CN100555348C - Intelligent video monitoring system of bank self-aid apparatus - Google Patents
Intelligent video monitoring system of bank self-aid apparatus Download PDFInfo
- Publication number
- CN100555348C CN100555348C CNB2007101059341A CN200710105934A CN100555348C CN 100555348 C CN100555348 C CN 100555348C CN B2007101059341 A CNB2007101059341 A CN B2007101059341A CN 200710105934 A CN200710105934 A CN 200710105934A CN 100555348 C CN100555348 C CN 100555348C
- Authority
- CN
- China
- Prior art keywords
- video
- module
- aid apparatus
- bank self
- target
- 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
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
- Alarm Systems (AREA)
Abstract
The invention provides a kind of intelligent video monitoring system of bank self-aid apparatus, described system is made up of with memory module and alarm module video acquisition module, analysis module, video coding.Described analysis module is used for the detection and tracking moving target, eliminates the identification of false target image block and goal behavior; Wherein, described analysis module uses background modeling, prediction search and point to follow the tracks of the tracking that combines, the use scene with described bank self-aid apparatus does not change, be applicable to the indoor and outdoors environment simultaneously, the picture element that point is followed the tracks of under only the foreground area of extracting with background modeling being covered is followed the tracks of, and the result of tracking is the image block sequence of motion.The present invention can eliminate false target and target detection mistake effectively, can adapt to the applied environment of bank self-aid apparatus, is convenient to the differentiation of target and the identification of goal behavior and also can be directly used in warning.
Description
Technical field
The present invention relates to the intelligent video monitoring system technical field, particularly a kind of stable bank self-aid apparatus real time intelligent control system that merges background modeling, prediction search and some tracking results.
The method that the present invention adopts comprises background modeling, target detection, target following, the goal behavior identification under the static scene.Monitored object is the various possible hazardous act that the bank self-aid apparatus periphery takes place.
Background technology
What existing bank self-aid apparatus used is the traditional video surveillance system, not the video monitoring function of intelligence.The traditional video surveillance system comprises video acquisition, mobile detection and compression storage.Video acquisition has used various camera, comprises CCD camera, CMOS camera, infrared camera etc.Mobile detection is to detect the frame that motion is arranged in the video sequence, and not detection and tracking target is not a kind of methods of video analyses of intelligence.The MPEG-4 or the decoding method of standard have H.264 mainly been adopted in compression storage.When the bank self-aid apparatus periphery was caused danger situation, the early warning in time of traditional video surveillance system was normally reported to the police by other modes, and the artificial enquiry monitoring video is collected evidence afterwards then.Simultaneously, when bank self-aid apparatus periphery takes place to violate the behavior of rule, can not initiatively remind the litigant to take care in time and people on every side keeps order.
In the use of bank self-aid apparatus, still there are some unsafe factors, such as going around of the robbery that occurs in the bank cashier machine, a suspect etc.This just needs to detect and record a suspect's image, remind the people that using bank self-aid apparatus, remind security personnel, remind the people of periphery to keep professional order and security distance etc., in time reports to the police when taking preventive measures, keeping order or causing danger situation.Response speed when therefore, the traditional video monitoring system of developing intellectual resource video monitoring system replacement can be eliminated the potential safety hazard when using bank self-aid apparatus and improve dangerous situation and the treatment effeciency of incident.
Bank self-aid apparatus may be placed in outdoor, also may be placed in indoorly, may be placed in the very strong background of illumination variation, also may be placed in the darker background of light application ratio.And present intelligent monitor system adapts to the indoor and outdoors environment by adjusting parameter, tracking performance instability, rate of false alarm height when being applied to bank self-aid apparatus, even can not the tenacious tracking target after the adjustment parameter, this just needs a kind of new intelligent processing method of exploitation, adapts to the applied environment of bank self-aid apparatus.
Existing intelligent video monitoring system generally uses the method for background modeling to detect target, on the basis of detected target, use track algorithm to follow the tracks of then, tracking results relies on the result of target detection, and this is the unsettled reason of performance in the bank self-aid apparatus monitoring just.The present invention has proposed a kind of background modeling and has followed the tracks of the method that combines with point then according to the applied environment of bank self-aid apparatus, can stablize the abnormal conditions that detect the bank self-aid apparatus periphery effectively, and in time remind or warning.
