CN101635026B - Method for detecting derelict without tracking process - Google Patents
Method for detecting derelict without tracking process Download PDFInfo
- Publication number
- CN101635026B CN101635026B CN2008101170593A CN200810117059A CN101635026B CN 101635026 B CN101635026 B CN 101635026B CN 2008101170593 A CN2008101170593 A CN 2008101170593A CN 200810117059 A CN200810117059 A CN 200810117059A CN 101635026 B CN101635026 B CN 101635026B
- Authority
- CN
- China
- Prior art keywords
- abandon
- zone
- picture element
- derelict
- term
- 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
- 238000000034 method Methods 0.000 title claims abstract description 80
- 230000008569 process Effects 0.000 title claims abstract description 43
- 238000001514 detection method Methods 0.000 claims abstract description 35
- 238000004040 coloring Methods 0.000 claims description 25
- 230000003068 static effect Effects 0.000 claims description 16
- 239000000284 extract Substances 0.000 claims description 14
- 238000012360 testing method Methods 0.000 claims description 12
- 238000012986 modification Methods 0.000 claims description 11
- 230000004048 modification Effects 0.000 claims description 11
- 230000008878 coupling Effects 0.000 claims description 6
- 238000010168 coupling process Methods 0.000 claims description 6
- 238000005859 coupling reaction Methods 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 4
- 230000000295 complement effect Effects 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000013459 approach Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 14
- 239000000203 mixture Substances 0.000 abstract 2
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 239000002360 explosive Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 206010003830 Automatism Diseases 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 231100001261 hazardous Toxicity 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a method for detecting a derelict without a tracking process, which comprises the following steps: extracting pixel points probably belonging to the derelict from each frame of an image input by a camera by using a method based on the pixel points, and updating a Gaussian mixture model of a corresponding position by using RGB color information of each pixel point probably belonging to the derelict; and inputting the RGB color information of each pixel point in the image into the updated corresponding Gaussian mixture model, detecting out the pixel points matching with the color information describing the derelict probably appearing at the pixel point of the corresponding position, finishing the detection of the derelict in a video, and giving an alarm if a derelict is detected. The derelict detection without the tracking process contains no tracking process so that the derelict detection plays a very important role of improving the precise detection on the derelict in a monitored scene by a monitoring system in a complex scene. The method can be applied to an intelligent video monitoring system for the help of recognizing the derelict in a public scene so as to improve the security of public places.
Description
Technical field
The invention belongs to area of pattern recognition, relate to technology such as Flame Image Process and computer vision, the abandon that particularly relates to based on video detects.
Background technology
Along with the development of technology and the reduction gradually of hardware device price, a large amount of monitoring cameras is installed to various occasions, and particularly those are to occasions of safety requirements sensitivity, like airport, community, bank, parking lot, military base etc.The video monitoring of dynamic scene is the forward position research direction that receives much concern in recent years, and its detection from the video camera sequences of images captured, identification, tracking target are also understood its behavior.Although the rig camera that extends as human vision at present ubiquity in commercial application, present supervisory system generally can only be used videotape to record and is used for collecting evidence afterwards, and analysis video data and Realtime Alerts that can not be real-time.Therefore, develop automatism with practical significance, intelligent video monitoring system becomes urgent and necessary day by day.This just requires and can not only replace human eye with video camera, and the general-purpose computers contributor, replaces the people, keeps watch on or control task to accomplish.
Abandon detects significant for the safety of safeguarding the public place.Here said abandon detects and is meant that detecting those is had a mind to abandon, throws in public places or some critical position by the people, and the knapsack, briefcase of explosive etc. possibly are housed.Usually the terrorist ignites the bomb that is contained in the bag through the mode of timing or remote control after placing such parcel.This crime means cost is low, harm is big, take precautions against and track down the difficulty height, becomes the offender gradually and carries out one of main mode of explosive attack.Similarly case emerged in an endless stream, like Madrid, ESP serial blast case in 2004; London in 2005 and Liverpudlian case of explosion or the like.Detect based on the abandon of video and to be meant and to utilize field erected video monitoring equipment, through the picture material analysis in the video image being surveyed the generation that abandon detects incident.Clearly, detect on the basis of existing supervisory system, need not to increase new hardware input based on the abandon of video.
