CN105163078B - A kind of screen removes intelligent video monitoring system - Google Patents
A kind of screen removes intelligent video monitoring system Download PDFInfo
- Publication number
- CN105163078B CN105163078B CN201510551491.3A CN201510551491A CN105163078B CN 105163078 B CN105163078 B CN 105163078B CN 201510551491 A CN201510551491 A CN 201510551491A CN 105163078 B CN105163078 B CN 105163078B
- Authority
- CN
- China
- Prior art keywords
- screen
- image
- edge
- module
- removes
- 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.)
- Active
Links
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to screens to monitor field, provides a kind of screen and removes intelligent video monitoring system, in the monitoring of particular place, judging whether computer screen removes the occlusion area of monitor setting, if screen removes occlusion area, carries out warning reminding.The present invention includes image collection module, screen edge detection module, screen angle detection module and alarm module, first log into the camera of monitor, gather the Y passages of image in camera, convolution is carried out to the area-of-interest for obtaining image using Sobel Operator, again linear edge is extracted with Hough transformation, utilize specific matrix convolution angle of departure, the presence of screen is determined according to the logical place relation at edge and angle, if there is the situation for meeting screen area to occur, then system judges that screen removes the region blocked set, which alarms.Monitoring system of the present invention has the advantages of accuracy is high, and real-time is good.
Description
Technical field
The invention belongs to screens to monitor field, is related in special occasions such as research institute, enterprises, computer screen is examined in real time
It surveys, it is ensured that in the occlusion area that computer screen is set in monitor, so as to realizing the secrecy to related data;It is specially a kind of
Screen removes intelligent video monitoring system.
Background technology
With the development of science and technology and the enhancing of people's protective awareness, the units such as research institute, company are to the guarantor of data
Close work is also gradually also increasingly paid attention to.The physical base that monitoring is every profession and trade key sector or important place is monitored in real time
Plinth, administrative department can obtain valid data, image or acoustic information by it, the process of paroxysmal abnormality event be carried out timely
Monitoring and memory, to provide efficiently, in time commander and height, arrangement police strength, settle a case.
Although monitor can monitor the situation of office space in real time, the letter in the computer screen lighted in office
Breath also tends to leak by monitor.Although monitor can set occlusion area to block computer screen, for
The artificial situation that computer screen is removed to the occlusion area set but can not discovery in time handle, i.e., monitor can not be real-time
Whether detection computer screen is moved out of the situation of occlusion area.Therefore, the present invention proposes that a kind of screen removes intelligent video monitoring
Whether system removes default occlusion area for intelligent measurement computer screen.
The content of the invention
It is an object of the invention to be directed in the monitoring of specific occasion, a kind of screen is provided and removes intelligent video monitoring system
System for judging whether computer screen removes the occlusion area of monitor setting, if screen removes occlusion area, is alarmed
It reminds.Compared to traditional monitor, occlusion area is set, the present invention computer screen can be removed the situation of occlusion area into
Row real-time monitoring alarming has real-time and pinpoint accuracy, and the information avoided in computer screen is leaked by monitor, realizes
Monitor maintains secrecy under specific occasion.
To achieve the above object, the technical scheme is that:
A kind of screen removes intelligent video monitoring system, including image collection module, screen edge detection module, screen angle
Detection module and alarm module,
Described image acquisition module is logs in the camera of monitor, for obtaining image information in monitor in real time:
The screen edge detection module, which is used to obtain in the area-of-interest of image from image collection module, screens screen
Edge line;
Screen angle detection module, which is used to obtain from image collection module, obtains screen angle in the area-of-interest of image;
The alarm module is used to carry out logical combination judgement to screen edge straight line and screen angle, meets screen shape i.e.
Assert that screen removes, and then system alarm.
In the present invention, described image acquisition module obtains the gray-scale map that image information in monitor is image in real time, and right
The area-of-interest of image carries out histogram equalization processing.