Summary of the invention
The object of the present invention is to provide a kind of intelligent video monitoring system of bank self-aid apparatus.
For solving the problems of the technologies described above, the invention provides a kind of intelligent video monitoring system of bank self-aid apparatus, described system is made up of with memory module and alarm module video acquisition module, analysis module, video coding, the output of described video acquisition module is divided into two-way, one the road to described analysis module, another road is to described video coding and memory module, and described video coding and memory module and described alarm module are controlled in the output of described analysis module respectively, wherein:
Described video acquisition module is used to obtain digital video sequences;
Described analysis module is used for the detection and tracking moving target, eliminates the identification of false target image block and goal behavior; Wherein, described analysis module uses background modeling, prediction search and point to follow the tracks of the tracking that combines, the use scene with described bank self-aid apparatus does not change, be applicable to the indoor and outdoors environment simultaneously, the picture element that point is followed the tracks of under only the foreground area of extracting with background modeling being covered is followed the tracks of, and the result of tracking is the image block sequence of motion;
Described video coding and memory module are stored the video sequence of unusual goal behavior according to the result of analysis module;
Described alarm module is used for the real-time warning message that transmits and carry out from analysis module.
Preferably, described system can expand to network monitoring system.
Preferably, described system can be applicable to the video frequency searching based on behavior, target or incident.
Preferably, described system can be applicable to be equipped with any bank self-aid apparatus of video camera.
In said system of the present invention, target following needs to detect earlier target, then detected result is followed the tracks of.Target detection can be divided into block-based detection and based on the detection of unique point.Target detection mode under the static background comparatively commonly used at present is a background modeling method, method for tracking target has a lot, as Mean Shift, feature Selection, active profile, manifold covariance, the tracking of KLT point, Sift etc., but going back the neither one method can be applicable to all occasions.
The surrounding enviroment of bank self-aid apparatus are ever-changing, and the indoor and outdoor situation has, and also have the situations such as appearance of overlapping, general objective of unexpected variation, shade, the target of light, make that the method with background modeling can not adapt to.The method of background modeling is that each picture element is set up a background model, with this model new image is differentiated then, determine which be the foreground point which be background dot.Form by image block with the target area that the method for background modeling obtains, some under indoor situation, often occur because the false target zone that illumination variation causes.In fact background model is the model of an illumination, is not the model of a target travel, and the prospect that obtains with background modeling method is the result of illumination variation, and not necessarily real target area.The motion of the unique point that the method that point is followed the tracks of is can tracking target apparent in view, it can be used for target detection, also can be used for the tracking of target.If target occupies the bigger ratio of image, and the different piece of target is when existing different motions, and as the swing of people's four limbs, a method of following the tracks of with point just is difficult to correctly detect moving target so.
Intelligent video monitoring system of bank self-aid apparatus is to come out moving object detection and Tracking Recognition, and detection is most important link, the present invention is directed to the applied environment characteristics of bank self-aid apparatus, has developed a kind of stabilizing effective intelligent video monitoring system.
Intelligent video monitoring system of bank self-aid apparatus is made up of four modules: video acquisition module, analysis module, video coding and memory module and alarm module.
Video acquisition module is used to obtain digital video sequences, and compatible different cameras can be as acquisition module as long as can produce the video camera of digital image information.
Analysis module is made up of submodules such as background modeling, foreground extraction, prediction search, feature point tracking, goal behavior identifications.Method with background modeling can obtain the foreground image piece that illumination variation causes, these image blocks may be that target travel causes, also may be that the ambient lighting variation causes.Then the foreground image piece is predicted that the picture element under search and the covering of foreground image piece carries out a tracking, just can eliminate the Vitua limage piece that illumination variation causes, because illumination variation is to the influence of image, show as the variation of brightness of image, do not change the texture of image, the method for following the tracks of with point just can determine foreground image piece that illumination variation causes less than moving.And, the method motion of tracking image piece always that point is followed the tracks of, as long as detect the target image piece, target just can not disappear.Therefore, in conjunction with the method for background modeling and some tracking, help improving the stability and the validity of system.It is real goal or false target that the goal behavior differentiation is used for differentiating, and whether according to the rule of conduct of formulating, it is unusual to differentiate goal behavior.
