video data preprocessing module further comprises profile detection module, is used
Facial contour and character contour in detection seating area；And
The human face region area and character contour area in image after seeking detection, and by Face datection area and people
Object contour area is sent to the identification and alarm module.
In a specific embodiment, the system also includes radio receiving transmitting module, for that will identify and alarm module
The data of transmission, are sent to wireless terminal；And receive the control information from wireless terminal.
In a specific embodiment, the system also includes feedback module, receives the control information of wireless terminal, when
The control information indicate current alerts belong to false alarm or it is processed when, eliminate on monitor supervision platform to corresponding seating area
It highlights.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the program can be stored in a computer-readable storage medium
In, the program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, read-only memory (Read-Only Memory, ROM) or random access memory (Random Access
Memory, RAM) etc..
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by those familiar with the art, all answers
It is included within the scope of the present invention.Therefore, protection scope of the present invention should be subject to the protection scope in claims.
Claims (6)
1. a kind of examination hall monitoring data processing method, which is characterized in that the described method comprises the following steps:
S1: when reaching the test time, start the video monitoring picture for obtaining monitoring system in real time；
S2: the video monitoring picture data stream is sampled；
S3: pre-processing the video data after sampling, and examinee seat is carried out region division according to preset regional scope,
Obtain each seating area；
S4: action recognition is carried out to the image behind division region, when identifying the movement more than preset threshold, is determined as in violation of rules and regulations
Movement；
Seek the difference in areas of the specific region of the continuous interframe of monitoring image, and preset area difference threshold value, based on the difference in areas and
Violation movement is preset as multiple alert levels by difference in areas threshold value；
It is that the setting of each alert levels focuses on display mode accordingly at monitor supervision platform end；
S5: when identifying movement in violation of rules and regulations, the beginning of the video pictures and the video pictures of one preset duration of start recording
Temporal information, and server is sent to using the video pictures and the time started information as data packet；
S6: the seating area where acting in violation of rules and regulations is focused on display in monitor supervision platform, and issues prompting；
It further comprise examining a block partiting step, which includes: before the step S1
S01: one time threshold k 1 of setting, before obtaining examinee's admission, monitor video data of the duration within K1；
S02: sampling the monitor video data, obtains N number of sampled images therein；
S03: each sampled images are filtered, and obtain the gray level image of filtered sampled images；
S04: edge processing, acquisition edge image, and the horizontal line in detection image are carried out to each above-mentioned gray level image and erected
Line；
S05: being divided into W1 sub-block for image, the sum of horizontal line quantity and vertical line quantity in every piece of statistics Z, when being immediately adjacent to each other
The Z of sub-block when being greater than a threshold value Z1, same block is arranged in the sub-block and is marked；The area of the sub-block is small
The 1/4 of the practical shared area in the picture in position is each examined in being equal to；
S06: merging the sub-block with same block label, obtains in each sampled images and examines a block diagram
Picture；
S07: comparing in each sampled images and examine a block, when the division for examining a block has differences, obtains and divides
Two most width of the identical quantity of a block of examining examine a block image；Same in a block image examining for position is examined when two width is examined
When position block division has differences, compare the Z for examining a block having differences in two images, retains the biggish area Kao Wei of Z value
Block image, to examine a block in the image, as seating area.
2. the method as described in claim 1, which is characterized in that the action recognition in the step S4 includes at least below one
Kind: it stands, movement of squatting down；It turns one's head, touch turn；Frequent movement of bowing；Arm substantially wobbling action.
3. according to the method described in claim 1, it is characterized by: obtaining the intensity histogram of gray level image first in step S04
Figure, and the peak-peak in histogram and minimum peak are obtained, it filters out gray value in image and is more than or equal to peak-peak and is less than
Equal to the pixel of minimum peak, obtain to edge-detected image；
Secondly, carrying out edge detection to edge-detected image to described, edge image is obtained.
4. according to the method described in claim 1, it is characterized by: first to the video data after the sampling in examination process,
Carry out Face datection and character contour detection；
Secondly, comparing in the video data after sampling, the face area change and character contour area of two continuous frames image become
Change；Above-mentioned area change is two field pictures area absolute value of the difference；
An at least first threshold L1, and an at least second threshold S1 are set, when face area change L >=L1 and character contour area
When changing S >=S1, it is set as highest alarm level；As L < L1 and S >=S1, it is set as the second alarm level；As L >=L1 and S
When < S1, it is set as third alarm level.