In the present invention, after the screen edge detection module obtains image, first with Sobel (Sobel) operator to figure
The area-of-interest of picture carries out the marginal information that convolution obtains image, Hough (Hough) is recycled to convert extraction edge line, really
Determine the position of screen edge straight line.
Further, the screen edge detection module screens concretely comprising the following steps for screen edge straight line:
Step 1. extracts the edge of image, and note A is original image information, and Sx and Sy are respectively to be calculated to horizontal Sobel
Son, Gx and Gy are the image intensity value of longitudinal direction and transverse edge detection respectively, then have Gx=Sx*A,Gy=Sy* A, then image
The Grad G of each pixel:Image slices vegetarian refreshments is made trade-offs according to predetermined threshold value;If Grad is less than threshold
Value, zero is arranged to by its gradient;If Grad is more than or equal to threshold value, marginal point is regarded as, so as to fulfill edge detection;
Step 2. obtains marginal information according to step 1 and carries out Hough (Hough) conversion, and throwing is weighted to conversion gained straight line
Ticket obtains the highest conversion line segment of confidence level, is the edge line of screen.
In the present invention, the detailed process that screen angle detection module obtains screen angle is:The Special matrix of use and its
Transposition, flip matrix carry out convolution to image, obtain four screen angles of screen.
It should be noted that in the present invention for the computer screen lighted, shape is substantially a rectangle, that
Determine computer screen position just only it needs to be determined that its four edges are closed with the position at four angles and the position at edge and angle
System can determine the position of screen.
The beneficial effects of the present invention are:
The present invention proposes a kind of screen and removes intelligent video monitoring system, in the monitoring of particular place, passes through monitoring
Device sets the region (computer screen of office worker) of need for confidentiality block after, the position of computer screen is monitored in real time
To guarantee to find that computer screen removes occlusion area, and warning reminding in time.In monitoring process, the phase of monitor is first logged into
Machine gathers the Y passages (gray-scale map) of image in camera, using Sobel (Sobel) operator to the area-of-interest of acquisition image
Convolution is carried out, then extraction linear edge is converted with Hough (Hough), using specific matrix convolution angle of departure, according to edge and angle
Logical place relation determine the presence of screen, if having meet screen area situation occur, system judge screen remove set
The region blocked put, the system are alarmed.Monitoring system of the present invention has the advantages of accuracy is high, and real-time is good.
Description of the drawings
Fig. 1 removes intelligent video for screen of the present invention and monitors system block diagram.
Fig. 2 removes Hough (Hough) Transformation Graphs in intelligent video monitoring system for screen of the present invention.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.
A kind of screen removes intelligent video monitoring system, including image collection module, screen edge detection module, screen angle
Detection module and alarm module;System module block diagram is as shown in Figure 1:Image collection module is to log in camera, according to corresponding end
Mouth obtains corresponding image information;Screen edge detection module is mainly area-of-interest (ROI) the screening symbol in image
The region of computer screen edge is closed, it is then determined that the marginal position of screen;Screen angle detection module is mainly according in image
Area-of-interest (ROI) screens the region for meeting computer screen angle, it is then determined that the Angle Position of screen;Alarm module is mainly logical
The logical place relation for crossing edge and angle determines the position of screen, and then judges whether screen removes occlusion area, if screen moves
Go out and set occlusion area, then system alarm.Specially:
A. image collection module
In the present invention, the monitoring real-time image information in place is obtained from monitor, it is ensured that detection need to block in real time
Whether computer screen is in the range of the blocking of setting;FOR ALL WE KNOW, the network monitoring device of networkmonitor front end has
There is provided the IP address of oneself, and open specified data communication port;To obtain image in monitor camera, the present invention
This IP address is logged in, goes to access this monitor by corresponding port, so as to obtain the image information in reflection monitoring place in real time.
In view of different indoor illumination conditions and the difference of monitoring camera image quality, the figure obtained from different cameral
As that may be very different in terms of resolution ratio;In order to adapt to different environment, the adaptability of algorithm is improved, the present invention is first
First the area-of-interest of the original image to collecting carries out histogram equalization processing, enhances image resolution ratio, consequently facilitating
Detection;Such processing has good enhancing effect in the case of screen content brightness is not high simultaneously.
B. screen edge detection module
After the gray-scale map for obtaining image, first with Sobel (Sobel) operator to the region of interest in the image of acquisition
Domain carries out the edge that convolution obtains image, and Hough (Hough) is recycled to convert extraction linear edge, determines the position of straight line.
Sobel (Sobel) operator is a kind of first difference operator, includes the matrix of two groups of 3x3, is respectively horizontal and vertical
To, it counts the gradient of a pixel using the Grad of pixel adjacent domain, is then accepted or rejected according to certain threshold value, and then
Obtain the edge of image.Original image, S are represented with A in the present embodimentxAnd SySobel Operator, i.e., specific convolution mask,
GxAnd GyThe longitudinal direction represented respectively and the image of transverse edge detection, then have following relation:
For the right and left frame of screen, respectively with longitudinal detective operators SyAnd its flip horizontal matrix is to the figure that collects
As carrying out convolution, two width testing result figures are obtained, upper and lower side frame is similarly;
For each pixel in image, gradient magnitude G is:
Threshold value, which can be ensured that, is clearly partitioned into marginal information;Since above-mentioned convolution results (gradient) are very big in edge,
In smooth region very little, excessive trend is presented in intermediate region;Grad is arranged to the Grad of slightly below marginal zone, energy by we
The influence of some miscellaneous points and non-frame region for result is filtered out well;Simultaneously because under different light conditions, gradient
Value also has certain variation, we set a threshold to adjustable numerical value, to adapt to the light condition in different monitoring place.At this
In invention, we make trade-offs pixel according to the Grad of the image pixel of acquisition and the threshold value of setting;For the every of image
A pixel, if its Grad is less than threshold value, its gradient is arranged to zero by us;If its Grad is all higher than being equal to threshold value,
Marginal point is then regarded as, so as to fulfill edge detection.
After the marginal information for obtaining image, we determine the edge line of screen using Hough (Hough) conversion;
Hough transform converts the test problems in image space using the point-line duality of image space and Hough parameter spaces
To parameter space;By carrying out simple cumulative statistics in parameter space, then accumulator peak is found in Hough parameter spaces
The method detection of straight lines of value;Thus the test problems of given curve in original image are converted into the peak found in parameter space
Value problem namely detection overall permanence is converted into detection local characteristics.
Become as shown in Fig. 2, the curve S1 and S2 in Hough spaces are rectangular coordinate system midpoint P1 and P2 respectively by Hough
It gets in return, then the intersection point P0 of curve S1 and S2 in Hough spaces then represent the straight line L1 of passing point P1 and P2.We
The four width images obtained above obtained with rope shellfish operator convolution are passed through into Hough transform, two ends of straight line can be obtained
Point coordinates is to get to the straight line for representing screen edge.
Since the straight line quantity that Hough transform obtains is very big, it would be desirable to which straight line obtained by Hough transform is weighted
Ballot (utilizes houghlinesP functions), to obtain the highest conversion line segment of confidence level, filters out and excessively repeats and short and small line
Section.We need to be adjusted ballot threshold value in the process, and transformation results quantity is made to be unlikely to very big and energy generation
Whole section of fringe region of table is preferred, and number of results can not be very little, to avoid important Eigenvector is lost.
C. screen angle detection module
For computer screen, it is understood that determine its position, in addition to edge lines to be determined, four are also predefined
Angle Position, the special matrix R and its transposition and flip matrix that we use herein carry out convolution to image, to extract screen
Curtain angle:
It is the lower right corner for detecting screen for matrix template R, if a certain area-of-interest and the phase after R convolution
Should be maximum, then it is the screen lower right corner to assert it.It is similar, by the overturning of matrix R, transposed matrix we can obtain screen
Its excess-three angle.
D. alarm module
Front we obtain Hough result of variations (the doubtful edge line of screen edge line from the area-of-interest of image
The extreme coordinates of section) and screen Angle Position coordinate.In view of screen sheet as rectangle, parallel four side can be rendered as in video
Shape.We compare the coordinate differences of line segment endpoint and its corresponding angles away from, it may be determined that this group of data whether meet screen feature (such as
Should occur the left end point of top edge and the upper extreme point of left hand edge near the upper left corner);It compares between line segment endpoint and angle point
Slope differences are away from can further determining that whether data are available, (such as the upper left corner should be with left hand edge line segment to the slope in the lower left corner
Slope differences are away from less).We determine the presence of screen according to grid deviation and the satisfactory one group of data of slope deviation.If
There is the situation for meeting screen shape to occur, that is, judge there is screen appearance, i.e. screen has moved out occlusion area, and system will be reported
It is alert to remind.
The above description is merely a specific embodiment, any feature disclosed in this specification, except non-specifically
Narration, can be replaced by other alternative features that are equivalent or have similar purpose;Disclosed all features or all sides
Method or in the process the step of, in addition to mutually exclusive feature and/or step, can be combined in any way.
Claims (4)
1. a kind of screen removes intelligent video monitoring system, including image collection module, screen edge detection module, the inspection of screen angle
Module and alarm module are surveyed,
Described image acquisition module is logs in the camera of monitor, for obtaining image information in monitor in real time:
The screen edge detection module, which is used to obtain in the area-of-interest of image from image collection module, screens screen edge
Straight line;
Screen angle detection module, which is used to obtain from image collection module, obtains screen angle in the area-of-interest of image;
The alarm module is used to carry out logical combination judgement to screen edge straight line and screen angle, meets screen shape and assert
Screen removes, and then system alarm;
The detailed process that screen angle detection module obtains screen angle is:The Special matrix and its transposition of use, flip matrix
Convolution is carried out to image, obtains four screen angles of screen;The Special matrix is:
2. the screen as described in claim 1 removes intelligent video monitoring system, which is characterized in that described image acquisition module is real-time
The gray-scale map that image information in monitor is image is obtained, and histogram equalization processing is carried out to the area-of-interest of image.
3. the screen as described in claim 1 removes intelligent video monitoring system, which is characterized in that the screen edge detection module
After obtaining image, the marginal information that convolution obtains image is carried out to the area-of-interest of image first with Sobel Operator, then
Edge line is extracted using Hough transformation, determines the position of screen edge straight line.
4. the screen as described in claim 3 removes intelligent video monitoring system, which is characterized in that the screen edge detection module
Screening screen edge straight line concretely comprises the following steps:
Step 1. extracts the edge of image, and note A is original image information, Sx and Sy be respectively to the Sobel Operator with transverse direction,
Gx and Gy is the image intensity value of longitudinal direction and transverse edge detection respectively, then has:Gx=Sx*A, Gy=Sy*A, then image
The Grad G of each pixel:Image slices vegetarian refreshments is made trade-offs according to predetermined threshold value;If Grad is less than threshold
Value, zero is arranged to by its gradient;If Grad is more than or equal to threshold value, marginal point is regarded as, so as to fulfill edge detection;
Step 2. obtains marginal information according to step 1 and carries out Hough transformation, is weighted ballot to conversion gained straight line, obtains credible
Highest conversion line segment is spent, is the edge line of screen.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510551491.3A CN105163078B (en) | 2015-09-01 | 2015-09-01 | A kind of screen removes intelligent video monitoring system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510551491.3A CN105163078B (en) | 2015-09-01 | 2015-09-01 | A kind of screen removes intelligent video monitoring system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105163078A CN105163078A (en) | 2015-12-16 |
CN105163078B true CN105163078B (en) | 2018-05-18 |
Family
ID=54803820
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510551491.3A Active CN105163078B (en) | 2015-09-01 | 2015-09-01 | A kind of screen removes intelligent video monitoring system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105163078B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112150493B (en) * | 2020-09-22 | 2022-10-04 | 重庆邮电大学 | Semantic guidance-based screen area detection method in natural scene |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103914830A (en) * | 2014-02-22 | 2014-07-09 | 小米科技有限责任公司 | Straight line detection method and device |
CN104469127A (en) * | 2013-09-22 | 2015-03-25 | 中兴通讯股份有限公司 | Photographing method and photographing device |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4573158B2 (en) * | 2004-02-26 | 2010-11-04 | 日本ビクター株式会社 | Monitoring image generation method and monitoring image generation apparatus |
-
2015
- 2015-09-01 CN CN201510551491.3A patent/CN105163078B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104469127A (en) * | 2013-09-22 | 2015-03-25 | 中兴通讯股份有限公司 | Photographing method and photographing device |
CN103914830A (en) * | 2014-02-22 | 2014-07-09 | 小米科技有限责任公司 | Straight line detection method and device |
Non-Patent Citations (1)
Title |
---|
基于Hough变换的快速矩形检测算法;李强兵等;《微计算机信息》;20071231;第23卷(第11-1期);第248页至第249页 * |
Also Published As
Publication number | Publication date |
---|---|
CN105163078A (en) | 2015-12-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109089160B (en) | Video analysis system and method for food processing violation behaviors of restaurants in colleges and universities | |
CN109308447A (en) | The method of equipment operating parameter and operating status is automatically extracted in remote monitoriong of electric power | |
CN103984961B (en) | A kind of image detecting method for being used to detect underbody foreign matter | |
CN106686377B (en) | A kind of video emphasis area determination method based on deep-neural-network | |
CN106682635B (en) | A kind of smog detection method based on random forest feature selecting | |
CN105744226B (en) | A kind of 1+N rifle ball interlock methods based on camera coordinate system | |
CN109190473A (en) | The application of a kind of " machine vision understanding " in remote monitoriong of electric power | |
CN105675623B (en) | It is a kind of based on the sewage color of sewage mouth video and the real-time analysis method of flow detection | |
CN107240079A (en) | A kind of road surface crack detection method based on image procossing | |
CN107845090A (en) | A kind of silicon chip detection method and silicon chip detection means | |
CN109118548A (en) | A kind of comprehensive intelligent water quality recognition methods | |
US11315374B2 (en) | Detection of fraudulent access at control gates | |
CN102404602A (en) | Vidicon definition detection method based on definition test card | |
CN106530281A (en) | Edge feature-based unmanned aerial vehicle image blur judgment method and system | |
CN110189375A (en) | A kind of images steganalysis method based on monocular vision measurement | |
CN104780361A (en) | Quality evaluation method for urban video monitoring system | |
CN108362382B (en) | A kind of thermal imaging monitoring method and its monitoring system | |
CN112183472A (en) | Method for detecting whether test field personnel wear work clothes or not based on improved RetinaNet | |
CN106682678A (en) | Image angle point detection and classification method based on support domain | |
CN109375068A (en) | A kind of target identification method and device based on ultraviolet imagery corona detection | |
CN109308448A (en) | A method of it prevents from becoming distribution maloperation using image processing techniques | |
WO2023231262A1 (en) | Hoisting wire rope tension measurement method based on visual vibration frequency identification | |
CN105163078B (en) | A kind of screen removes intelligent video monitoring system | |
CN109034038A (en) | A kind of fire identification device based on multi-feature fusion | |
CN111753587A (en) | Method and device for detecting falling to ground |
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 |