The video coding is used to store the video sequence that those abnormal conditions occur with memory module, can reduce the requirement to the storage medium capacity like this, improves the efficient of query video.
The information of alarm module receiver, video analysis module, and produce or transmit alerting signal.Alerting signal can be used for reminding the people of periphery to note maintaining vigilance with the prompting litigant toe the mark, also can be used for reporting to the police and handles.
In sum, compare with other intelligent video monitoring systems, the intelligent video monitoring system of bank self-aid apparatus that the present invention proposes has following difference:
1) analysis module has adopted background modeling, prediction search and point to follow the tracks of the method that combines, eliminated the false target that causes owing to illumination variation effectively, and general objective different piece motion inconsistent and the target detection mistake that causes, as the inconsistent target detection mistake that cause of the people in the walking owing to limb motion.The applied environment that has so just in time adapted to bank self-aid apparatus.
2) result who traces into is made up of the image block sequence by obvious change color of motion, is convenient to the differentiation of target and the identification of goal behavior.The motion of target generally is clocklike continuous, shows as the image block of continuous motion, and the motion of false target generally is not continuous, shows as the discontinuous fluid of image block, perhaps the discontinuous appearance of image block.
3) detect tracking results and can be used for reminding the people of periphery to maintain vigilance with the prompting litigant toe the mark, improve the service efficiency of bank self-aid apparatus, also can be directly used in warning.
Description of drawings
Fig. 1 is an intelligent video monitoring system of bank self-aid apparatus of the present invention.
Fig. 2 is an analysis module of the present invention.
Embodiment
The intelligent video monitoring system of bank self-aid apparatus block diagram as shown in Figure 1.Total system is made up of four parts: video acquisition module, analysis module, video coding and memory module and alarm module.
Video acquisition module is used to gather video, compatible different video cameras; Analysis module is used for the detection and tracking moving target, eliminates the false target image block, improves system stability, and goal behavior identification; Video coding and memory module are stored the video sequence of unusual goal behavior according to the result of analysis module; Alarm module is used for transmitting in real time and carrying out warning message.
As shown in Figure 2, analysis module can be divided into five submodules again: background modeling module, foreground extracting module, prediction search module, some tracking module, goal behavior identification module.The background modeling module is used to set up background model.Foreground extracting module is used to extract the foreground area of reflection illumination variation.The prediction search module is according to the position of the target prodiction present frame of former frame, the foreground image piece that search is corresponding.Picture element under the some tracking module covers the prospect of extracting is followed the tracks of.The goal behavior identification module is then analyzed the result of a tracking, extracts real target trajectory, and carries out goal behavior and differentiate.
The concrete implementing procedure of intelligent video monitoring system of bank self-aid apparatus is as follows:
The first, set the zone relevant with early warning, comprise the position and the prewarning area of bank self-aid apparatus itself.
The second, terminal camera acquisition image sequence is divided into two-way, and one the road to analysis module, and another road is to video coding and memory module.
The 3rd, in analysis module, at first set up background model with the method for background modeling, to each frame image to be detected, determine that with the background model of having set up which picture element belongs to prospect, and form the foreground image piece.Then the foreground image piece is done the prediction search of target travel and the foreground image piece picture element under covering and do a trace analysis, determine which image block belongs to the moving image piece, and the image block of motion done continuous tracking on time dimension, form the movement locus of image block.Movement locus according to real goal is continuous hypothesis at last, extracts real moving target, and goal behavior is differentiated.
The 4th, in video coding and memory module,, only store those and unusual video sequence occurs according to the result of video analysis.
The 5th, if violate the goal behavior of setting rule, output alarm signal immediately.
Characteristics of the present invention and effect have:
1) performance of analysis module is stable, and object detecting and tracking has used background modeling, pre-Survey search and point is followed the tracks of the method that combines, the result of object detecting and tracking is not with bank self-aid apparatus Use scenes and changing is applicable to the environment of indoor and outdoors simultaneously. General intelligent video monitoring system What consider is general applied environment, but when being used for bank self-aid apparatus, the existence and stability problem. Silver The application scenario of row Possum is ever-changing, and rainy day, snowy day, daytime, evening, indoor chamber are arranged , light and no light are arranged outward, so general objective and little target etc. are the bank self-aid apparatus intelligent video monitoring System must be able to adapt to very intense light irradiation and environmental evolution, just can be practical, and this just requires the video branch It must be stable analysing.
2) result who traces into is made up of the image block sequence by obvious change color of moving, and is convenient to The differentiation of target and the identification of goal behavior. The motion of target generally is regular continuous, performance Be the image block of continuous motion, the motion of false target generally is not continuous, shows as image block not Continuous motion, the perhaps discontinuous appearance of image block.
3) detect tracking results and can be used for reminding the people of periphery to maintain vigilance with the prompting party toe the mark, Improve the service efficiency of bank self-aid apparatus, also can be directly used in warning.
4) reduced requirement to memory space. Only the video sequence that the abnormal object motion is arranged is stored, Can greatly reduce memory space, be convenient to simultaneously inquire about afterwards, because each video sequence corresponding Individual goal behavior.
5) can be expanded into the video monitoring system of management Network Based. Along with the monitoring camera number Roll up, coming the management and monitoring video storage with computer is extremely urgent demand, adopts based on intelligence The Network Management Model of video analysis is the development trend of monitoring system, also is conducive to monitor video simultaneously Retrieval.
6) realization of system can be one of following three kinds of modes: hardware is realized, software is realized or soft or hard Part is in conjunction with realization.
Claims (4)
1, a kind of intelligent video monitoring system of bank self-aid apparatus, described system is made up of with memory module and alarm module video acquisition module, analysis module, video coding, the output of described video acquisition module is divided into two-way, one the road to described analysis module, another road is to described video coding and memory module, described video coding and memory module and described alarm module are controlled in the output of described analysis module respectively, it is characterized in that:
Described video acquisition module is used to obtain digital video sequences;
Described analysis module is used for the detection and tracking moving target, eliminates the identification of false target image block and goal behavior; Wherein, described analysis module uses background modeling, prediction search and point to follow the tracks of the tracking that combines, the use scene with described bank self-aid apparatus does not change, be applicable to the indoor and outdoors environment simultaneously, the picture element that point is followed the tracks of under only the foreground area of extracting with background modeling being covered is followed the tracks of, and the result of tracking is the image block sequence of motion;
Described video coding and memory module are stored the video sequence of unusual goal behavior according to the result of analysis module;
Described alarm module is used for the real-time warning message that transmits and carry out from analysis module.
2, according to the intelligent video monitoring system of bank self-aid apparatus of claim 1, wherein, described system can expand to network monitoring system.
3, according to the intelligent video monitoring system of bank self-aid apparatus of claim 1, wherein, described system can be applicable to the video frequency searching based on behavior, target or incident.
4, according to the intelligent video monitoring system of bank self-aid apparatus of claim 1, wherein, described system can be applicable to be equipped with any bank self-aid apparatus of video camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007101059341A CN100555348C (en) | 2007-06-01 | 2007-06-01 | Intelligent video monitoring system of bank self-aid apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007101059341A CN100555348C (en) | 2007-06-01 | 2007-06-01 | Intelligent video monitoring system of bank self-aid apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101079171A CN101079171A (en) | 2007-11-28 |
| CN100555348C true CN100555348C (en) | 2009-10-28 |
Family
ID=38906629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2007101059341A Expired - Fee Related CN100555348C (en) | 2007-06-01 | 2007-06-01 | Intelligent video monitoring system of bank self-aid apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100555348C (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103400120A (en) * | 2013-08-02 | 2013-11-20 | 上海泓申科技发展有限公司 | Video analysis-based bank self-service area push behavior detection method |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009246935A (en) * | 2008-03-14 | 2009-10-22 | Sanyo Electric Co Ltd | Image processing device, and imaging device equipped with the same |
| CN102073844A (en) * | 2010-11-10 | 2011-05-25 | 无锡中星微电子有限公司 | Intelligent monitoring system and method |
| CN102209256A (en) * | 2011-05-19 | 2011-10-05 | 北京国铁华晨通信信息技术有限公司 | Video-based method and device for testing intrusion detection performance |
| CN103400148B (en) * | 2013-08-02 | 2015-04-01 | 上海泓申科技发展有限公司 | Video analysis-based bank self-service area tailgating behavior detection method |
| CN103956007B (en) * | 2014-04-27 | 2016-06-15 | 王美金 | Bank abnormity situation detection system based on mobile phone acceleration sensor and gyroscope |
| CN104320609A (en) * | 2014-09-22 | 2015-01-28 | 南开大学 | Low storage monitoring method based on optical correlation |
| CN105957271B (en) * | 2015-12-21 | 2018-12-28 | 中国银联股份有限公司 | A kind of financial terminal safety protecting method and system |
| CN108399691B (en) * | 2017-02-08 | 2021-01-12 | 新谊整合科技股份有限公司 | ATM exception management method and system |
| CN107040759A (en) * | 2017-04-12 | 2017-08-11 | 合肥才来科技有限公司 | Intelligent monitor system applied to bank |
| CN107644519A (en) * | 2017-10-09 | 2018-01-30 | 中电科新型智慧城市研究院有限公司 | A kind of intelligent alarm method and system based on video human Activity recognition |
| CN109961587A (en) * | 2017-12-26 | 2019-07-02 | 天地融科技股份有限公司 | A kind of monitoring system of self-service bank |
| CN116612594A (en) * | 2023-05-11 | 2023-08-18 | 深圳市云之音科技有限公司 | Intelligent monitoring and outbound system and method based on big data |
-
2007
- 2007-06-01 CN CNB2007101059341A patent/CN100555348C/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103400120A (en) * | 2013-08-02 | 2013-11-20 | 上海泓申科技发展有限公司 | Video analysis-based bank self-service area push behavior detection method |
| CN103400120B (en) * | 2013-08-02 | 2015-04-15 | 上海泓申科技发展有限公司 | Video analysis-based bank self-service area push behavior detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101079171A (en) | 2007-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100555348C (en) | Intelligent video monitoring system of bank self-aid apparatus | |
| KR102104088B1 (en) | Uwb-based location tracking and ai combined intelligent object tracking video monitoring system | |
| US9652863B2 (en) | Multi-mode video event indexing | |
| Park et al. | Construction worker detection in video frames for initializing vision trackers | |
| CN100487739C (en) | Multi-layer real time forewarning system based on the intelligent video monitoring | |
| CN104866827B (en) | Personnel based on video monitoring platform cross the detection method of behavior | |
| CN103942850B (en) | Based on medical personnel's monitoring method on duty of video analysis and RFID technique | |
| CN105426820B (en) | More people's anomaly detection methods based on safety monitoring video data | |
| EP2801078B1 (en) | Context aware moving object detection | |
| CN101751677B (en) | Target continuous tracking method based on multi-camera | |
| CN102663743B (en) | Personage's method for tracing that in a kind of complex scene, many Kameras are collaborative | |
| CN101872526B (en) | Smoke and fire intelligent identification method based on programmable photographing technology | |
| Kim et al. | Smoke detection in video | |
| CN102665071A (en) | Intelligent processing and search method for social security video monitoring images | |
| Kumar et al. | Study of robust and intelligent surveillance in visible and multi-modal framework | |
| CN111144291A (en) | Method and device for distinguishing personnel invasion in video monitoring area based on target detection | |
| CN101635835A (en) | Intelligent video monitoring method and system thereof | |
| CN109151388A (en) | A kind of video frequency following system that multichannel video camera is coordinated | |
| CN101854516A (en) | Video monitoring system, video monitoring server and video monitoring method | |
| JP2005506740A (en) | Video trip wire | |
| CN108376246A (en) | A kind of identification of plurality of human faces and tracking system and method | |
| CN101266710A (en) | An all-weather intelligent video analysis monitoring method based on a rule | |
| CN103281518B (en) | A kind of multi-network all-weather intelligent video supervisory control system | |
| KR20060031832A (en) | A smart visual security system based on real-time behavior analysis and situation cognizance | |
| CN102892007A (en) | Method and system for facilitating color balance synchronization between a plurality of video cameras as well as method and system for obtaining object tracking between two or more video cameras |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20091028 Termination date: 20200601 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |