CN105611233A - Online video monitoring method for static scene - Google Patents
Online video monitoring method for static scene Download PDFInfo
- Publication number
- CN105611233A CN105611233A CN201510959344.XA CN201510959344A CN105611233A CN 105611233 A CN105611233 A CN 105611233A CN 201510959344 A CN201510959344 A CN 201510959344A CN 105611233 A CN105611233 A CN 105611233A
- Authority
- CN
- China
- Prior art keywords
- target
- static scene
- online video
- supervising
- training sample
- 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.)
- Pending
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
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
- G08B13/19608—Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position
Abstract
The invention provides an online video monitoring method for a static scene, and the method comprises the steps: carrying out target detection of a monitoring video through employing the technology of pattern recognition and the technology of background difference; and giving an alarm if a suspected object is detected. The method carries out the target detection in a static monitoring scene based on the pattern recognition and background difference, determines an end point search region through employing priori knowledge, guarantees the real-time operation, carries out the updating of a recognition model through employing new positive and negative samples generated in an operation process, and avoids the problems that manual on-duty burden is large and long-time working causes negligence and leakage in a conventional video monitoring method. A person on duty just needs to confirm the detection result, so the method can effectively reduce the work intensity of the person on duty, and greatly reduces the dependence on the person on duty. The method effectively avoids detection leakage hidden troubles caused by the accidental leaving of the person on duty, greatly improves the monitoring efficiency and quality, and can quickly cut and store a detection object so as to facilitate the searching once there is a suspected object.
Description
Technical field
The present invention relates to video frequency graphic monitoring technical field, relate in particular to a kind of static scene Online Video prisonControl method.
Background technology
Traditional video surveillance method relies on artificial on duty carrying out, when more in the region of needs monitoring, on dutyPersonnel need to be simultaneously in the face of tens even dozens of monitored picture judged whether one by one suspicious situation, prisonControl work is heavy, and very easily careless omission, neglects in low period of notice (as the late into the night) or duty personnelWhen post, more easily cannot ensure monitoring effect.
Summary of the invention
The object of the present invention is to provide a kind of static scene Online Video method for supervising, existing in order to solveIn technology, rely on the problem that artificial post is prone to careless omission.
The invention provides a kind of static scene Online Video method for supervising, comprising:
By mode identification technology and background differential technique, monitor video is carried out to target detection;
If suspected target being detected reports to the police.
Further, by mode identification technology and background differential technique, monitor video is carried out to target detection,Specifically comprise:
By object detector, monitor video is carried out to target detection, if detect in all key areasNumber of targets sum is greater than 1, determines and suspected target detected; Or, by background model, monitoring is lookedFrequently carry out target detection, if the area of difference image foreground area is greater than default alarm threshold value, determineSuspected target detected.
Adopt the beneficial effect of the invention described above technical scheme to be: based target identification and background subtraction divide and carry outTarget detection under static monitoring scene, utilizes priori to determine terminal region of search, ensures fortune in real timeOK, and utilize the new positive negative sample producing in running to upgrade model of cognition, avoided biographySystem video frequency monitoring method relies on artificial post workload greatly completely and the easily appearance that works long hours is neglected,The problem of omitting. Duty personnel only need confirm testing result, can effectively reduce duty personnelWorking strength, and significantly reduced the degree of dependence to duty personnel, effectively stop duty personnel unexpected fromThe undetected hidden danger that cause on hilllock, has improved monitoring efficiency and quality greatly, and can be rapid to the target detectingShear and preserve, once occur that suspicious object is convenient to search.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of static scene Online Video method for supervising of the present invention;
Fig. 2 is the flow chart of static scene Online Video method for supervising embodiment of the present invention.
Detailed description of the invention
For making object, technical scheme and the advantage of the embodiment of the present invention clearer, below in conjunction with thisAccompanying drawing in bright embodiment, is clearly and completely described the technical scheme in the embodiment of the present invention,Obviously, described embodiment is the present invention's part embodiment, instead of whole embodiment.
The invention discloses a kind of static scene Online Video method for supervising, as shown in Figure 1, the method bagDraw together following steps:
Step S101, carries out target detection by mode identification technology and background differential technique to monitor video;
Step S102, if detect, suspected target reports to the police.
Mode identification technology belongs to the object detection method in vision field, due to traditional mode identification technologyWhile application in vision field, exist the all-purpose detector training adapting under different scenes, visual angle more difficult,And the needed time of target detection is longer, especially when need to be in different scale and larger range of videoThe time needing while retrieval is longer, thereby cannot requirement of real time. Therefore, the present invention is directed toStatic scene, fixes at visual angle, yardstick fixes, and the known situation in the zone of action of target in video,Therefore can realize and utilize mode identification technology to carry out real-time target detection; Meanwhile, draw in the present inventionEnter background differential technique, thereby can further reduce target loss.
In embodiments of the present invention, by mode identification technology and background differential technique, monitor video is carried outTarget detection, wherein, mode identification technology refers to by object detector carries out target inspection to monitor videoSurvey, if the number of targets sum detecting in all key areas is greater than 1, determines and suspected target detected;Background differential technique refers to by background model monitor video is carried out to target detection, if difference image prospectThe area in region is greater than default alarm threshold value, determines and suspected target detected.
Fig. 2 is the flow chart of static scene Online Video method for supervising embodiment of the present invention, as shown in Figure 2,Comprise:
Step S201, gathers positive and negative training sample;
Artificial from historical monitoring image, gather respectively with respect to the each M of positive and negative training sample of target withUpper, wherein M > 2000, concrete gatherer process is:
Positive sample is the image outline that comprises target, wherein the width of sketch figure picture and height and target realityWidth and aspect ratio are 1.1:1; Negative sample is the image outline that comprises part target area; Then alignNegative sample carries out size normalization processing. For example: if desired training of human detector, can be by positive and negativeSize is all normalized to (H:110, W:40).
Step S202, utilizes training sample training objective detector;
Specifically can use gathered positive negative sample, choose suitable target detection feature, useThe object detector that AdaBoost Algorithm for Training is made up of some Weak Classifier cascades.
Step S203, determines the key monitoring region in monitor video;
Be specially the upper left corner and lower right corner coordinate by extracting the each key area in monitoring image, fromGather Z={z and obtain a key area1,…,zn}。
Step S204, initializes background model;
Obtain the front N frame monitoring image of monitor video, for initializing background model, specify the back of the body simultaneouslyScape difference alarm threshold value, in the present embodiment, N >=200, can establish background difference alarm threshold value is α.
Step S205, carries out target detection by object detector and background model to monitor video;
Testing process is specially: start to extract every frame monitoring image from the N+1 two field picture of monitor video,In each key monitoring region of every frame monitoring image, utilize the object detector scan image trainingAnd carry out target detection. Background model by the every frame monitoring image extracting and after initializing is poor simultaneouslyPoint, obtain difference image Id, and calculate difference image IdThe area S of middle foreground area.
Step S206, judge detect number of targets sum whether be greater than 1 or S > α;
Judge the target whether object detector detects, the number of targets sum that object detector detectsWhether be greater than 1, if so, determine and now perform step S207 by the suspected target that object detector detects,Otherwise execution step S210.
Judge difference image IdWhether the area S of middle foreground area is greater than background difference alarm threshold value α, ifBe, determine and now perform step S207 by the suspected target detecting, otherwise execution step S210.
Step S207, determines and suspected target detected and report to the police;
Determine and suspected target detected, report to the police to remind supervisor to note, and working as suspected targetPrior image frame storage is for future reference, preserves the position coordinates of each suspected target in image simultaneously.
Step S208, determines whether false alarm;
Supervisor judges alarming result, performs step S209 if be judged as false alarm, otherwiseExecution step S210.
Step S209, upgrades object detector and alarm threshold value α;
Position coordinates according to the suspected target of preserving in step S207 in image, extraction target areaImage, and deposit negative example base in. In negative example base, new samples quantity exceedes at 1000 o'clock, can utilizeNew negative example base is upgraded object detector, and upgrades alarm threshold value α according to monitoring effect.
Step S210, judges whether to exist next frame monitoring image;
If there is next frame monitoring image, return to execution step S205, otherwise execution step S211.
Step S211, process ends.
Static scene Online Video method for supervising of the present invention, based target identification and background subtraction divide carry out quietTarget detection under state monitoring scene, utilizes priori to determine terminal region of search, ensures real time execution,And utilize the new positive negative sample producing in running to upgrade model of cognition, avoid tradition to lookFrequently method for supervising relies on artificial post workload completely greatly and works long hours and easily occurs neglecting, omittingProblem. Duty personnel only need confirm testing result, can effectively reduce duty personnel's workIntensity, and significantly reduced the degree of dependence to duty personnel, effectively stop duty personnel and surprisingly leave the post to makeThe undetected hidden danger becoming, has greatly improved monitoring efficiency and quality, and can shear rapidly the target detectingPreserve, once occur that suspicious object is convenient to search.
One of ordinary skill in the art will appreciate that: all or part of step that realizes above-mentioned each embodiment of the methodSuddenly can complete by the relevant hardware of programmed instruction. Aforesaid program can be stored in a computer canRead in storage medium. This program, in the time carrying out, is carried out the step that comprises above-mentioned each embodiment of the method; AndAforesaid storage medium comprises: ROM, RAM, magnetic disc or CD etc. are various can be program code storedMedium.