Having had at present more algorithm that the abandon based on video is detected studies; But often all exist certain shortcoming: present method need detect moving targets all in the scene; Confirm the position of same target in the image sequence different frame through tracing process then; Thereby obtain the movement locus of each moving target in scene; If there is the track of certain target from the track of another target, to separate, and this target keeps stationary state to surpass a period of time, and then this target is an abandon.But monitoring scene tends to more complicated, and for example: dynamic background and moving target are too much.This brings great challenge can for the accurate tracking of moving target.So the accuracy of these class methods is relatively poor, practicality is bad.
Summary of the invention
The abandon detection algorithm of prior art needs accurate detection and tracking process; When scene is complicated; Not only can cause a large amount of erroneous detection that the computation complexity of system is increased substantially, in order to solve prior art problems, the objective of the invention is effective, convenient, accurately detect abandon; For this reason, provide a kind of based on abandon detection method video and that need not tracing process.
To achieve these goals, the abandon detection method based on video provided by the invention need not tracing process, and comprises that model modification and abandon detect two processes, and step is following:
Step S1: to every two field picture of camera input, utilize based on the method for picture element and extract the picture element that possibly belong to abandon, and remove to upgrade the mixed Gauss model of relevant position with the RGB colouring information that each possibly belong to the abandon picture element;
Step S2: the RGB colouring information of each picture element in the said image is input in the corresponding mixed Gauss model that upgraded; Detect and describe the picture element that the colouring information of the abandon that the relevant position picture element possibly occur is complementary, accomplish the detection of abandon in the video.
Method of the present invention can be given prominence to its superiority in the scene (like dynamic background, moving target is too much) of complicacy.Method of the present invention is different with other, and method of the present invention need not carried out moving object detection and tracking, thereby is easier to realize and use.The present invention detects abandon can be applied in following aspect:
(1) is applied to the intelligent video monitoring field, is used for helping to provide the potential hazardous location of monitoring scene and warning being provided, thereby avoid the massive losses that causes because of possible explosive incident.
(2) be applied to the object identification of computer vision field, at first target carried out rough sort, dwindle the search volume, improve recognition efficiency and accuracy rate.
(3) semantization that is applied in the intelligent monitor system is understood, and the classification of the main body in its semantization is provided, and help system is understood event in the scene.
Description of drawings
Fig. 1 illustrates the process flow diagram that the present invention need not the abandon detection of tracing process.
Fig. 2 a illustrates the frame example in the sport video of the present invention.
Fig. 2 b illustrates and the present invention is based on the zone that the detected background of background modeling method changes.
Fig. 2 c illustrates the detected stagnant zone in short-term of the present invention.
Fig. 2 d illustrates the present invention stagnant zone is in short-term removed the result behind the noise.
Fig. 3 a illustrates the frame example in the sport video of the present invention.
Fig. 3 b illustrates a frame example adjacent with Fig. 3 a.
Fig. 3 c illustrates among the embodiment because the slow erroneous detection that causes of target travel.
Fig. 4 a illustrates the frame example image before the long-time static target motion among the embodiment.
Fig. 4 b illustrates long-time static target post exercise one frame example image among the embodiment.
Fig. 4 c illustrates among the embodiment because the erroneous detection that causes after the long-time static target motion.
Fig. 5 a illustrates a frame warning example among the embodiment.
Fig. 5 b illustrates and removes the preceding abandon testing result of noise among the embodiment.
Fig. 5 c illustrates the abandon testing result behind the removal noise among the embodiment.
Embodiment
Specify each related detailed problem in the technical scheme of the present invention below in conjunction with accompanying drawing.Be to be noted that described embodiment only is intended to be convenient to understanding of the present invention, and it is not played any qualification effect.
Traditional abandon detection method based on video needs tracing process usually, and complicated scene can cause great challenge to accurate tracking.So the accuracy of these class methods is relatively poor, practicality is bad.The abandon that need not tracing process detects because do not comprise tracing process, plays a very important role for improving the accurate detection of supervisory system to abandon in the monitoring scene.Utilize the characteristic of abandon, the present invention has realized an abandon detection system that need not tracing process.Like Fig. 1 the FB(flow block) of the abandon detection method that need not tracing process is shown, comprises that model modification and abandon detect two parts:
Described model modification process comprises step: to every two field picture of camera input, utilize background modeling and inter-frame difference method to extract static in short-term picture element; To the stagnant zone in short-term that extracts, utilize abandon to have a certain size and have the characteristic of limbus to remove noise, thereby obtain the zone that possibly belong to abandon; To the extracted region that possibly the belong to abandon corresponding mixed Gauss model of each picture element RGB information updating in should the zone, update method is the K averaging method.
