CN202435528U - Video monitoring system - Google Patents
Video monitoring system Download PDFInfo
- Publication number
- CN202435528U CN202435528U CN2012200208595U CN201220020859U CN202435528U CN 202435528 U CN202435528 U CN 202435528U CN 2012200208595 U CN2012200208595 U CN 2012200208595U CN 201220020859 U CN201220020859 U CN 201220020859U CN 202435528 U CN202435528 U CN 202435528U
- Authority
- CN
- China
- Prior art keywords
- module
- moving target
- input port
- acquisition
- port
- 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
Images
Landscapes
- Alarm Systems (AREA)
Abstract
The utility model discloses a video monitoring system, which comprises an acquisition module, a detection module, a tracking module, an analysis module and an alarming module, wherein the image output port of the acquisition module is connected with the image input port of the detection module; the moving target information output port of the detection module is connected with the moving target information input port of the tracking module; the moving target motion state output port of the tracking module is connected with the moving target motion state input port of the analysis module; and the abnormity signal output port of the analysis module is connected with the abnormity signal input port of the alarming module. With the adoption of the system, the relevant information about an acquired moving target is analyzed intelligently, and the multi-time bottom layer calculation to the same video flow is performed for only one time, so that the rate of report omission and false report about the abnormal actions during the monitoring is reduced. In addition, the system is linked with a security system, thus monitoring personnel can respond immediately to confirm whether the moving target behaves abnormally or not, and further, the accuracy of security monitoring can be improved.
Description
Technical field
The utility model relates to the intelligent video monitoring system technical field, relates in particular to a kind of through the video monitoring system of middleware Technology management and monitoring scene video stream with the monitoring of completion high efficiency smart.
Background technology
Along with the application about intelligent video monitoring system of the universalness and hugeization of supervisory control system, particularly recent years has also obtained developing widely.The development of video analysis technology, multimedia database, artificial intelligence technology now; Intelligent video monitoring has progressively been come into the security protection application market; Intellectualized technology can in time, automatically extract a large amount of useful informations from original video information; Be used for accomplishing the transmission preservation and the retrieval of video, also can drive other data, trigger other behaviors, accomplish the task that manpower is difficult to completion easily.Intellectualized monitoring is one of important directions of monitoring technique development, also is the necessary core technology of setting up the large-scale monitoring network system.Intellectualized technology can solve a lot of problems that exist in the present monitoring, and the combination of computer and image technique makes that image detects automatically, video analysis becomes possibility.Simultaneously intellectualized technology can improve monitoring efficiency greatly, can be from the data of complicacy identification behavior and type, operational order, data and information can be provided.Realize reporting to the police, reminding functions such as concern, intelligent retrieval.The more important thing is that intellectualized technology passes through the video flowing that intelligent video analysis is on-the-spot or write down; Thereby detect suspicious activity, incident or behavior pattern; Can make video monitoring system from search suspect's means afterwards, convert a kind of supplementary means that stops crime to take place into.
But; At present intelligent video monitoring system is to adopt different algorithms that the video flowing from monitoring scene is carried out processing and identification to go out in essence: cross over warning line, get into warning region, drive in the wrong direction, steal, be detained, pace up and down, incidents such as velocity anomaly and density anomaly; And for need discern the anomalous event of a plurality of types the time from Same Scene; Carried out the processing that repeats for many video informations; Increased amount of calculation, a high performance computer only can be handled seldom several roads video stream signal often, particularly need carry out multiple anomalous event when analyzing to one tunnel vision signal; One tunnel vision signal just needs a high performance computer, and these application for large-scale intelligent video monitoring system have all caused very big resistance
The utility model content
The technical problem that the utility model mainly solves provides a kind of video monitoring system, and this system can the intellectual analysis video flowing, and unified management is carried out in the identification and the processing of anomalous event, guarantees that identical process only handles once, reduces the redundancy of system.
For solving the problems of the technologies described above, the technical scheme that the utility model adopts is: a kind of video monitoring system is provided, comprises acquisition module, detection module, tracking module, analysis module and alarm module; Said acquisition module has the output end of image mouth; Detection module has image input port and moving target information outlet; Tracking module has moving target information input port and moving target motion state delivery outlet; Analysis module has moving target motion state input port and abnormal signal delivery outlet, and alarm module has the abnormal signal input port; The output end of image mouth of said acquisition module links to each other with the image input port of detection module; The moving target information outlet of detection module links to each other with the moving target information input port of tracking module; The moving target motion state delivery outlet of tracking module links to each other with the moving target motion state input port of analysis module, and the abnormal signal delivery outlet of analysis module links to each other with the abnormal signal input port of alarm module; Acquisition module is passed detection module with the video flowing of gathering through the output end of image oral instructions; Detection module is used for detecting video flowing and whether has motion and moving target information is exported to tracking module; Supply tracking module to carry out target following, analysis module receives the target state of tracking module, and the configuration of monitoring environment is also analyzed this target whether abnormal behaviour is arranged; Whether there is abnormal behaviour to take place thereby differentiate, if there is abnormal behaviour to take place just to think that alarm module sends warning message.