Finally it should be noted that: above each embodiment is only in order to technical scheme of the present invention to be described, but not rightIts restriction; Although the present invention is had been described in detail with reference to aforementioned each embodiment, this area commonTechnical staff is to be understood that: its technical scheme that still can record aforementioned each embodiment is modified,Or some or all of technical characterictic is wherein equal to replacement; And these amendments or replacement, andDo not make the essence of appropriate technical solution depart from the scope of various embodiments of the present invention technical scheme.
Claims (10)
1. a static scene Online Video method for supervising, is characterized in that, comprising:
By mode identification technology and background differential technique, monitor video is carried out to target detection;
If suspected target being detected reports to the police.
2. static scene Online Video method for supervising according to claim 1, is characterized in that instituteState by mode identification technology and background differential technique monitor video carried out to target detection, specifically comprise:
By object detector, monitor video is carried out to target detection, if detect in all key areasNumber of targets sum is greater than 1, determines and suspected target detected; Or,
By background model, monitor video is carried out to target detection, if the area of difference image foreground area is largeIn default alarm threshold value, determine and suspected target detected.
3. static scene Online Video method for supervising according to claim 2, is characterized in that instituteState by object detector monitor video is carried out to target detection, if the order detecting in all key areasMark number sum is greater than 1, determines and detects that the operation of suspected target is specially:
Determine the key monitoring region in described monitor video, by described object detector to described monitoringImage scanning is carried out to carry out target detection, if institute in each key monitoring region of the current frame image of videoState the number of targets sum detecting in key monitoring region and be greater than 1, determine and suspected target detected.
4. static scene Online Video method for supervising according to claim 2, is characterized in that instituteState by background model monitor video is carried out to target detection, if the area of difference image foreground area is greater thanDefault alarm threshold value, determine and detect that the operation of suspected target is specially:
The current frame image of described monitor video and background model are carried out to difference, to obtain difference image,If the area of foreground area is greater than default alarm threshold value in described difference image, determine detect doubtfulTarget.
5. according to the static scene Online Video method for supervising described in claim 2 or 4, it is characterized in that,Described background model initializes by the front N two field picture of described monitor video, described N >=200.
6. according to the static scene Online Video method for supervising described in claim 2 or 3, it is characterized in that,Described object detector is from historical monitoring image, to gather the training sample with respect to target, according to describedThe set of the definite target detection feature of training sample.
7. static scene Online Video method for supervising according to claim 6, is characterized in that instituteState with respect to the training sample of target and comprise with respect to the positive training sample of target with respect to the negative instruction of targetPractice sample, described positive training sample is the image outline that comprises target; Described negative training sample is for comprising portionThe image outline in partial objectives for region.
8. static scene Online Video method for supervising according to claim 7, is characterized in that,Described determine detect after suspected target and also comprise: preserve the coordinate position of each suspected target in imagePut and described current frame image.
9. static scene Online Video method for supervising according to claim 8, is characterized in that,Detect that if described suspected target also comprises after reporting to the police:
If determine that according to described suspected target described warning is false alarm, extracts described suspected target placeThe image in region is negative training sample material, upgrades negative training sample according to described negative training sample material.
10. static scene Online Video method for supervising according to claim 5, is characterized in that,If detect that described suspected target also comprises after reporting to the police: periodically update according to monitoring effectDescribed alarm threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510959344.XA CN105611233A (en) | 2015-12-18 | 2015-12-18 | Online video monitoring method for static scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510959344.XA CN105611233A (en) | 2015-12-18 | 2015-12-18 | Online video monitoring method for static scene |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105611233A true CN105611233A (en) | 2016-05-25 |
Family
ID=55990736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510959344.XA Pending CN105611233A (en) | 2015-12-18 | 2015-12-18 | Online video monitoring method for static scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105611233A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107742385A (en) * | 2017-09-13 | 2018-02-27 | 武汉安保通科技有限公司 | A kind of micro-vibration warning system and method |
| CN108900813A (en) * | 2018-07-24 | 2018-11-27 | 安徽康能电气有限公司 | A kind of video monitoring device that intelligent power transmission line foreign object is identified with amiable shape |
| CN109919058A (en) * | 2019-02-26 | 2019-06-21 | 武汉大学 | A kind of multisource video image highest priority rapid detection method based on Yolo V3 |
| CN110852253A (en) * | 2019-11-08 | 2020-02-28 | 杭州宇泛智能科技有限公司 | Ladder control scene detection method and device and electronic equipment |
| CN110944159A (en) * | 2019-12-31 | 2020-03-31 | 联想(北京)有限公司 | Information processing method, electronic equipment and information processing system |
| CN111597917A (en) * | 2020-04-26 | 2020-08-28 | 河海大学 | Target detection method based on frame difference method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101409004A (en) * | 2008-11-24 | 2009-04-15 | 浙江大学 | Safety defense monitoring method based on Symbian intelligent mobile phone platform |
| US20120314064A1 (en) * | 2011-06-13 | 2012-12-13 | Sony Corporation | Abnormal behavior detecting apparatus and method thereof, and video monitoring system |
| CN103473570A (en) * | 2013-08-16 | 2013-12-25 | 电子科技大学 | Self-adaptive video scene pedestrian detection method |
| CN103577795A (en) * | 2012-07-30 | 2014-02-12 | 索尼公司 | Detection equipment and method, detector generation equipment and method and monitoring system |
-
2015
- 2015-12-18 CN CN201510959344.XA patent/CN105611233A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101409004A (en) * | 2008-11-24 | 2009-04-15 | 浙江大学 | Safety defense monitoring method based on Symbian intelligent mobile phone platform |
| US20120314064A1 (en) * | 2011-06-13 | 2012-12-13 | Sony Corporation | Abnormal behavior detecting apparatus and method thereof, and video monitoring system |
| CN103577795A (en) * | 2012-07-30 | 2014-02-12 | 索尼公司 | Detection equipment and method, detector generation equipment and method and monitoring system |
| CN103473570A (en) * | 2013-08-16 | 2013-12-25 | 电子科技大学 | Self-adaptive video scene pedestrian detection method |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107742385A (en) * | 2017-09-13 | 2018-02-27 | 武汉安保通科技有限公司 | A kind of micro-vibration warning system and method |
| CN107742385B (en) * | 2017-09-13 | 2020-05-01 | 武汉安保通科技有限公司 | Micro-vibration alarm system and method |
| CN108900813A (en) * | 2018-07-24 | 2018-11-27 | 安徽康能电气有限公司 | A kind of video monitoring device that intelligent power transmission line foreign object is identified with amiable shape |
| CN109919058A (en) * | 2019-02-26 | 2019-06-21 | 武汉大学 | A kind of multisource video image highest priority rapid detection method based on Yolo V3 |
| CN110852253A (en) * | 2019-11-08 | 2020-02-28 | 杭州宇泛智能科技有限公司 | Ladder control scene detection method and device and electronic equipment |
| CN110944159A (en) * | 2019-12-31 | 2020-03-31 | 联想(北京)有限公司 | Information processing method, electronic equipment and information processing system |
| CN111597917A (en) * | 2020-04-26 | 2020-08-28 | 河海大学 | Target detection method based on frame difference method |
| CN111597917B (en) * | 2020-04-26 | 2022-08-05 | 河海大学 | Target detection method based on frame difference method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105611233A (en) | Online video monitoring method for static scene | |
| US20240062518A1 (en) | Method, apparatus, computer device and storage medium for detecting objects thrown from height | |
| Li et al. | 3D parallel fully convolutional networks for real-time video wildfire smoke detection | |
| RU2393544C2 (en) | Method and device to detect flame | |
| US10691947B2 (en) | Monitoring device | |
| CN103093212B (en) | The method and apparatus of facial image is intercepted based on Face detection and tracking | |
| CN108010242B (en) | Security alarm method, system and storage medium based on video identification | |
| CN109757395A (en) | A kind of pet behavioral value monitoring system and method | |
| CN105554462A (en) | Remnant detection method | |
| CN104616006A (en) | Surveillance video oriented bearded face detection method | |
| CN104866842A (en) | Human body target intrusion detection method | |
| Cai et al. | Intelligent video analysis-based forest fires smoke detection algorithms | |
| CN110674753A (en) | Theft early warning method, terminal device and storage medium | |
| CN111524189A (en) | Door opening and closing state detection method and device and computer storage medium | |
| CN115205581A (en) | Fishing detection method, fishing detection device and computer readable storage medium | |
| CN103093197A (en) | Monitoring method and system for recognizing hanging behavior | |
| CN103489202A (en) | Intrusion detection method based on videos | |
| CN107729811B (en) | Night flame detection method based on scene modeling | |
| CN104866844A (en) | Crowd gathering detection method for monitor video | |
| CN112070033A (en) | Video carry-over detection method based on finite-state machine analysis | |
| CN114821978B (en) | Method, device and medium for eliminating false alarm | |
| CN104574340A (en) | Video intrusion detection method based on historical images | |
| CN115810161A (en) | Transformer substation fire identification method and system | |
| WO2018110377A1 (en) | Video monitoring device | |
| CN105243355A (en) | Event-driven remote wireless coalbed methane well station abnormal scene safety monitoring method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160525 |
|
| RJ01 | Rejection of invention patent application after publication |