5. according to the method described in claim 4, it is characterized in that, the Face datection and the specific method of character contour detection
Are as follows:
Firstly, the image to the video data after sampling carries out gray processing processing, gray level image is obtained, and carry out to gray level image
Enhancing, specifically in the following way:
Wherein, f (x, y) indicates that the gray value of original image, g (x, y) indicate that the gray value of enhancing image, h1 indicate to meetWhen maximum gradation value, h2 indicate meetWhen minimum gradation value, hist [] indicate present image gray value
Histogram, N indicate the total pixel number of present image；
Secondly, carrying out region detection, specific detection mode to enhanced image are as follows:
Wherein, f (xi,yj) grey scale pixel value of the expression in the M1*M2 neighborhood centered on (x, y), the length of M1, M2 expression neighborhood,
Width, C indicate offset, and when carrying out Face datection, C takes C1, and when carrying out character contour detection, C takes C2, described C, C1, C2
The constant being positive.
6. a kind of automatic invigilator's system, applicable method as claimed in claim 1 to 5, which is characterized in that the system
Include:
End module, including at least one monitoring camera are monitored, for obtaining the video monitoring picture in examination hall in real time；
Server module, for receiving the data by acting associated video pictures in violation of rules and regulations and the video pictures time started is constituted
Packet；
Video data preprocessing module, the video monitoring picture for obtaining to monitoring end module sample, and to adopting
Image after sample is pre-processed, and divides seating area；
Identification and alarm module, the violation movement in seating area for identification, when identifying the movement more than preset threshold,
The rank that calibration acts in violation of rules and regulations, and the rank and violation are acted the associated data packet and be sent to server, simultaneously
The seating area information being related to will be acted in violation of rules and regulations is sent to monitor supervision platform；
Monitor supervision platform, the video monitoring picture sent for showing monitoring end module, and sent according to identification and alarm module
Information, the seating area for acting be related in violation of rules and regulations is highlighted in monitored picture；
Automatic seating area demarcating module is sampled, is mentioned for the video monitoring data of the preset duration before examinee's admission
The monitored picture of a preset quantity is taken, is identified according to the seating area to those monitored pictures, obtains the seat in monitored picture
Region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610286408.9A CN105959624B (en) | 2016-05-03 | 2016-05-03 | Examination hall monitoring data processing method and its automatic invigilator's system of realization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610286408.9A CN105959624B (en) | 2016-05-03 | 2016-05-03 | Examination hall monitoring data processing method and its automatic invigilator's system of realization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105959624A CN105959624A (en) | 2016-09-21 |
| CN105959624B true CN105959624B (en) | 2019-04-05 |
Family
ID=56914550
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610286408.9A Expired - Fee Related CN105959624B (en) | 2016-05-03 | 2016-05-03 | Examination hall monitoring data processing method and its automatic invigilator's system of realization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105959624B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107181934A (en) * | 2017-06-26 | 2017-09-19 | 长春嘉诚信息技术股份有限公司 | Instruction transmission method and system in a kind of court trial process |
| CN107977614A (en) * | 2017-11-23 | 2018-05-01 | 长春嘉诚信息技术股份有限公司 | A kind of court's trial behavior method for early warning, apparatus and system |
| CN110119862A (en) * | 2018-02-07 | 2019-08-13 | 中国石油化工股份有限公司 | Based on enterprise it is external enter factory personnel smoke danger classes diagnostic method |
| CN108535899B (en) * | 2018-04-13 | 2021-06-01 | Tcl华星光电技术有限公司 | TFT-LCD data stream detection method, electronic device and computer readable storage medium |
| CN108647509A (en) * | 2018-05-11 | 2018-10-12 | 北京北信源信息安全技术有限公司 | A kind of method and device for preventing sensitive document from revealing |
| CN110659542B (en) * | 2018-06-29 | 2022-10-21 | 杭州海康威视数字技术股份有限公司 | Monitoring method and device |
| CN108989770A (en) * | 2018-08-31 | 2018-12-11 | 温州大学 | A kind of education monitoring system |
| CN111510664A (en) * | 2019-01-30 | 2020-08-07 | 武汉景思睿创新科技有限公司 | Examination room monitoring system based on video image recognition |
| CN110059619B (en) * | 2019-04-17 | 2022-05-13 | 阿波罗智联(北京)科技有限公司 | Automatic alarm method and device based on image recognition |
| CN110032992B (en) * | 2019-04-25 | 2023-05-23 | 沈阳图为科技有限公司 | Examination cheating detection method based on gestures |
| CN110135282B (en) * | 2019-04-25 | 2021-06-04 | 沈阳图为科技有限公司 | Examinee return plagiarism cheating detection method based on deep convolutional neural network model |
| CN111586356B (en) * | 2020-05-08 | 2021-07-20 | 国家邮政局邮政业安全中心 | Violation monitoring method, device and system, electronic equipment and storage medium |
| CN111651132B (en) * | 2020-06-02 | 2023-03-24 | 马鞍山芯乔科技有限公司 | Picture-in-picture synchronous display system based on visual inspection picture |
| CN112087603B (en) * | 2020-08-31 | 2022-10-11 | 浙江蓝鸽科技有限公司 | Intelligent examination room supervision method |
| CN116486324B (en) * | 2023-03-08 | 2024-02-09 | 深圳市旗畅科技有限公司 | Subway seat trampling behavior detection method, device, equipment and storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI588778B (en) * | 2012-01-17 | 2017-06-21 | 國立臺灣科技大學 | Activity recognition method |
| CN103136510B (en) * | 2013-01-17 | 2016-08-03 | 京东方科技集团股份有限公司 | A kind of invigilator's system and method |
| CN103971381A (en) * | 2014-05-16 | 2014-08-06 | 江苏新瑞峰信息科技有限公司 | Multi-target tracking system and method |
| CN105491354A (en) * | 2016-01-18 | 2016-04-13 | 广州爱九游信息技术有限公司 | Video monitoring method and video monitoring platform |
-
2016
- 2016-05-03 CN CN201610286408.9A patent/CN105959624B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN105959624A (en) | 2016-09-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105959624B (en) | Examination hall monitoring data processing method and its automatic invigilator's system of realization | |
| CN106056079B (en) | A kind of occlusion detection method of image capture device and human face five-sense-organ | |
| WO2021036636A1 (en) | Vibration detection method and apparatus for lifting device, server and storage medium | |
| US20140177946A1 (en) | Human detection apparatus and method | |
| CN109711318B (en) | Multi-face detection and tracking method based on video stream | |
| US20110090358A1 (en) | Image pickup apparatus, information processing apparatus, and information processing method | |
| CN107240092A (en) | A kind of image blur detection method and device | |
| CN113096158A (en) | Moving object identification method and device, electronic equipment and readable storage medium | |
| CN103475800B (en) | Method and device for detecting foreground in image sequence | |
| Alioua et al. | Eye state analysis using iris detection based on Circular Hough Transform | |
| KR20090044957A (en) | Theft and left baggage survellance system and meothod thereof | |
| CN109698906A (en) | Dithering process method and device, video monitoring system based on image | |
| CN109905670A (en) | A kind of multi-stage platform monitoring system | |
| CN207233038U (en) | Face is called the roll and number system | |
| CN103473779A (en) | Method and device for detecting stripe interference in image | |
| CN109711313B (en) | Real-time video monitoring method for identifying sewage poured into river channel | |
| CN105740752A (en) | Method and system for sensitive image filtering | |
| KR20080057500A (en) | System and method for tracking movement images of people | |
| JP2007312271A (en) | Surveillance system | |
| CN108563997A (en) | It is a kind of establish Face datection model, recognition of face method and apparatus | |
| CN102542304B (en) | Region segmentation skin-color algorithm for identifying WAP (Wireless Application Protocol) mobile porn image | |
| CN115984973A (en) | Human body abnormal behavior monitoring method for peeping-proof screen | |
| CN106611417A (en) | A method and device for classifying visual elements as a foreground or a background | |
| CN110390313A (en) | A kind of violent action detection method and system | |
| Baek et al. | Detecting functional connectivity changes in fMRI data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20191023 Address after: Yuen Road Ya Dong Qixia District of Nanjing City, Jiangsu province 210003 New District No. 9 Patentee after: Nanjing Post & Telecommunication Univ. Address before: 212006 Zhenjiang City, Jiangsu province Jingkou District is East Street No. 80 Xuefu hengmeijiayuan Building 5 room 404 Patentee before: Fang Junjie |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190405 Termination date: 20210503 |