CN1801930A - Dubious static object detecting method based on video content analysis - Google Patents
Dubious static object detecting method based on video content analysis Download PDFInfo
- Publication number
- CN1801930A CN1801930A CN 200510061858 CN200510061858A CN1801930A CN 1801930 A CN1801930 A CN 1801930A CN 200510061858 CN200510061858 CN 200510061858 CN 200510061858 A CN200510061858 A CN 200510061858A CN 1801930 A CN1801930 A CN 1801930A
- Authority
- CN
- China
- Prior art keywords
- video
- static
- dubious
- frame
- detection
- 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.)
- Granted
Links
Landscapes
- Image Analysis (AREA)
Abstract
Present invention relates to content analysis based questionable quiescence object detection method. It utilizes motional detection of previous background frame successive frame detecting moving target, doing target tracking through connected domain marking, identifying questionable quiescence object utilizing pattern recognition technique. It contains A, object model video target detection, B, target tracking, C, video target signature calculation and quiescence dubiety detecting doing quiescence questionable object detecting according to features and quiescence questionable object point, according to system maintenance video object model calculating video objective features. Said invention can be used to video monitoring.
Description
Technical field
The present invention relates to a kind of detection method of Dubious static object, specifically is a kind of Dubious static object detecting method based on video content analysis.Be applicable to the video surveillance applications field.
Background technology
Current intelligent monitor system generally is by limiting suitable application area, in particular, limiting the classification of object and assist the Security Officer to work by the incident of safety regulation regulation.The video content analysis method that is used for intelligent monitor system is made up of three main portions: target detection, target following and target classification and identification.
Because follow-up process depends on accurate target and follows the tracks of very much, correct detection and tracking target is very important.Though have many target detection, algorithm of target tracking to be studied and to deliver, have only less several products that can be used to use one-level.For example, the AlertVideo product of the Northrop Grumman company in Los Angeles has used one to subtract the target detection track algorithm of cutting based on adaptive background, used the corresponding method in edge and zone to obtain foreground target afterwards, its background model is made up of the Gauss model of one group of RGB, to reach the influence of the variation of removing light slowly, rain, snow etc.Background subtract cut after, by the region method of connected domain and surrounded by edges, candidate pixel can be formed the target area.Afterwards, the translational speed of object, how much feature of color composition and some is used to tracking target in the frame that ensues mutually.Each object that is detected is assigned with a unique ID number.
Target identification in the rest image and classification are positive research fields in recent decades, have used for reference the method for rest image target identification in the nearest video object identification.There are some video object recognition systems to use based on the method for learning in the hope of obtaining maximum flexibility.
Find that by literature search Changick Kim etc. provides the moving object detection based on successive frames in " Object-based video abstraction for video surveillancesystems " (IEEE Transactions on circuits and systems for video technology.Vol.12.No.12.Dec2002) literary composition; Shao-Yi Chen etc. has proposed the moving object detection based on preceding background frames in " Efficient moving objectsegmentation algorithm using background registration Technique " literary composition, willow hall etc. has proposed the background frames maintaining method based on each pixel Gaussian distribution in " based on the video object segmentation method of motion detection " (Chinese patent, patent No. 03151406.5).
Summary of the invention
The technical problem to be solved in the present invention is: a kind of Dubious static object detecting method based on video content analysis is provided, that purpose is to use is preceding, the motion detection of background frames and successive frames, to detect moving target, do target following by connected component labeling, last application mode recognition technology identification Dubious static object.
The technical solution adopted in the present invention is: based on the Dubious static object detecting method of video content analysis, its step is as follows:
A) video object detects, and by the object video that the motion detection in the successive frames obtains moving, is not belonged to the object video of background by the frame difference of prospect frame and background frames;
B) video frequency object tracking merges the result that two kinds of video objects detect, and the zone that obtains is made marks, and getting in touch between the object video in the searching present frame and the object video of system maintenance is in order to the object video model in the maintenance system;
C) detection of the calculating of video object feature and static suspicious item according to the feature of the object video Model Calculation video object of system maintenance, is done the detection of static suspicious item according to the characteristics of feature and static suspicious item.
Described video object detects and is divided into two kinds of differences but complementary method, i.e. the object that obtains moving by the motion detection in the successive frames is not belonged to the object of background by the frame difference of prospect frame and background frames; Wherein, the motion detection in the successive frames has been utilized the movement edge feature, and main process has the Canny edge to ask for, movement edge is asked for, moving target is asked for.