Described abandon testing process comprises step: in the mixed Gauss model that the input of the RGB colouring information of each picture element in the said image is corresponding, detect and describe the picture element that the colouring information of the abandon that the relevant position picture element possibly occur is complementary; Utilize abandon to have a certain size and have the characteristic of limbus to remove noise; If have abandon to report to the police immediately, otherwise then continue testing process.
The hardware minimalist configuration that method of the present invention needs is: P43.0G CPU, the computing machine of 512M internal memory; Lowest resolution is the monitoring camera of 320x240; Frame per second is the video frequency collection card of 25 frame per seconds.Here camera and capture card are used to the view data that provides real-time.On the hardware of this configuration level, adopt the C Plus Plus programming to realize this method, can reach the effect of real-time detection.The present invention comprises two processes: model modification process and abandon testing process.In the face of two related committed steps of process in the method for the present invention specify one by one, concrete form is described below down:
The model modification process:
At first, be that stagnant zone extracts in short-term:
Usually abandon has following two characteristics:
(1) abandon can cause the background in occupied zone to change;
(2) object keeps static after abandoned.
This step is exactly above-mentioned two characteristics according to abandon, extracts the stagnant zone in short-term in the scene.It comprises two processes:
The first step is the method for utilizing based on background modeling, extracts the zone that background changes in the scene.Said background modeling is the colouring information that extracts each frame of movement destination image sequence, carries out the mixed Gaussian background modeling through the colouring information to each picture element and makes this model can describe the background color information of corresponding picture element.The zone that therefore can change with method detection background based on background modeling.Mixed Gauss model is widely used in background modeling at present, and more stable performance is arranged.Present image shown in accompanying drawing 2a, the zone that the detected background shown in accompanying drawing 2b changes.
Second step was utilized the inter-frame difference method in the zone that the background of extracting changes, and extracted static in short-term zone.The frame-to-frame differences point-score is exactly the method for carrying out difference with current frame image and consecutive frame image.If the difference value of a certain picture element, just thinks that this picture element is static in short-term less than certain threshold value.Detected stagnant zone in short-term shown in accompanying drawing 2c.
Its two, to the stagnant zone in short-term that extracts, carry out noise region and remove:
We can find out that the stagnant zone in short-term that detects is not an abandon entirely from accompanying drawing 2c, include a large amount of noises.These generating noise mainly contain following two reasons:
(1) when moving target motion slowly, and the interior of articles distribution of color is relatively evenly the time, the interior zone of moving target can be detected as stagnant zone by the inter-frame difference method.Simultaneously should the zone have obviously also caused the variation in short-term of background.Adjacent two frames shown in accompanying drawing 3a and accompanying drawing 3b utilize static picture element in the scene that the inter-frame difference method extracts shown in accompanying drawing 3c.The interior zone that can find out moving target from accompanying drawing 3c is mistaken as stagnant zone.
(2) long-time static object, the background area of exposing after the motion suddenly can be considered to background the zone that changes has taken place.Long-time static object can be thought the background area based on the method for background modeling.Spill the real background that comes after the target travel because its colouring information is inconsistent with the model of describing this regional background colouring information, and treated as prospect, promptly be mistaken as this regional background change has taken place.Obviously also belonged to simultaneously stagnant zone in short-term by the zone of flase drop.Image before the long-time static target motion shown in accompanying drawing 4a, image after the target travel shown in accompanying drawing 4b, the zone that the background that the background modeling method shown in accompanying drawing 4c extracts changes.Can find out the target travel of long-time stop from Fig. 4 c after, the zone that background changes has been treated as in the background area of exposing.