In order to make this video monitoring system can be applied to different working environments, the acquisition module that the present technique scheme provides can compatible various video stream.Concrete, realize that through this technical scheme acquisition module is provided with first acquisition port, second acquisition port and the 3rd acquisition port; First acquisition port is used for the collection network video flowing; Second acquisition port is used to gather the video flowing that video frequency collection card provides; The 3rd acquisition port is used to gather the file video flowing.
The beneficial effect of the utility model is: this video monitoring system can be applied to complicated scene; Acquisition module is passed detection module with the video flowing of gathering through the output end of image oral instructions; Detection module is used for detecting video flowing and whether has motion and moving target information is exported to tracking module; Supply tracking module to carry out target following, analysis module receives the target state of tracking module, and the configuration of monitoring environment is also analyzed this target whether abnormal behaviour is arranged; Whether there is abnormal behaviour to take place thereby differentiate, if there is abnormal behaviour to take place just to send warning message to alarm module.This system is through analysis module, the relevant information of the moving target that collects carried out intellectual analysis, thereby the repeatedly bottom computing of same video flowing is only carried out once, can reduce the failing to report of abnormal behaviour in the monitoring, rate of false alarm.In addition, this system can also open safety-protection system automatically, thereby the monitor staff can be given a response immediately, through this supervisory control system, confirms whether the behavior of moving target is abnormal behaviour, improves the accuracy rate of safety monitoring.
Description of drawings
Fig. 1 is the structure chart of the video monitoring system of the utility model embodiment.
Embodiment
By the technology contents, the structural feature that specify the utility model, realized purpose and effect, know clearly below in conjunction with execution mode and conjunction with figs. and give explanation.
See also Fig. 1; The video monitoring system of the utility model; Comprise acquisition module, detection module, tracking module, analysis module and alarm module; Said acquisition module has the output end of image mouth, and detection module has image input port and moving target information outlet, and tracking module has moving target information input port and moving target motion state delivery outlet; Analysis module has moving target motion state input port and abnormal signal delivery outlet, and alarm module has the abnormal signal input port; The output end of image mouth of said acquisition module links to each other with the image input port of detection module; The moving target information outlet of detection module links to each other with the moving target information input port of tracking module; The moving target motion state delivery outlet of tracking module links to each other with the moving target motion state input port of analysis module, and the abnormal signal delivery outlet of analysis module links to each other with the abnormal signal input port of alarm module.
Concrete, the acquisition module of this video monitoring system is used to gather the video flowing that current needs are analyzed; Detection module; Be used for detecting video flowing and whether have moving target,, carry out adaptive background modeling if there is moving target; Preliminary acquisition prospect; Prospect is carried out morphology handle, fill the target area and reduce the medium and small fragment zone of image, utilize HSV space shadow detection method that the shade of target is removed acquisition final objective zone; Tracking module is used to utilize particle filter or Kalman filtering tracking that moving target is followed the tracks of; Analysis module is used to analyze the shape information and the kinetic characteristic of the moving target of being followed the tracks of, and utilizes decentralization and area information to analyze abnormal conditions, with moving object classification, accomplishes abnormal behaviour identification Treatment Analysis; Alarm module is used for when having abnormal conditions in the video flowing, starting emergency alarm according to abnormal behaviour identification Treatment Analysis data.
Below will specify the course of work of the video monitoring system of present technique scheme:
Video acquisition module is gathered the current video flowing that needs analysis earlier, and concrete, acquisition module is provided with first acquisition port, second acquisition port and the 3rd acquisition port; First acquisition port is used for the collection network video flowing; Second acquisition port is used to gather the video flowing that video frequency collection card provides; The 3rd acquisition port is used to gather the file video flowing, and the video flowing of acquisition module collection comprises the network video stream that DVR provides, video flowing that video frequency collection card provides and file video flowing.Acquisition module passes to detection module through the image delivery outlet with image information, detects in the video streaming image whether have moving target through detection module, and is concrete; This detection module carries out adaptive background modeling to video flowing, tentatively obtains prospect, prospect is carried out morphology handle; Fill the target area; And the zone of the fractionlet in the minimizing image, utilize HSV space shadow detection method that the shade of target is removed acquisition final objective zone, behind the acquisition motion target area; Moving target information is passed to tracking module; The background modeling of detection module is based on adaptive background modeling method, and the average that adopts the preceding N two field picture get video flowing is B as a setting, and establishing the K frame image data is I
K, corresponding background image is B
K, detection module with context update is:
Bk+1＝a*Ik+(1-a)*Bk
Detection module obtains foreground data Fk and is based on following formula acquisition:
Fk+1＝|Ik-Bk|＞T
Wherein obtaining of threshold value T can adaptively obtain.Calculate the difference image Dk=|Ik-Ik-1| of two two field pictures.Big young pathbreaker's difference image of image is divided into plurality of sub-regions, and the size of each subregion is m*m (0＜m＜15).To i sub regions Dki calculated threshold Ti.
Wherein F (k-1) i is foreground data and the corresponding data of Dki piece that the k-1 frame detects.The pixel of prospect more then is judged to prospect relatively in F (k-1) i, and vice versa.
Final threshold value T is:
Among the step S200, in order to obtain complete motion target area and the empty scape of minimizing, the prospect Fk that adopts morphology methods that background modeling is extracted here carries out in the place, back.If the size of structural elements S is s*s (0＜s＜10), the foreground image after the processing is Fk ':
Wherein, ⊙ is the corrosion operation;
is expansive working.
When removing shade; Adopted HSV space shadow detection method; Basic principle is a same object (background or prospect) at the tone of shadow region and nonshaded area is approximate consistent, and shade makes mainly that brightness changes in this zone, and dash area is necessarily low than the brightness of background.
Be transformed on the HSV space at current frame image Ik and background image Bk.Extract tone H and brightness V with the corresponding zone of prospect Fk ' of initial examination and measurement.
Wherein, F "
kBe the final moving target identification image that detects.FALSE representes background information, and TRUE representes foreground information.H (B
k(F '
k)) expression and subregion chrominance information on the HVS space of the corresponding background image Bk of prospect.T1 and t2 are threshold parameter, and value is 0＜t1＜10,0＜t2＜40 respectively
Tracking module is used particle filter or two kinds of trackings of Kalman filtering are followed the tracks of moving target.
Analysis module is used for moving target is classified; Concrete, moving target is divided into: people, car and chaotic disturbance, principle of classification are that the people is shown as strip on image and area occupied is less relatively; Car is a squarish, when a plurality of targets are concentrated relatively, thinks chaotic disturbance.Calculate the area O of target
S, promptly the number of the shared pixel of target is tentatively judged target according to criterion Os＞th1 (300＜th1＜900).When Os＞th1 thinks that then the target that detects is a car, otherwise combining target area Os and girth Oc further judge target.When (Oc) 2/Os＞th2 (0.2＜th2＜0.9), then detect target and behave, otherwise think that target is a car.
If total N target (N＞1); The center of gravity of i target is Oi, and the center of gravity of prospect is calculated the decentralization Od of target according to center of gravity for
:
When Od＜th3 (0＜th3＜50), then think the disturbance that causes confusion.
Tracking module is used particle filter or two kinds of trackings of Kalman filtering are followed the tracks of moving target; Then the motion state with moving target passes to analysis module; The shape information and the kinetic characteristic of the moving target that the analysis module analysis is followed the tracks of; Utilize decentralization and area information to analyze abnormal conditions, with moving object classification, concrete; The abnormal behaviour classification comprises: cross over warning line, get into the warning line zone, drive in the wrong direction, be detained, pace up and down, velocity anomaly and density anomaly; This analysis module can also be accomplished the configuration of identification monitoring environment, and the configuration of this monitoring environment comprises: the position of warning line, orientation, length and size, the retrograde deflection of moving target etc.
Seven kinds of abnormal behaviours are divided into into three major types according to needed target information: the abnormal behaviour of a) differentiating according to moving object classification information has density anomaly; B) abnormal behaviour of differentiating according to motion target tracking information has the warning region of entering and crosses over warning line; C) abnormal behaviour of differentiating according to the united information of moving object classification and motion target tracking, have retrograde, be detained, pace up and down and velocity anomaly etc.