Frame difference detection before described, in the background frames is to compare with the reference frame of no any target object by the image of relatively input, just can obtain the difference of two two field pictures, the zone at this difference place comprises all zones different with the background frames color, the object that had both comprised motion also comprises static object.
The result of two kinds of video object segmentation is merged, and promptly by successive frames and zone preceding, that the background frames motion detection obtains are sought common ground, doing mathematics morphology is handled then, can obtain two video object segmentation results' fusion.
Described video frequency object tracking has comprised that to the video segmentation result after merging, with the way mark connected region of following the tracks of, each connected region is thought an object video; Video frequency object tracking has also comprised object video set corresponding of object video in the present frame and system maintenance, when first frame of video segmentation result is arranged, all object videos are created corresponding structure, in each calculating frame thereafter, at first seek and its size according to the position of Kalman filter forecasting by each object, the object video that features such as grey level histogram are close, found object video is removed from the present frame object video, in present frame object video set, seek the nearer system video object that is not mated of being separated by with it then in the remaining object and do coupling, be considered to emerging object video according to remaining present frame object video afterwards, in system, do corresponding building work, for the remaining object video that does not have in the matching system, do the object existence work that stops.
The calculating of described video object feature and the detection of static suspicious item comprise that shape information obtains, according to the object video of safeguarding in the system that upgraded with the object video in the present frame, utilization is based on the descriptor of area, describe the shape of video image, utilization Hu Moment is as the characteristic vector of shape description.
The calculating of described video object feature and the detection of static suspicious item comprise that velocity information obtains, and for each object video, adopt Kalman filtering to obtain the average movement velocity of the object video barycenter followed the tracks of.
The calculating of described video object feature and the detection of static suspicious item comprise the Dubious static object judgement, Dubious static object is that certain effective area scope is arranged, by other static object that surpasses certain hour in moving object that enters from border division back, in view of the above, according to Hu Moment, the duration of record that object enters, divides and the very little state of speed, make the judgement of static suspicious item.
The process that described movement edge is asked for is: it is poor that the edge image of in succession two frame video images is done, and to eliminate the influence of static scene, two two field pictures of establishing in succession are respectively f
nAnd f
N-1, then movement edge can be defined as:
|φ(f
n-1)-φ(f
n)|=|θ(G*f
n-1)-θ(G*f
n)|
Wherein G is a Gauss operator, and * is a convolution, and is a gradient operator, and θ is the edge detection operator of canny;
The process that described moving target is asked for is: in the movement edge image that obtains, the object of motion can stay the next edge line of sealing basically, and the zone that the line between first and last marginal point in each row is formed calls horizontal candidate region; Same, the zone that the line between first in each row and last marginal point is formed is called vertical candidate region; Same, the scan line of positive and negative 45 degree also can obtain two candidate regions; By these candidate regions are asked also, then morphology is done in the zone that obtains and handled, promptly obtain object video testing result based on successive frames.
Comprised during frame difference before described, in the background frames detects and asked color distinction, the difference of prospect frame and background frames need be calculated by the color distinction of pixel, compare color spaces such as RGB, YUV, HSV calculating pixel color distinction is more suitable, and the HSV value of establishing two pixels is respectively (H
1, S
1, V
1), (H
2, S
2, V
2), consider the characteristics in HSV space, the discrimination formula of the color distinction of Cai Yonging is here:
|(H
1-H
2)|*|(S
1-S
2)|>Th
hsor|V
1-V
2|>Th
v
Th wherein
HsAnd Th
vIt is corresponding threshold values.
The invention has the beneficial effects as follows: the technology of target detection that the present invention is integrated, target following, pattern recognition aspect, motion detection in conjunction with preceding, background frames and successive frames, to detect moving target, do target following by connected component labeling, last application mode recognition technology identification Dubious static object.The present invention can improve the automaticity of airport, harbour, station and other public place security monitoring, makes the Security Officer can monitor more place more easily, improves efficiency for monitoring, reduces the possibility of omission.
Description of drawings
Fig. 1 is a step block diagram of the present invention.
Fig. 2 is the exploded view that moving target is asked for.
Embodiment
As shown in Figure 1, this implementation method mainly is divided into three big steps:
The first step, video object detects 1, and it is divided into two kinds of differences but complementary method: by the object video (object) that the motion detection in the successive frames obtains moving, do not belonged to the object video (object) of background by the frame difference of prospect frame and background frames.
Motion detection in the successive frames has been utilized the movement edge feature, and main process has the Canny edge to ask for, and movement edge is asked for, and moving target is asked for.
Referring to Fig. 2, the process that Canny asks at the edge 1-1 is that at first image to be done Gaussian convolution level and smooth, then uses the non-maximum of Grad to constrain the refinement edge, at last the weak edge adding edge image that will link to each other with strong edge with the threshold values that lags behind.
The process that movement edge is asked for 1-2 be the edge image of in succession two frame video images is done poor, to eliminate the influence of static scene.If two two field pictures in succession are respectively f
nAnd f
N-1, then movement edge can be defined as:
|φ(f
n-1)-φ(f
n)|=|θ(G*f
n-1)-θ(G*f
n)|
Wherein G is a Gauss operator, and * is a convolution, and is a gradient operator, and θ is the edge detection operator of canny.
The process that moving target is asked for 1-3 is in the movement edge image that obtains, and the object of motion can stay the next edge line of sealing basically.The zone that line between first and last marginal point in each row is formed is called horizontal candidate region, and is same, and the zone that the line between first in each row and last marginal point is formed is called vertical candidate region.Same, the scan line of positive and negative 45 degree also can obtain two candidate regions.By these candidate regions are asked also, then morphology is done in the zone that obtains and handled, can obtain object video testing result based on successive frames.
(a) expression movement edge among Fig. 2 (b) is represented horizontal candidate region, (c) represents vertical candidate region, (d) expression morphology processed video object detection result.
In addition, by the image (prospect frame) of relatively input and the reference frame (background frames) of no any target object, can obtain the difference of two two field pictures, the zone at this difference place comprises all zones different with the background frames color, the object that had both comprised motion also comprises static object.
Background frames safeguards that 1-4 takes the method for Gaussian Profile array:
The background model characteristic value adopts the YC value RGB of pixel, wherein I
Ij=(R
Ij, G
Ij, B
Ij) represent the rgb value on j frame, the i pixel, distributed model is described as:
The estimated value of the probability density function p (x) of background distributions
Satisfy:
1. in the formula, on rgb space, it is the input feature value on a certain frame pixel that each pixel supposition has N Gaussian Profile, x
x=(R,G,B)
T
Be the weight of i Gaussian Profile of this pixel, wherein μ
iBe the average of i Gaussian Profile, μ
i=(μ
IR, μ
IG, μ
IB)
T, σ
iBe the mean square deviation of i Gaussian Profile, σ
i=(σ
IR, σ
IG, σ
IB)
T
Initial model, the value of input feature value of using first frame exactly be as the average of each Gaussian Profile in the model, as mean square deviation, and supposes that the weights of first distribution are 1 with predefined value, and all the other are 0.
Read current video frame by video frequency collection card, adjust the weight and the parameter of the single Gaussian Profile of coupling in real time with new video sampling value, the real background that comes new model more to approach after the variation distributes, and its matching criterior is:
|x-μ
i|<τσ
i,
And simultaneously
Hour, be only coupling,
The parameter update that coupling distributes is followed following formula:
μ
i(t)=(1-α)μ
i(t-1)+αx(t) ③
σ
i(t)=((1-β)σ
i 2(t-1)+β(x(t)-μ
i(t))
2)
④
The size of factor-alpha has characterized the influence size of the sampled value of time distance to the background object state, and the size of β has then mainly characterized the speed that video camera self parameter changes;
Distribution of weights is upgraded and is followed:
When new sampling and i distribution coupling, S (t)=1, otherwise S (t)=0, the size of factor gamma has reflected the sensitivity that background model changes background object;
When new value does not match, under the certain situation of distribution number N, give up the Gaussian Profile of weight minimum, replace with new distribution, and initializes weights is
Simultaneously other weights are done normalized:
I ≠ min.
The average of choosing the Gaussian Profile of weight maximum in the background model of current each pixel is as maintained background.
When safeguarding background frames, do not upgrade the zone at the object video place of system maintenance cited below.
Ask color distinction 1-5, the difference of prospect frame and background frames need be calculated by the color distinction of pixel, compares color spaces such as RGB, YUV, and HSV calculating pixel color distinction is more suitable.If the HSV value of two pixels is respectively (H
1, S
1, V
1), (H
2, S
2, V
2), consider the characteristics in HSV space, the discrimination formula of the color distinction of Cai Yonging is here:
|(H
1-H
2)|*|(S
1-S
2)|>Th
hs?or|V
1-V
2|>Th
v
Th wherein
HsAnd Th
vIt is corresponding threshold values.
Frame difference target is asked for 1-6, and the picture element of prospect frame and background frames correspondence is judged whether it is that frame is not good enough according to color distinction.Because camera has random noise, tiny noise spot can appear on the frame difference image, by the opening operation in the mathematical morphology, can eliminate these noise spots.
Second step, video frequency object tracking 2, by successive frames and zone preceding, that the background frames motion detection obtains are sought common ground, doing mathematics morphology is handled then, can obtain two kinds of video object segmentation results' fusion 2-1.
For the video segmentation result after merging, with the way mark connected region of following the tracks of, each connected region is thought an object video, and this step is called object video mark 2-2.
Following step is the corresponding 2-3 with system object of detected object (object video in the present frame is corresponding with the object video of system maintenance): when first frame of video segmentation result was arranged, all object videos were created corresponding structure.In each calculating frame thereafter, at first seek and its size according to the position of Kalman filter forecasting by each object, the object video that features such as grey level histogram are close, found object video is removed from the present frame object video; In present frame object video set, seek the nearer system video object that is not mated of being separated by with it then in the remaining object and do coupling; Be considered to emerging object video according to remaining present frame object video afterwards, in system, do corresponding building work; For the remaining object video that does not have in the matching system, do the object existence work that stops.
In the 3rd step, the calculating of video object feature and the detection of static suspicious item 3 specifically comprise:
Shape information is obtained 3-1, according to the object video of safeguarding in the system that upgraded with the object video in the present frame, utilization can be described the shape of video image based on the descriptor (region-based descriptor) of area, and utilization Hu Moment is as the characteristic vector of shape description.
Velocity information is obtained 3-2, and for each object video, the average movement velocity of its barycenter is a key parameter of judging that object video is whether static.This method adopts Kalman filtering to obtain the average movement velocity of the object video barycenter followed the tracks of.
Dubious static object is judged 3-3, Dubious static object is that certain effective area scope is arranged, by other static object that surpasses certain hour in moving object that enters from border division back, in view of the above, according to Hu Moment, the duration of record that object enters, divides and the very little state of speed, make the judgement of static suspicious item.
By the moving object detection in conjunction with preceding, background frames and successive frames, the present invention can the comparatively stable moving object detection with in the visual field come out, and keeps the integrality of target to a great extent.And environment can adapt to following specific background automatically to be changed: the 1) variation of illumination condition; 2) rule of background object state changes: as the screen of indoor flicker etc.; 3) variation of camera self-condition: the camera lens that causes as external force slightly rocks; 4) conversion of foreground object state: as moving in and out of background object.In addition, by the target detection result is done connected component labeling, and the set of the object video of maintenance system, can overcome the situation that a target is divided into a plurality of parts to a great extent.
Claims (10)
1, a kind of Dubious static object detecting method based on video content analysis is characterized in that method step is as follows:
A) video object detects (1), by the object video that the motion detection in the successive frames obtains moving, is not belonged to the object video of background by the frame difference of prospect frame and background frames;
B) video frequency object tracking (2) merges the result that two kinds of video objects detect, and the zone that obtains is made marks, and getting in touch between the object video in the searching present frame and the object video of system maintenance is in order to the object video model in the maintenance system;
C) detection of the calculating of video object feature and static suspicious item (3) according to the feature of the object video Model Calculation video object of system maintenance, is done the detection of static suspicious item according to the characteristics of feature and static suspicious item.
2, the Dubious static object detecting method based on video content analysis according to claim 1, it is characterized in that: described video object detects (1) and is divided into two kinds of differences but complementary method, the i.e. object that obtains moving by the motion detection in the successive frames is not belonged to the object of background by the frame difference of prospect frame and background frames; Wherein, the motion detection in the successive frames has been utilized the movement edge feature, and main process has the Canny edge to ask for, movement edge is asked for, moving target is asked for.
3, the Dubious static object detecting method based on video content analysis according to claim 1 and 2, it is characterized in that: the frame difference detection before described, in the background frames is to compare with the reference frame of no any target object by the image of relatively input, just can obtain the difference of two two field pictures, the zone at this difference place comprises all zones different with the background frames color, the object that had both comprised motion also comprises static object.
4, the Dubious static object detecting method based on video content analysis according to claim 1 and 2, it is characterized in that: the result of two kinds of video object segmentation is merged, promptly by successive frames and zone preceding, that the background frames motion detection obtains are sought common ground, doing mathematics morphology is handled then, can obtain two video object segmentation results' fusion.
5, the Dubious static object detecting method based on video content analysis according to claim 1, it is characterized in that: described video frequency object tracking (2) has comprised the video segmentation result after merging, with the way mark connected region of following the tracks of, each connected region is thought an object video; Video frequency object tracking (2) has also comprised object video set corresponding of the object video in the present frame and system maintenance, when first frame of video segmentation result is arranged, all object videos are created corresponding structure, in each calculating frame thereafter, at first seek and its size according to the position of Kalman filter forecasting by each object, the object video that features such as grey level histogram are close, found object video is removed from the present frame object video, in present frame object video set, seek the nearer system video object that is not mated of being separated by with it then in the remaining object and do coupling, be considered to emerging object video according to remaining present frame object video afterwards, in system, do corresponding building work, for the remaining object video that does not have in the matching system, do the object existence work that stops.
6, the Dubious static object detecting method based on video content analysis according to claim 1, it is characterized in that: the calculating of described video object feature and the detection of static suspicious item (3) comprise that shape information obtains, according to the object video of safeguarding in the system that upgraded with the object video in the present frame, utilization is based on the descriptor of area, describe the shape of video image, utilization Hu Moment is as the characteristic vector of shape description.
7, the Dubious static object detecting method based on video content analysis according to claim 1, it is characterized in that: the calculating of described video object feature and the detection of static suspicious item (3) comprise that velocity information obtains, for each object video, adopt Kalman filtering to obtain the average movement velocity of the object video barycenter followed the tracks of.
8, the Dubious static object detecting method based on video content analysis according to claim 1, it is characterized in that: the calculating of described video object feature and the detection of static suspicious item (3) comprise the Dubious static object judgement, Dubious static object is that certain effective area scope is arranged, by other static object that surpasses certain hour in moving object that enters from border division back, in view of the above, according to Hu Moment, the duration of record that object enters, divides and the very little state of speed, make the judgement of static suspicious item.
9, the Dubious static object detecting method based on video content analysis according to claim 2, it is characterized in that: the process that described movement edge is asked for is: it is poor that the edge image of in succession two frame video images is done, to eliminate the influence of static scene, two two field pictures of establishing in succession are respectively f
nAnd f
N-1, then movement edge can be defined as:
|φ(f
n-1)-φ(f
n)|=|θ(G*f
n-1)-θ(G*f
n)|
Wherein G is a Gauss operator, and * is a convolution, and is a gradient operator, and θ is the edge detection operator of canny;
The process that described moving target is asked for is: in the movement edge image that obtains, the object of motion can stay the next edge line of sealing basically, and the zone that the line between first and last marginal point in each row is formed calls horizontal candidate region; Same, the zone that the line between first in each row and last marginal point is formed is called vertical candidate region; Same, the scan line of positive and negative 45 degree also can obtain two candidate regions; By these candidate regions are asked also, then morphology is done in the zone that obtains and handled, promptly obtain object video testing result based on successive frames.
10, the Dubious static object detecting method based on video content analysis according to claim 1 and 2, it is characterized in that: comprised during the frame difference before described, in the background frames detects and asked color distinction, the difference of prospect frame and background frames need be calculated by the color distinction of pixel, compare color spaces such as RGB, YUV, HSV calculating pixel color distinction is more suitable, and the HSV value of establishing two pixels is respectively (H
1, S
1, V
1), (H
2, S
2, V
2), consider the characteristics in HSV space, the discrimination formula of the color distinction of Cai Yonging is here:
|(H
1-H
2)|*|(S
1-S
2)|>Th
hs?or|V
1-V
2|>Th
v
Th wherein
HsAnd Th
vIt is corresponding threshold values.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200510061858A CN100589561C (en) | 2005-12-06 | 2005-12-06 | Dubious static object detecting method based on video content analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200510061858A CN100589561C (en) | 2005-12-06 | 2005-12-06 | Dubious static object detecting method based on video content analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN1801930A true CN1801930A (en) | 2006-07-12 |
CN100589561C CN100589561C (en) | 2010-02-10 |
Family
ID=36811644
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN200510061858A Expired - Fee Related CN100589561C (en) | 2005-12-06 | 2005-12-06 | Dubious static object detecting method based on video content analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN100589561C (en) |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN100527842C (en) * | 2007-01-26 | 2009-08-12 | 清华大学 | Background-based motion estimation coding method |
CN101751669A (en) * | 2009-12-17 | 2010-06-23 | 北京中星微电子有限公司 | Static object detection method and device |
CN101765025A (en) * | 2008-12-23 | 2010-06-30 | 北京中星微电子有限公司 | System for abnormal detection of surveillance camera and method thereof |
CN102034240A (en) * | 2010-12-23 | 2011-04-27 | 北京邮电大学 | Method for detecting and tracking static foreground |
CN102105904A (en) * | 2008-08-11 | 2011-06-22 | 欧姆龙株式会社 | Detection information registration device, object detection device, electronic device, method for controlling detection information registration device, method for controlling object detection device, program for controlling detection information registration device |
CN102314591A (en) * | 2010-07-09 | 2012-01-11 | 株式会社理光 | Method and equipment for detecting static foreground object |
WO2012016374A1 (en) * | 2010-08-03 | 2012-02-09 | Empire Technology Development Llc | Method for identifying objects in video |
CN102402273A (en) * | 2010-09-07 | 2012-04-04 | 原相科技股份有限公司 | Interactive control system, dynamic detection method of object, host device and control method thereof |
CN101299812B (en) * | 2008-06-25 | 2012-12-05 | 北京中星微电子有限公司 | Method, system for analyzing, storing video as well as method, system for searching video |
CN101587620B (en) * | 2008-05-21 | 2013-01-02 | 上海新联纬讯科技发展有限公司 | Method for detecting stationary object based on visual monitoring |
CN101764923B (en) * | 2008-12-23 | 2013-03-27 | 北京中星微电子有限公司 | Method and device for eliminating noise in motion detection |
US8456528B2 (en) | 2007-03-20 | 2013-06-04 | International Business Machines Corporation | System and method for managing the interaction of object detection and tracking systems in video surveillance |
CN103729844A (en) * | 2013-12-20 | 2014-04-16 | 乐视致新电子科技(天津)有限公司 | Motionlessness detecting method and device |
CN103839255A (en) * | 2013-12-05 | 2014-06-04 | 福建师范大学 | Method and device for detecting video image matting tampering |
CN104038792A (en) * | 2014-06-04 | 2014-09-10 | 紫光软件***有限公司 | Video content analysis method and device for IPTV (Internet Protocol Television) supervision |
CN105550692A (en) * | 2015-12-30 | 2016-05-04 | 南京邮电大学 | Unmanned aerial vehicle automatic homing landing method based on landmark color and outline detection |
CN106033616A (en) * | 2015-03-17 | 2016-10-19 | 联想(北京)有限公司 | Electronic equipment and image processing method |
CN106408577A (en) * | 2016-09-21 | 2017-02-15 | 安徽慧视金瞳科技有限公司 | Successive frame connected domain parallel tagging method used for projection interactive system |
CN106846357A (en) * | 2016-12-15 | 2017-06-13 | 重庆凯泽科技股份有限公司 | A kind of suspicious object detecting method and device |
CN110222575A (en) * | 2019-05-07 | 2019-09-10 | 杭州智尚云科信息技术有限公司 | Overall equipment efficiency.comprehensive efficiency of equipment analysis method, device, system, equipment and storage medium based on attention mechanism target detection |
CN111353954A (en) * | 2020-02-20 | 2020-06-30 | 新疆大学 | Video image processing method and device and electronic equipment |
WO2021087773A1 (en) * | 2019-11-05 | 2021-05-14 | 深圳市欢太科技有限公司 | Recognition method and apparatus, electronic device, and storage medium |
CN113689472A (en) * | 2021-10-26 | 2021-11-23 | 城云科技(中国)有限公司 | Moving target detection method, device and application |
CN114067367A (en) * | 2022-01-14 | 2022-02-18 | 南京甄视智能科技有限公司 | Small animal detection method in infrared scene, computer equipment and storage medium |
-
2005
- 2005-12-06 CN CN200510061858A patent/CN100589561C/en not_active Expired - Fee Related
Cited By (36)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN100527842C (en) * | 2007-01-26 | 2009-08-12 | 清华大学 | Background-based motion estimation coding method |
US8456528B2 (en) | 2007-03-20 | 2013-06-04 | International Business Machines Corporation | System and method for managing the interaction of object detection and tracking systems in video surveillance |
CN101587620B (en) * | 2008-05-21 | 2013-01-02 | 上海新联纬讯科技发展有限公司 | Method for detecting stationary object based on visual monitoring |
CN101299812B (en) * | 2008-06-25 | 2012-12-05 | 北京中星微电子有限公司 | Method, system for analyzing, storing video as well as method, system for searching video |
US8774456B2 (en) | 2008-08-11 | 2014-07-08 | Omron Corporation | Detective information registration device and target object detection device for detecting an object in an image |
CN102105904B (en) * | 2008-08-11 | 2014-06-25 | 欧姆龙株式会社 | Detection information registration device, object detection device, electronic device, method for controlling detection information registration device, method for controlling object detection device, program for controlling detection information registration device |
CN102105904A (en) * | 2008-08-11 | 2011-06-22 | 欧姆龙株式会社 | Detection information registration device, object detection device, electronic device, method for controlling detection information registration device, method for controlling object detection device, program for controlling detection information registration device |
CN101764923B (en) * | 2008-12-23 | 2013-03-27 | 北京中星微电子有限公司 | Method and device for eliminating noise in motion detection |
CN101765025A (en) * | 2008-12-23 | 2010-06-30 | 北京中星微电子有限公司 | System for abnormal detection of surveillance camera and method thereof |
CN101751669B (en) * | 2009-12-17 | 2014-03-26 | 北京中星微电子有限公司 | Static object detection method and device |
CN101751669A (en) * | 2009-12-17 | 2010-06-23 | 北京中星微电子有限公司 | Static object detection method and device |
CN102314591B (en) * | 2010-07-09 | 2014-07-23 | 株式会社理光 | Method and equipment for detecting static foreground object |
CN102314591A (en) * | 2010-07-09 | 2012-01-11 | 株式会社理光 | Method and equipment for detecting static foreground object |
WO2012016374A1 (en) * | 2010-08-03 | 2012-02-09 | Empire Technology Development Llc | Method for identifying objects in video |
US8873801B2 (en) | 2010-08-03 | 2014-10-28 | Empire Technology Development Llc | Identification of objects in a video |
CN102402273A (en) * | 2010-09-07 | 2012-04-04 | 原相科技股份有限公司 | Interactive control system, dynamic detection method of object, host device and control method thereof |
CN102034240A (en) * | 2010-12-23 | 2011-04-27 | 北京邮电大学 | Method for detecting and tracking static foreground |
CN103839255A (en) * | 2013-12-05 | 2014-06-04 | 福建师范大学 | Method and device for detecting video image matting tampering |
CN103839255B (en) * | 2013-12-05 | 2017-03-01 | 福建师范大学 | Video keying altering detecting method and device |
CN103729844A (en) * | 2013-12-20 | 2014-04-16 | 乐视致新电子科技(天津)有限公司 | Motionlessness detecting method and device |
CN103729844B (en) * | 2013-12-20 | 2017-01-11 | 乐视致新电子科技(天津)有限公司 | Motionlessness detecting method and device |
CN104038792A (en) * | 2014-06-04 | 2014-09-10 | 紫光软件***有限公司 | Video content analysis method and device for IPTV (Internet Protocol Television) supervision |
CN104038792B (en) * | 2014-06-04 | 2017-06-16 | 紫光软件***有限公司 | For the video content analysis method and apparatus of IPTV supervision |
CN106033616B (en) * | 2015-03-17 | 2020-01-31 | 联想(北京)有限公司 | Electronic device and image processing method |
CN106033616A (en) * | 2015-03-17 | 2016-10-19 | 联想(北京)有限公司 | Electronic equipment and image processing method |
CN105550692A (en) * | 2015-12-30 | 2016-05-04 | 南京邮电大学 | Unmanned aerial vehicle automatic homing landing method based on landmark color and outline detection |
CN105550692B (en) * | 2015-12-30 | 2018-08-31 | 南京邮电大学 | The homing vector landing concept of unmanned plane based on marker color and contour detecting |
CN106408577A (en) * | 2016-09-21 | 2017-02-15 | 安徽慧视金瞳科技有限公司 | Successive frame connected domain parallel tagging method used for projection interactive system |
CN106408577B (en) * | 2016-09-21 | 2019-12-31 | 安徽慧视金瞳科技有限公司 | Continuous frame connected domain parallel marking method for projection interactive system |
CN106846357A (en) * | 2016-12-15 | 2017-06-13 | 重庆凯泽科技股份有限公司 | A kind of suspicious object detecting method and device |
CN110222575A (en) * | 2019-05-07 | 2019-09-10 | 杭州智尚云科信息技术有限公司 | Overall equipment efficiency.comprehensive efficiency of equipment analysis method, device, system, equipment and storage medium based on attention mechanism target detection |
CN110222575B (en) * | 2019-05-07 | 2021-06-29 | 杭州智尚云科信息技术有限公司 | Equipment comprehensive efficiency analysis method, device, system, equipment and storage medium based on attention mechanism target detection |
WO2021087773A1 (en) * | 2019-11-05 | 2021-05-14 | 深圳市欢太科技有限公司 | Recognition method and apparatus, electronic device, and storage medium |
CN111353954A (en) * | 2020-02-20 | 2020-06-30 | 新疆大学 | Video image processing method and device and electronic equipment |
CN113689472A (en) * | 2021-10-26 | 2021-11-23 | 城云科技(中国)有限公司 | Moving target detection method, device and application |
CN114067367A (en) * | 2022-01-14 | 2022-02-18 | 南京甄视智能科技有限公司 | Small animal detection method in infrared scene, computer equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN100589561C (en) | 2010-02-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN1801930A (en) | Dubious static object detecting method based on video content analysis | |
Cao et al. | Vehicle detection and motion analysis in low-altitude airborne video under urban environment | |
WO2018130016A1 (en) | Parking detection method and device based on monitoring video | |
Ge et al. | Real-time pedestrian detection and tracking at nighttime for driver-assistance systems | |
Chen et al. | Efficient hierarchical method for background subtraction | |
Hsieh et al. | Shadow elimination for effective moving object detection by Gaussian shadow modeling | |
US10127448B2 (en) | Method and system for dismount detection in low-resolution UAV imagery | |
CN109918971B (en) | Method and device for detecting number of people in monitoring video | |
CN102915433A (en) | Character combination-based license plate positioning and identifying method | |
Lee et al. | Human detection using histogram of oriented gradients and human body ratio estimation | |
Cai et al. | Real-time arrow traffic light recognition system for intelligent vehicle | |
CN101719280B (en) | Method for detecting petty infrared target | |
Bullkich et al. | Moving shadow detection by nonlinear tone-mapping | |
Su et al. | A new local-main-gradient-orientation HOG and contour differences based algorithm for object classification | |
Teutsch et al. | Evaluation of object segmentation to improve moving vehicle detection in aerial videos | |
Sheng et al. | Real-time anti-interference location of vehicle license plates using high-definition video | |
Miller et al. | Person tracking in UAV video | |
Cao et al. | Visual attention accelerated vehicle detection in low-altitude airborne video of urban environment | |
CN104200455A (en) | Key posture extraction method based on motion statistical feature analysis | |
Satish et al. | Edge assisted fast binarization scheme for improved vehicle license plate recognition | |
Jia et al. | A novel moving cast shadow detection of vehicles in traffic scene | |
Chandrasekhar et al. | A survey of techniques for background subtraction and traffic analysis on surveillance video | |
Angeline et al. | Tracking and localisation of moving vehicle license plate via Signature Analysis | |
Liu et al. | Shadow Elimination in Traffic Video Segmentation. | |
CN111583341A (en) | Pan-tilt camera displacement detection method |
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 | ||
C17 | Cessation of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100210 Termination date: 20111206 |