But we find from experiment, and the profile in these erroneous detection zones corresponds to after the original image, and marginal information is also inadequate.In other words, abandon generally all has tangible edge.We utilize this characteristic to remove by above-mentioned two kinds of erroneous detection that reason causes.Its specific practice is: utilize blob analytical approach stagnant zone piecemeal in short-term, and extract the edge of each piece; Present image is extracted the edge with the canny algorithm; If the coincident degree of a certain edge and present image during greater than certain threshold value, just thinks that this piece is possible abandon.Stagnant zone in short-term shown in accompanying drawing 2d is removed the result behind the noise.Registration is by computes:
Its three, upgrade the mixed Gauss model of relevant position:
Be applied to background classes seemingly with mixed Gauss model, here we also make up a mixed Gauss model at each picture element, are used for describing the colouring information that abandon possibly appear in this picture element.The colouring information of the picture element that step 2 is extracted adopts the mixed Gauss model of K averaging method online updating corresponding position, is used to describe the abandon colouring information.Suppose that then the average of this Gaussian function and variance are upgraded by following formula within the twice variance of rgb value certain Gaussian function average in mixed Gauss model of certain picture element:
μ
r＝(1-α)μ
r+αR μ
g＝(1-α)μ
g+αG μ
b＝(1-α)μ
b+αB
Wherein, μ
r, μ
g, μ
bBe respectively the RGB component of Gaussian function average, R, G, B are respectively the values of the RGB component of picture element, and σ is the variance of Gaussian function.Obviously, the update strategy of its Gauss's parameter is similar with the gauss hybrid models in the background modeling, but renewal process also has the following steps:
(1) the Gauss's weights in the mixed Gauss model are to be confirmed by the time length that this Gauss is not updated here.The time of not upgrading is long more, and its weights are more little.So each Gauss's the recent renewal time needs to preserve.
(2) each Gauss's the structure time need be confirmed, just can report to the police because have only after parcel abandons certain hour.
(3) mixed Gauss model is not before being upgraded by any sample, and its Gauss's number is 0.
The abandon testing process:
At first, detect the picture element that is complementary with the mixed Gauss model of describing abandon information:
Here matching process also with background modeling in the mixed Gauss model matching process similar, its difference is:
What (1) we detected here is the picture element with Model Matching, and detect in the background modeling be and the unmatched picture element of model.
(2) except colouring information need mate, the current time need be greater than a threshold value, because have only parcel to be dropped above certain hour and could to report to the police with the Gauss's of being mated creation-time.This threshold value is confirmed by manual.
(3) if in certain picture element mixed Gaussian, Gauss's number is 0, does not then match.
Detection effect shown in accompanying drawing 5b, we find to have a large amount of noises.
Its two, utilize the characteristic of abandon to remove detection noise:
The characteristic that this step adopted with the strategy with the model modification process in step 2 identical.If still there is abandon to be detected, system alarm then, if there is not abandon to be detected, then system's step 1 of forwarding the abandon testing process to continues testing process.Frame warning example shown in accompanying drawing 5a.Abandon testing result behind the removal noise shown in accompanying drawing 5c.
The abandon detection scheme that need not tracing process mainly comprises model modification and two processes of abandon detection, in order to specify the specific embodiment of this invention, is the example explanation with some intelligent video monitoring systems that is used for station hall.This system can detect the abandon in the monitoring scene in real time.
The purpose of model modification is on-line study and upgrade the model that is used to describe the abandon distribution of color that each picture element possibly occur.
Model modification is following:
Step S11:, utilize in this frame of background modeling and inter-frame difference method extraction stagnant zone in short-term to every two field picture of camera input; Extracting in short-term, stagnant zone comprises following process: utilize the method based on background modeling, detect the zone that background changes in the present frame; In the zone that background changes, utilize the inter-frame difference method to detect static in short-term zone;
Step S12: to the stagnant zone in short-term that S11 obtained, utilize abandon to have a certain size and the tangible characteristics in edge, further get rid of the noise in the stagnant zone in short-term that step S11 extracted, thereby obtain to belong to the zone of abandon;
Step S13:, each the picture element colouring information in this zone is upgraded mixed Gaussian (Gaussian MixtureModel) model of relevant position to the zone that possibly belong to abandon that S12 obtained.
Train be used to describe the model of abandon color characteristics after, we detect the zone that possibly have abandon in every frame in real time.Its concrete steps are following:
Step S21: the RGB colouring information that every two field picture of camera input is extracted each picture element; And the RGB colouring information of each picture element is input in the mixed Gauss model of correspondence position; Detect the picture element of RGB colouring information and correspondence position mixed Gauss model coupling, obtain the zone of forming by the picture element of mixed Gauss model coupling;
Step S22: to the zone of forming by the picture element of coupling, utilize abandon to have limbus and a certain size characteristics, further get rid of noise region, thereby obtain the abandon zone;
Step S23:,, then report to the police if there is abandon to be detected to every two field picture of input; Otherwise, then continue the abandon testing process of execution in step S21.