(1) density anomaly: when the moving target content is chaotic disturbance, then think density anomaly to occur;
(2) cross over warning line and entering warning region: when sampling video flow, just warning line locality and length information are collected; Calculate the center of gravity of tracking target according to the target of being followed the tracks of; Warning line is converted into the line segment equation; When forwarding the center of gravity of target to another side on one side by the line segment equation (looking the judgement face that warning line is a belt restraining), then think occur unusual.The four edges of warning region (only considering regular domain here) is converted into four linear equations, and whether the center of gravity of judging target according to four linear equations is in warning region.
(3) drive in the wrong direction: driving in the wrong direction is meant that mainly the car direction of going is opposite with assigned direction; The direction of driving in the wrong direction, is judged when moving target is not car according to the data analysis that moving object classification carried out in the analysis of carrying out just having accomplished when video flowing is gathered data; Then stop abnormality detection; According to movement locus, use the fitting a straight line Moving Target, calculated line slope angle θ ' to the Tracking Estimation moving target of moving target; When deflection θ and the straight inclined angle θ ' of regulation satisfy following condition, then think occur retrograde.
mod(θ-θ′+360，360)＜t
θ
Wherein mod () is a modulo operation, and t θ is a threshold value, and span is 0＜t θ＜30
(4) pace up and down
Pace up and down and mainly be meant people's reciprocating motion.Judge when carrying out moving object classification whether target is the people.The center of gravity of track human target when moving target is followed the tracks of.Consider the view data of M (M＞100) frame.If the target center of gravity of k frame is Ok, satisfies then to be judged to when descending condition and pace up and down:
Wherein, O
K+i-O
kBe vector, tp is that threshold value has determined the length of pacing up and down.
(5) velocity anomaly
Velocity anomaly is whether inspection vehicle exceeds the speed limit.Judge through classification of motions whether target is car.When the relative position of car between two frames is big, then think velocity anomaly (thinking that here car is a rectilinear motion).The selection of threshold value is determined by the scene restrictive condition.
Alarm module can be discerned the Treatment Analysis data according to abnormal behaviour, when having abnormal conditions in the video flowing, starts emergency alarm.This alarm module is through the driver of the warning device of exploitation bottom, and is concrete, comprises the driving of note cat, the driving of warning lamp signal, sound card driving etc., receive to think that the type of alarm of setting reaches relevant additional information, as: phone number, alarm sound etc.; Receive and transfer the corresponding driving program after the anomalous event of video intelligent analysis module and report to the police, in addition, this alarm module can be preserved each warning message, the convenient inquiry of reporting to the police.
The above is merely the embodiment of the utility model; Be not thus the restriction the utility model claim; Every equivalent structure transformation that utilizes the utility model specification and accompanying drawing content to be done; Or directly or indirectly be used in other relevant technical fields, all in like manner be included in the scope of patent protection of the utility model.
Claims (2)
1. a video monitoring system is characterized in that: comprise acquisition module, detection module, tracking module, analysis module and alarm module;
Said acquisition module has the output end of image mouth; Detection module has image input port and moving target information outlet; Tracking module has moving target information input port and moving target motion state delivery outlet; Analysis module has moving target motion state input port and abnormal signal delivery outlet, and alarm module has the abnormal signal input port;
The output end of image mouth of said acquisition module links to each other with the image input port of detection module; The moving target information outlet of detection module links to each other with the moving target information input port of tracking module; The moving target motion state delivery outlet of tracking module links to each other with the moving target motion state input port of analysis module, and the abnormal signal delivery outlet of analysis module links to each other with the abnormal signal input port of alarm module.
2. video monitoring system according to claim 1 is characterized in that: acquisition module is provided with first acquisition port, second acquisition port and the 3rd acquisition port;
First acquisition port is used for the collection network video flowing;
Second acquisition port is used to gather the video flowing that video frequency collection card provides;
The 3rd acquisition port is used to gather the file video flowing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012200208595U CN202435528U (en) | 2012-01-17 | 2012-01-17 | Video monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012200208595U CN202435528U (en) | 2012-01-17 | 2012-01-17 | Video monitoring system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN202435528U true CN202435528U (en) | 2012-09-12 |
Family
ID=46784933
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012200208595U Expired - Fee Related CN202435528U (en) | 2012-01-17 | 2012-01-17 | Video monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN202435528U (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103079052A (en) * | 2012-12-25 | 2013-05-01 | 南京航空航天大学 | Grid-type demarcation method based on video monitoring early-warning area of mobile phone client |
| CN105306912A (en) * | 2015-12-07 | 2016-02-03 | 成都比善科技开发有限公司 | Intelligent cat-eye system triggering shooting based on luminous intensity and distance detection |
| CN105741467A (en) * | 2016-04-25 | 2016-07-06 | 美的集团股份有限公司 | Safety monitoring robot and robot safety monitoring method |
| CN107623842A (en) * | 2017-09-25 | 2018-01-23 | 西安科技大学 | A kind of video monitoring image processing system and method |
| CN109313847A (en) * | 2016-06-07 | 2019-02-05 | 罗伯特·博世有限公司 | Method, apparatus and system for wrong road driver identification |
| CN110324574A (en) * | 2018-03-29 | 2019-10-11 | 京瓷办公信息***株式会社 | Monitoring system |
| CN111986373A (en) * | 2020-09-09 | 2020-11-24 | 重庆电子工程职业学院 | Security system for training place |
-
2012
- 2012-01-17 CN CN2012200208595U patent/CN202435528U/en not_active Expired - Fee Related
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103079052A (en) * | 2012-12-25 | 2013-05-01 | 南京航空航天大学 | Grid-type demarcation method based on video monitoring early-warning area of mobile phone client |
| CN105306912A (en) * | 2015-12-07 | 2016-02-03 | 成都比善科技开发有限公司 | Intelligent cat-eye system triggering shooting based on luminous intensity and distance detection |
| CN105306912B (en) * | 2015-12-07 | 2018-06-26 | 成都比善科技开发有限公司 | Intelligent peephole system based on luminous intensity and apart from detection triggering camera shooting |
| CN105741467A (en) * | 2016-04-25 | 2016-07-06 | 美的集团股份有限公司 | Safety monitoring robot and robot safety monitoring method |
| CN105741467B (en) * | 2016-04-25 | 2018-08-03 | 美的集团股份有限公司 | A kind of security monitoring robot and robot security's monitoring method |
| CN109313847A (en) * | 2016-06-07 | 2019-02-05 | 罗伯特·博世有限公司 | Method, apparatus and system for wrong road driver identification |
| US11315417B2 (en) | 2016-06-07 | 2022-04-26 | Robert Bosch Gmbh | Method, device and system for wrong-way driver detection |
| CN107623842A (en) * | 2017-09-25 | 2018-01-23 | 西安科技大学 | A kind of video monitoring image processing system and method |
| CN110324574A (en) * | 2018-03-29 | 2019-10-11 | 京瓷办公信息***株式会社 | Monitoring system |
| CN110324574B (en) * | 2018-03-29 | 2021-06-22 | 京瓷办公信息***株式会社 | Monitoring system |
| CN111986373A (en) * | 2020-09-09 | 2020-11-24 | 重庆电子工程职业学院 | Security system for training place |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102547244A (en) | Video monitoring method and system | |
| CN202435528U (en) | Video monitoring system | |
| Singh et al. | Real-time anomaly recognition through CCTV using neural networks | |
| CN102663452B (en) | Suspicious act detecting method based on video analysis | |
| Myrans et al. | Automated detection of faults in sewers using CCTV image sequences | |
| CN103517042A (en) | Nursing home old man dangerous act monitoring method | |
| CN106355162A (en) | Method for detecting intrusion on basis of video monitoring | |
| CN102254394A (en) | Antitheft monitoring method for poles and towers in power transmission line based on video difference analysis | |
| CN104392464A (en) | Human intrusion detection method based on color video image | |
| CN103281518A (en) | Multifunctional networking all-weather intelligent video monitoring system | |
| CN117035419B (en) | Intelligent management system and method for enterprise project implementation | |
| CN113642403B (en) | Crowd abnormal intelligent safety detection system based on edge calculation | |
| Malhi et al. | Vision based intelligent traffic management system | |
| CN117437599B (en) | Pedestrian abnormal event detection method and system for monitoring scene | |
| CN113379099A (en) | Machine learning and copula model-based highway traffic flow self-adaptive prediction method | |
| CN115410324A (en) | Car as a house night security system and method based on artificial intelligence | |
| CN117876966A (en) | Intelligent traffic security monitoring system and method based on AI analysis | |
| CN103152558A (en) | Intrusion detection method based on scene recognition | |
| CN105741503A (en) | Parking lot real time early warning method under present monitoring device | |
| CN106128105B (en) | A kind of traffic intersection pedestrian behavior monitoring system | |
| CN110278285A (en) | Intelligent safety helmet remote supervision system and method based on ONENET platform | |
| CN202904792U (en) | Intelligent visualized alarm system | |
| Rahangdale et al. | Event detection using background subtraction for surveillance systems | |
| CN112032567A (en) | Buried gas pipeline leakage risk prediction system | |
| Masood et al. | Identification of anomaly scenes in videos using graph neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| 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: 20120912 Termination date: 20200117 |