In a word, the present invention proposes a kind of simple and effective abandon detection method that need not tracing process.Test findings on the PETS06 database has been verified the validity of our algorithms.The present invention is easy to realize, stable performance.The present invention is for improving the understandability of supervisory system to monitoring scene, and the security that promotes the public place has very important effect.
The above; Be merely the embodiment among the present invention, but protection scope of the present invention is not limited thereto, anyly is familiar with this technological people in the technical scope that the present invention disclosed; Can understand conversion or the replacement expected; All should be encompassed in of the present invention comprising within the scope, therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.
Claims (6)
1. abandon detection method that need not tracing process is characterized in that: comprise that model modification and abandon detect two processes, its step is following:
Step S1: to every two field picture of camera input, cause the background in occupied zone to change and object keeps static characteristics after abandoned according to abandon, utilization extracts the zone that background changes in the scene based on the method for background modeling; Zone in that the background of extracting changes utilizes the inter-frame difference method that current frame image and consecutive frame image are carried out difference processing, if the difference value of a certain picture element less than threshold value, then this picture element is static in short-term, extracts static in short-term zone; Utilize blob analytical approach stagnant zone piecemeal in short-term, and extract the edge of each piece; Present image is extracted the edge with the canny algorithm; If the coincident degree of a certain edge and present image during greater than threshold value, just thinks that this piece is possible abandon; Utilization is extracted the picture element that possibly belong to abandon based on the method for picture element, and removes to upgrade the mixed Gauss model of relevant position with the RGB colouring information that each possibly belong to the abandon picture element; Said registration is:
Step S2: the RGB colouring information of each picture element in the said image is input in the corresponding mixed Gauss model that upgraded; Detect and describe the picture element that the colouring information of the abandon that the relevant position picture element possibly occur is complementary, accomplish the detection of abandon in the video.
2. by the described abandon detection method of claim 1, it is characterized in that: said model modification is following:
Step S11:, utilize in this two field picture of background modeling and frame-to-frame differences point-score extraction stagnant zone in short-term to every two field picture of camera input;
Step S12: to stagnant zone in short-term, utilize abandon to have limbus and a certain size characteristic, get rid of the noise in the stagnant zone in short-term, thereby obtain to belong to the zone of abandon;
Step S13: to the RGB colouring information of the extracted region that possibly belong to abandon each picture element in should the zone and adopt the K averaging method to upgrade the mixed Gauss model of relevant position.
3. by the described abandon detection method of claim 2, it is characterized in that: said extraction stagnant zone in short-term comprises following process: utilize the method based on background modeling, detect the zone that background changes in the present image; In the zone that background changes, utilize the inter-frame difference method to detect static in short-term zone.
4. by the described abandon detection method of claim 2, it is characterized in that: utilize and extract in short-term that the picture element of stagnant zone is used to upgrade mixed Gauss model.
5. by the described abandon detection method of claim 2, it is characterized in that: adopt the mixed Gauss model of K averaging method online updating corresponding position, be used to describe the abandon colouring information.
6. by the described abandon detection method of claim 1, it is characterized in that: it is following that said abandon detects step:
Step S21: the RGB colouring information that every two field picture of camera input is extracted each picture element; And the RGB colouring information of each picture element is input in the mixed Gauss model of correspondence position; Detect the picture element of RGB colouring information and correspondence position mixed Gauss model coupling, obtain the zone of forming by the picture element of mixed Gauss model coupling;
Step S22: to the zone of forming by the picture element of coupling, utilize abandon to have limbus and a certain size characteristics, further get rid of noise region, thereby obtain the abandon zone;
Step S23:,, then report to the police if there is abandon to be detected to every two field picture of input; Otherwise, then continue the abandon testing process of execution in step S21.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101170593A CN101635026B (en) | 2008-07-23 | 2008-07-23 | Method for detecting derelict without tracking process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101170593A CN101635026B (en) | 2008-07-23 | 2008-07-23 | Method for detecting derelict without tracking process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101635026A CN101635026A (en) | 2010-01-27 |
| CN101635026B true CN101635026B (en) | 2012-05-23 |
Family
ID=41594211
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008101170593A Expired - Fee Related CN101635026B (en) | 2008-07-23 | 2008-07-23 | Method for detecting derelict without tracking process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101635026B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129688B (en) * | 2011-02-24 | 2012-09-05 | 哈尔滨工业大学 | Moving target detection method aiming at complex background |
| CN102831384B (en) * | 2011-06-13 | 2018-01-23 | 索尼公司 | The method and apparatus that abandon is detected from video |
| US8675917B2 (en) * | 2011-10-31 | 2014-03-18 | International Business Machines Corporation | Abandoned object recognition using pedestrian detection |
| CN103455997B (en) * | 2012-06-04 | 2016-05-04 | 深圳大学 | A kind of abandon detection method and system |
| CN104754179B (en) * | 2013-12-31 | 2017-11-07 | 澜起科技（上海）有限公司 | The fully-automated synthesis method and system of static nature information in dynamic image |
| JP5888348B2 (en) * | 2014-01-23 | 2016-03-22 | カシオ計算機株式会社 | Imaging apparatus, imaging control method, and program |
| CN107145861A (en) * | 2017-05-05 | 2017-09-08 | 中国科学院上海高等研究院 | A kind of abandon automatic testing method |
| CN113144591A (en) * | 2021-04-09 | 2021-07-23 | 广州三七互娱科技有限公司 | Virtual character edge drawing method and device and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1984236A (en) * | 2005-12-14 | 2007-06-20 | 浙江工业大学 | Method for collecting characteristics in telecommunication flow information video detection |
| CN101025862A (en) * | 2007-02-12 | 2007-08-29 | 吉林大学 | Video based mixed traffic flow parameter detecting method |
| CN101094413A (en) * | 2007-07-06 | 2007-12-26 | 浙江大学 | Real time movement detection method in use for video monitoring |
-
2008
- 2008-07-23 CN CN2008101170593A patent/CN101635026B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1984236A (en) * | 2005-12-14 | 2007-06-20 | 浙江工业大学 | Method for collecting characteristics in telecommunication flow information video detection |
| CN101025862A (en) * | 2007-02-12 | 2007-08-29 | 吉林大学 | Video based mixed traffic flow parameter detecting method |
| CN101094413A (en) * | 2007-07-06 | 2007-12-26 | 浙江大学 | Real time movement detection method in use for video monitoring |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101635026A (en) | 2010-01-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101635026B (en) | Method for detecting derelict without tracking process | |
| Khan et al. | A survey of advances in vision-based vehicle re-identification | |
| CN101389004B (en) | Moving target classification method based on on-line study | |
| CN101226597B (en) | Method and system for recognizing nights pedestrian based on thermal infrared gait | |
| CN109711370B (en) | Data fusion method based on WIFI detection and face clustering | |
| CN100545867C (en) | Aerial shooting traffic video frequency vehicle rapid checking method | |
| Kamthe et al. | Suspicious activity recognition in video surveillance system | |
| CN101770648B (en) | Video monitoring based loitering system and method thereof | |
| CN102411703A (en) | Device and method for detecting specific object in image sequence as well as video camera equipment | |
| CN101295405A (en) | Portrait and vehicle recognition alarming and tracing method | |
| CN104156942A (en) | Detection method for remnants in complex environment | |
| CN101727672A (en) | Method for detecting, tracking and identifying object abandoning/stealing event | |
| CN104008371A (en) | Regional suspicious target tracking and recognizing method based on multiple cameras | |
| CN101325690A (en) | Method and system for detecting human flow analysis and crowd accumulation process of monitoring video flow | |
| CN102521565A (en) | Garment identification method and system for low-resolution video | |
| CN102065275B (en) | Multi-target tracking method in intelligent video monitoring system | |
| CN109033175A (en) | A kind of method and system to scheme to search vehicle | |
| Liu et al. | Anomaly detection in surveillance video using motion direction statistics | |
| CN102136076A (en) | Method for positioning and tracing underground personnel of coal mine based on safety helmet detection | |
| CN102254394A (en) | Antitheft monitoring method for poles and towers in power transmission line based on video difference analysis | |
| Karpagavalli et al. | Estimating the density of the people and counting the number of people in a crowd environment for human safety | |
| Shi et al. | Using maximum consistency context for multiple target association in wide area traffic scenes | |
| Buch et al. | Vehicle localisation and classification in urban CCTV streams | |
| Bansal et al. | A necessary review on optical character recognition (OCR) system for vehicular applications | |
| CN102254396B (en) | Intrusion detection method and device based on video |
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: 20120523 Termination date: 20210723 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |