CN102054277A - Method and system for detecting moving target, and video analysis system - Google Patents
Method and system for detecting moving target, and video analysis system Download PDFInfo
- Publication number
- CN102054277A CN102054277A CN2009101100345A CN200910110034A CN102054277A CN 102054277 A CN102054277 A CN 102054277A CN 2009101100345 A CN2009101100345 A CN 2009101100345A CN 200910110034 A CN200910110034 A CN 200910110034A CN 102054277 A CN102054277 A CN 102054277A
- Authority
- CN
- China
- Prior art keywords
- background model
- gray scale
- gradient
- module
- initial
- 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
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000004458 analytical method Methods 0.000 title claims abstract description 13
- 238000001514 detection method Methods 0.000 claims abstract description 69
- 238000001914 filtration Methods 0.000 claims description 10
- 230000000877 morphologic effect Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 238000013398 bayesian method Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention is suitable for the technical field of monitoring, and provides a method and a system for detecting a moving target, and a video analysis system. The method comprises the following steps of: establishing a gray scale background model and a gradient background model according to pre-acquired gray scale and gradient information; and respectively detecting the moving target according to the gray scale background model and the gradient background model, and combining and segmenting results of detecting the moving target based on the gray scale background model and the gradient background model to obtain the moving target. In the embodiment of the invention, the gray scale background model and the gradient background model are established according to the pre-acquired gray scale and gradient information; and the moving target is respectively detected according to the gray scale background model and the gradient background model, the results of detecting the moving target based on the gray scale background model and the gradient background model are combined and segmented to obtain the moving target, so that the detection accuracy of the moving target is improved, and high applicability is achieved.
Description
Technical field
The invention belongs to the monitoring technique field, relate in particular to a kind of motion target detection method, system and video analytic system.
Background technology
The intelligent video analysis technology of being mentioned in the field of video monitoring is the combination of image voice technology, computer technology, mechanics of communication and multimedia storage development technique.Mainly be meant by computing machine the vision signal of continuous acquisition is handled, detect moving target wherein, and to moving target classify, subsequent treatment such as tracking, identification, and then the behavior of target analyzed, see if there is cross the border, more line, object is left over or incident such as theft takes place, when having detected such incident and take place, just send alerting signal and remind the monitor staff to note or take appropriate measures.
Moving object detection is the prerequisite and the basis of whole intelligent video analysis technology, and moving object detection whether accurate directly has influence on the order of accuarcy of characteristic matching, tracking and the behavioural analysis of the extraction of target signature and succeeding target.At the difference of monitoring environment, moving target detecting method can be divided into two kinds: a kind of is moving object detection under the static situation of video camera, and another kind is the moving object detection under the camera motion situation.At present, have a lot, generally speaking be divided into: background subtraction, frame-to-frame differences point-score and optical flow method etc. for the moving target detecting method under the video camera fixation case.Because background subtraction can provide more information about moving target than additive method, so be widely used in the moving object detection.
Background subtraction has been subjected to both at home and abroad scholar and researchist's extensive concern in a large number in recent years, benefit our pursuits also having carried out some aspect moving object detection theoretical research and the practical application, have much about the article of the moving object detection aspect that subtracts based on background and deliver, mainly contain based on single Gauss, based on mixed Gaussian, based on statistics with based on Bayesian method etc., but the accuracy that these methods detect is not enough, maybe can not reach the real-time requirement of detection.The background interference of especially complicated monitoring scene as the influence of weather conditions such as leaf swing, water wave fluctuation, sleet, illumination variation, usually makes the accuracy of detection of these algorithms reduce, and makes the applicability of algorithm reduce.
Summary of the invention
The purpose of the embodiment of the invention is to provide a kind of motion target detection method, the problem that the applicability that is intended to solve motion target detection algorithm in the prior art is lower, accuracy of detection is low.
The embodiment of the invention is achieved in that a kind of motion target detection method, and described method comprises the steps:
According to gray scale of obtaining in advance and gradient information, set up gray scale background model and gradient background model;
Carry out motion target detection respectively according to gray scale background model and gradient background model;
To combine based on the result that gray scale background model and gradient background model are carried out moving object detection, cut apart and obtain moving target.
Another purpose of the embodiment of the invention is to provide a kind of motion target detection system, and described system comprises:
Background model is set up module, is used for setting up gray scale background model and gradient background model according to the gray scale and the gradient information that obtain in advance;
Detection module is used for setting up gray scale background model and the gradient background model that module sets up according to described background model and carries out motion target detection; And
Cut apart module, be used for described detection module is carried out respectively combining based on the result of described gray scale background model and gradient background model moving object detection, cut apart and obtain moving target.
Another purpose of the embodiment of the invention is to provide a kind of video analytic system, and described video analytic system comprises the motion target detection system.
In embodiments of the present invention, according to gray scale of obtaining in advance and gradient information, set up gray scale background model and gradient background model; Carry out motion target detection respectively according to described gray scale background model and gradient background model; Obtain moving target with respectively the result who carries out moving object detection based on described gray scale background model and gradient background model being carried out combination, cutting apart, improved the precision of moving object detection, the applicability height.
Description of drawings
Fig. 1 is the realization flow figure of the motion target detection method that provides of the embodiment of the invention;
Fig. 2 is gray scale and the gradient information that basis that the embodiment of the invention provides is obtained in advance, sets up the realization flow figure of gray scale background model and gradient background model;
Fig. 3 is the structured flowchart of the motion target detection system that provides of first embodiment of the invention;
Fig. 4 is the structured flowchart that background model that the embodiment of the invention provides is set up module;
Fig. 5 is the structured flowchart of the motion target detection system that provides of second embodiment of the invention.
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer,, the present invention is further elaborated below in conjunction with drawings and Examples.Should be appreciated that specific embodiment described herein only in order to explanation the present invention, and be not used in qualification the present invention.
In embodiments of the present invention, according to gray scale of obtaining in advance and gradient information, set up gray scale background model and gradient background model; Carry out motion target detection according to described gray scale background model and gradient background model; The result who carries out moving object detection based on described gray scale background model and gradient background model is respectively carried out combination, cut apart and obtain moving target.
Fig. 1 shows the realization flow of the motion target detection method that the embodiment of the invention provides, and its concrete step is as described below:
In step S101,, set up gray scale background model and gradient background model according to gray scale of obtaining in advance and gradient information.
In embodiments of the present invention, obtain gray scale and gradient information in advance, set up background model according to this gray scale and gradient then, following to have be the description that embodiment carries out with gray scale and gradient respectively, do not repeat them here, but not in order to restriction the present invention.
In step S102, carry out motion target detection respectively according to described gray scale background model and gradient background model.
In embodiments of the present invention, after gray scale background model of setting up and gradient background model,, can utilize difference image and variance to compare for each pixel of later every two field picture, judge that current point is motor point or rest point, its concrete determination methods is:
In step S103, the result who carries out motion target detection based on described gray scale background model and gradient background model is respectively carried out combination, cut apart and obtain moving target.
In embodiments of the present invention,, can obtain final moving target information, get rid of because the influence of factors such as illumination gradual change by combining based on the result of gray scale moving object detection with based on the result of gradient moving object detection.
As another embodiment of the present invention, gray scale that the basis that Fig. 2 shows the embodiment of the invention to be provided is obtained in advance and gradient information are set up the realization flow of gray scale background model and gradient background model, and its concrete step is as described below:
In step S201, establish the initial average and the variance of half-tone information and gradient information.
In embodiments of the present invention, the acquisition of gradient information can be carried out convolution algorithm as template and original image by utilization sobel operator and obtain, and for the computing method of the gradient information of certain pixel is:
For each pixel, with the half-tone information value of first frame and gradient information value respectively as initial gray average and gradient mean value, each pixel for second frame, obtain corresponding difference image by gray scale and the gradient difference of calculating present frame and background model, and then obtain initial difference and corresponding initialization standard variance.
In step S202, set up initial gray scale background model and initial gradient background model according to the initial average and the variance of described half-tone information and gradient information.
In embodiments of the present invention, suppose the variation Gaussian distributed of the pixel of each except that the moving region in the background image, similar with single Gauss method, can be by the average of respective pixel point and the Gauss model that variance is set up background in the background in observation a period of time.
In step S203, according to the pixel value of each frame, described initial gray scale background model and initial gradient background model are trained, obtain gray scale background model and gradient background model.
In embodiments of the present invention, need be by the training of a period of time, obtain stable gray scale and gradient background model, that is: for each pixel of later every frame, the background value of each frame carries out incremented according to the magnitude relationship of current pixel value and last frame background value or unit successively decreases, and the method for building up of gray scale and gradient background model is:
μ
t(i, j)=μ
T-1(i, j)+sgn (I
t(i, j)-μ
T-1(i, j)), wherein, average μ
t(i j) represents the background value of present frame, μ
T-1(i j) represents the background value of previous frame.And sgn () expression sign function, its computing method are:
If average μ
T-1(i is j) than current pixel value I
t(i, j) little, then with average μ
t(i j) increases by 1, if average μ
T-1(i is j) than current pixel value I
t(i, j) big, then with average μ
t(i j) reduces 1.As can be seen, μ
t(i j) is I
t(i, j) approximate only be one embodiment of the present of invention herein, at this not in order to limit the present invention.
In embodiments of the present invention, for each pixel of later every frame, the computing method of gray scale and gradient standard variance are as follows: (present embodiment hypothesis Δ
t(x) ≠ 0)
σ
t(i, j)=σ
T-1(i, j)+sgn (N * Δ
t(i, j)-σ
T-1(i, j)), wherein, σ is decided in choosing on concrete scene of N
t(i j) represents present frame (i, j) the standard variance value of pixel, σ
T-1(i j) represents previous frame (i, j) the standard variance value of pixel.If difference is bigger, i.e. σ
T-1(i, j)＜N * Δ
t(bigger variation j), then with the variance incremented, has taken place with the reflection background, thereby has made background model have adaptivity for the variation of background in i.
In embodiments of the present invention, gray scale background model and gradient background model to set up mode similar, no longer separately give unnecessary details at this, but not in order to restriction the present invention.
As an alternative embodiment of the invention, the average and the variance of gray scale background model and gradient background model are upgraded respectively, to reach the purpose of upgrading gray scale background model and gradient background model, what it was concrete is embodied as:
For the renewal of average, be judged as the number of impact point by adding up every each pixel of M frame, if be judged as that the number of impact point has surpassed P time and this point when being impact point just, then do not upgrade, otherwise upgrade, its mode is: μ
t(i, j)=μ
t(i, j)+asgn (I
r(i, j)-μ
t(i, j)), wherein, a=4;
For the renewal of variance, be judged as the number of impact point by adding up every each pixel of M frame, the number of impact point has surpassed P time and this pixel point value of present frame I if be judged as
t(i, j)=μ
t(i in the time of j), does not then upgrade, otherwise upgrades, and mode is: σ
t(i, j)=σ
t(i, j) a*sgn (N* Δ
t(i, j)-σ
t(i, j)), wherein, a=4.
As another embodiment of the present invention, the result who the gray scale background model is carried out moving object detection carries out morphologic filtering processing and connected domain analysis, and what it was concrete is embodied as:
For morphologic filtering, in order to keep target information as much as possible, and the influence of removing noise spot needs that behind detected moving target pixel binary image is carried out morphologic filtering to handle, the influence of removing noise spot as much as possible.Present embodiment at the deficiency that wherein exists, improves on the basis of basic morphologic filtering algorithm, before doing closed operation, has added the airspace filter processing, and it specifically comprises the steps:
(1) from left to right, from top to bottom, scan image finds first binaryzation in the image to be labeled as 255 point (with the 3*3 structural element is example, for preventing to cross the border, does not handle and goes up most, the most following, the most left, the pictorial element on the rightest four limits);
(2) for being labeled as 255 points (impact point), judge 25 points in its field,, then current point is changed to 255 if N point arranged for impact point 255 in 25 points, otherwise, the current pixel location value is changed to 0 repeating step (1) and (2), until the traversal entire image.
For connected component analysis, be to find all pixels that belong to same connected region by using labeling algorithm, and give identical mark, and give different marks for the pixel that belongs to different connected regions, then the connected region of area greater than predetermined threshold value extracted, be the moving target that will obtain.
Above-mentioned only is one embodiment of the present of invention, and this morphologic filtering and connected domain analysis are at the half-tone information of moving target, this not in order to the restriction the present invention.
Fig. 3 shows the structured flowchart of the motion target detection system that first embodiment of the invention provides, for convenience of explanation, only provided the part relevant among the figure with the embodiment of the invention, wherein, the motion target detection system can be built in software unit, hardware cell or the software and hardware unit of video analytic system.
Background model is set up module 11 according to the gray scale and the gradient information that obtain in advance, sets up gray scale background model and gradient background model; Detection module 12 carries out motion target detection according to described gray scale background model and gradient background model that described background model is set up module 11 foundation; Cut apart module 13 described detection module 12 is carried out combination based on the result that described gray scale background model and gradient background model are carried out moving object detection respectively, cut apart and obtain moving target.
In embodiments of the present invention, as shown in Figure 4, establish initial average and variance that module 111 is established half-tone information and gradient information; First model building module 112 is set up initial gray scale background model and initial gradient background model according to the half-tone information of described establishment module 111 establishments and the initial average and the variance of gradient information; Training module 113 is set up the initial gray scale background model and the initial gradient background model of module foundation and is trained according to the pixel value of each frame to described first model 112, obtain gray scale background model and gradient background model.
In embodiments of the present invention, the detailed realization flow of above-mentioned functions module as described in above-mentioned method embodiment, does not repeat them here, but not in order to restriction the present invention.
As an alternative embodiment of the invention, Fig. 5 shows the structured flowchart of the motion target detection system that second embodiment of the invention provides, for convenience of explanation, only provided the part relevant among the figure with the embodiment of the invention, wherein, the motion target detection system can be built in software unit, hardware cell or the software and hardware unit of video analytic system.
Background model is set up module 11 according to the gray scale and the gradient information that obtain in advance, sets up gray scale background model and gradient background model; Detection module 12 carries out motion target detection according to described gray scale background model and gradient background model that described background model is set up module 11 foundation; Cut apart module 13 described detection module 12 is carried out combination based on the result that described gray scale background model and gradient background model are carried out moving object detection respectively, cut apart and obtain moving target, wherein detailed realization flow, as described in above-mentioned method embodiment, do not repeat them here, but not in order to restriction the present invention.
In embodiments of the present invention, standard variance is set up the standard variance that module 14 is set up gray scale and gradient; The average and the variance of 15 pairs of described gray scale background models of update module and gradient background model are upgraded respectively; The result that 16 pairs of gray scale background models of Filtering Processing module are carried out moving object detection carries out the morphologic filtering processing; The result that 17 pairs of gray scale background models of connected domain analysis module are carried out moving object detection carries out the connected domain analysis.
In embodiments of the present invention, establish initial average and the variance that module 111 is established half-tone information and gradient information; First model building module 112 is set up initial gray scale background model and initial gradient background model according to the half-tone information of described establishment module 111 establishments and the initial average and the variance of gradient information; Training module 113 is set up the initial gray scale background model and the initial gradient background model of module foundation and is trained according to the pixel value of each frame to described first model 112, obtain gray scale background model and gradient background model.
In embodiments of the present invention, the detailed realization flow of above-mentioned functions module as described in above-mentioned method embodiment, does not repeat them here, but not in order to restriction the present invention.
In embodiments of the present invention, according to gray scale of obtaining in advance and gradient information, set up gray scale background model and gradient background model; Carry out motion target detection according to described gray scale background model and gradient background model; The result who carries out moving object detection based on described gray scale background model and gradient background model is respectively carried out combination, cut apart and obtain moving target, improved the precision of moving object detection, the applicability height, simultaneously, increased and increased progressively the computing of successively decreasing, and reduced floating-point operations such as asking average, extraction of square root, made the arithmetic speed of algorithm improve greatly.
The above only is preferred embodiment of the present invention, not in order to restriction the present invention, all any modifications of being done within the spirit and principles in the present invention, is equal to and replaces and improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. a motion target detection method is characterized in that, described method comprises the steps:
According to gray scale of obtaining in advance and gradient information, set up gray scale background model and gradient background model;
Carry out motion target detection respectively according to gray scale background model and gradient background model;
To combine based on the result that gray scale background model and gradient background model are carried out moving object detection, cut apart and obtain moving target.
2. the method for claim 1 is characterized in that, gray scale that described basis is obtained in advance and gradient information, and the concrete steps of setting up gray scale background model and gradient background model comprise:
Establish the initial average and the variance of half-tone information and gradient information;
Initial average and variance according to described half-tone information and gradient information are set up initial gray scale background model and initial gradient background model;
According to the pixel value of each frame, described initial gray scale background model and initial gradient background model are trained, obtain gray scale background model and gradient background model.
3. method as claimed in claim 2 is characterized in that, described pixel value according to each frame is trained described initial gray scale background model and initial gradient background model, and the step that obtains gray scale background model and gradient background model also comprises afterwards:
Set up the standard variance of gray scale and gradient.
4. method as claimed in claim 2 is characterized in that described method also comprises the steps:
Average and variance to described gray scale background model and gradient background model are upgraded respectively.
5. the method for claim 1 is characterized in that, described will combining based on the result that gray scale background model and gradient background model are carried out moving object detection cut apart the step that obtains moving target and also comprised the steps:
The result who carries out moving object detection based on the gray scale background model is carried out morphologic filtering to be handled;
The result who carries out moving object detection based on the gray scale background model is carried out the connected domain analysis.
6. a motion target detection system is characterized in that, described system comprises:
Background model is set up module, is used for setting up gray scale background model and gradient background model according to the gray scale and the gradient information that obtain in advance;
Detection module is used for setting up gray scale background model and the gradient background model that module sets up according to described background model and carries out motion target detection; And
Cut apart module, be used for described detection module is carried out respectively combining based on the result of described gray scale background model and gradient background model moving object detection, cut apart and obtain moving target.
7. system as claimed in claim 6 is characterized in that, described background model is set up module and specifically comprised:
Establish module, the initial average and the variance that are used to establish half-tone information and gradient information;
First model building module is used for setting up initial gray scale background model and initial gradient background model according to the half-tone information of described establishment module establishment and the initial average and the variance of gradient information; And
Training module is used for the pixel value according to each frame, and initial gray scale background model and initial gradient background model that described first model building module is set up are trained, and obtains gray scale background model and gradient background model.
8. system as claimed in claim 7 is characterized in that, described system also comprises:
Standard variance is set up module, is used to set up the standard variance of gray scale and gradient;
Update module is used for the average and the variance of described gray scale background model and gradient background model are upgraded respectively.
9. system as claimed in claim 6 is characterized in that, described system also comprises:
The Filtering Processing module is used for that the result that the gray scale background model is carried out moving object detection is carried out morphologic filtering and handles; And
The connected domain analysis module, the result who is used for the gray scale background model is carried out moving object detection carries out the connected domain analysis.
10. a video analytic system is characterized in that, described video analytic system comprises the described motion target detection of claim 6 to 9 system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009101100345A CN102054277B (en) | 2009-11-09 | 2009-11-09 | Method and system for detecting moving target, and video analysis system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009101100345A CN102054277B (en) | 2009-11-09 | 2009-11-09 | Method and system for detecting moving target, and video analysis system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102054277A true CN102054277A (en) | 2011-05-11 |
| CN102054277B CN102054277B (en) | 2013-08-21 |
Family
ID=43958563
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009101100345A Active CN102054277B (en) | 2009-11-09 | 2009-11-09 | Method and system for detecting moving target, and video analysis system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102054277B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102236902A (en) * | 2011-06-21 | 2011-11-09 | 杭州海康威视软件有限公司 | Method and device for detecting targets |
| CN102665068A (en) * | 2012-04-26 | 2012-09-12 | 中南林业科技大学 | Panoramic type moving object surveillance method based on random update strategies |
| WO2014047856A1 (en) * | 2012-09-27 | 2014-04-03 | 华为技术有限公司 | Method and device for determining video foreground main image area |
| CN105184820A (en) * | 2015-09-15 | 2015-12-23 | 杭州中威电子股份有限公司 | Background modeling and motion object detection method and apparatus with image gradient and gray scale integration |
| CN110610150A (en) * | 2019-09-05 | 2019-12-24 | 北京佳讯飞鸿电气股份有限公司 | Tracking method, device, computing equipment and medium of target moving object |
| CN113052047A (en) * | 2021-03-18 | 2021-06-29 | 北京百度网讯科技有限公司 | Traffic incident detection method, road side equipment, cloud control platform and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101489121B (en) * | 2009-01-22 | 2013-02-13 | 北京中星微电子有限公司 | Background model initializing and updating method based on video monitoring |
-
2009
- 2009-11-09 CN CN2009101100345A patent/CN102054277B/en active Active
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102236902A (en) * | 2011-06-21 | 2011-11-09 | 杭州海康威视软件有限公司 | Method and device for detecting targets |
| CN102236902B (en) * | 2011-06-21 | 2013-01-09 | 杭州海康威视数字技术股份有限公司 | Method and device for detecting targets |
| CN102665068A (en) * | 2012-04-26 | 2012-09-12 | 中南林业科技大学 | Panoramic type moving object surveillance method based on random update strategies |
| WO2014047856A1 (en) * | 2012-09-27 | 2014-04-03 | 华为技术有限公司 | Method and device for determining video foreground main image area |
| CN105184820A (en) * | 2015-09-15 | 2015-12-23 | 杭州中威电子股份有限公司 | Background modeling and motion object detection method and apparatus with image gradient and gray scale integration |
| CN105184820B (en) * | 2015-09-15 | 2018-03-13 | 杭州中威电子股份有限公司 | A kind of background modeling and moving target detecting method and device for merging figure gradient and gray scale |
| CN110610150A (en) * | 2019-09-05 | 2019-12-24 | 北京佳讯飞鸿电气股份有限公司 | Tracking method, device, computing equipment and medium of target moving object |
| CN113052047A (en) * | 2021-03-18 | 2021-06-29 | 北京百度网讯科技有限公司 | Traffic incident detection method, road side equipment, cloud control platform and system |
| CN113052047B (en) * | 2021-03-18 | 2023-12-29 | 阿波罗智联(北京)科技有限公司 | Traffic event detection method, road side equipment, cloud control platform and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102054277B (en) | 2013-08-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102054277B (en) | Method and system for detecting moving target, and video analysis system | |
| Nienaber et al. | Detecting potholes using simple image processing techniques and real-world footage | |
| CN112669349B (en) | Passenger flow statistics method, electronic equipment and storage medium | |
| CN101311964B (en) | Method and device for real time cutting motion area for checking motion in monitor system | |
| CN102509291B (en) | Pavement disease detecting and recognizing method based on wireless online video sensor | |
| CN102609682B (en) | Feedback pedestrian detection method for region of interest | |
| CN103208123B (en) | Image partition method and system | |
| CN102436656A (en) | Animal-diversity monitoring method based on computer vision | |
| CN109919002B (en) | Yellow stop line identification method and device, computer equipment and storage medium | |
| CN107346547B (en) | Monocular platform-based real-time foreground extraction method and device | |
| CN104463903A (en) | Pedestrian image real-time detection method based on target behavior analysis | |
| CN104252709B (en) | Multiple-target foreground detection method for look-down group-housed pigs in look-down state under complicated background | |
| CN101470809A (en) | Moving object detection method based on expansion mixed gauss model | |
| CN105069441A (en) | Moving vehicle detection method based on background updating and particle swarm optimization algorithm | |
| CN103793715B (en) | The personnel in the pit's method for tracking target excavated based on scene information | |
| Li et al. | A lane marking detection and tracking algorithm based on sub-regions | |
| CN111967396A (en) | Processing method, device and equipment for obstacle detection and storage medium | |
| CN101950424A (en) | Feature associated cell tracking method based on centroid tracking frame | |
| CN103914829B (en) | Method for detecting edge of noisy image | |
| CN103489254A (en) | Lottery recognition method and lottery recognition system | |
| CN101908214A (en) | Moving object detection method with background reconstruction based on neighborhood correlation | |
| CN104599291A (en) | Structural similarity and significance analysis based infrared motion target detection method | |
| CN110033425B (en) | Interference area detection device and method and electronic equipment | |
| CN103093481B (en) | A kind of based on moving target detecting method under the static background of watershed segmentation | |
| CN105205833A (en) | Moving object detection method and device based on space-time background model |
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 | ||
| ASS | Succession or assignment of patent right |
Owner name: SHENZHEN LAUNCH DIGITAL TECHNOLOGY CO., LTD. Effective date: 20140314 Owner name: GUANGZHOU POWER SUPPLY BUREAU Free format text: FORMER OWNER: SHENZHEN LAUNCH DIGITAL TECHNOLOGY CO., LTD. Effective date: 20140314 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| C53 | Correction of patent for invention or patent application | ||
| CB03 | Change of inventor or designer information |
Inventor after: Wang Quanming Inventor after: Zhou Jianxiong Inventor after: Shi Gaimei Inventor after: He Zhitao Inventor after: Liu Yigang Inventor before: Wang Quanming Inventor before: Zhou Jianxiong Inventor before: Shi Gaimei |
|
| COR | Change of bibliographic data |
Free format text: CORRECT: INVENTOR; FROM: WANG QUANMING ZHOU JIANXIONG SHI GAIMEI TO: WANG QUANMING ZHOU JIANXIONG SHI GAIMEI HE ZHITAO LIU YIGANG Free format text: CORRECT: ADDRESS; FROM: 518055 SHENZHEN, GUANGDONG PROVINCE TO: 510620 GUANGZHOU, GUANGDONG PROVINCE |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20140314 Address after: 510620 Tianhe District, Guangzhou, Tianhe South Road, No. two, No. 2, No. Patentee after: Guangzhou Power Supply Bureau Patentee after: Shenzhen Launch Digital Technology Co., Ltd. Address before: 518055 Shenzhen, Nanshan District hi tech park, North West New West Road, building No. five galaxy on the third floor Patentee before: Shenzhen Launch Digital Technology Co., Ltd. |
|
| CP01 | Change in the name or title of a patent holder |
Address after: 510620, No. two, No. 2, Tianhe South Road, Guangzhou, Guangdong, Tianhe District Patentee after: GUANGZHOU POWER SUPPLY Co.,Ltd. Patentee after: LAUNCH DIGITAL TECHNOLOGY Co.,Ltd. Address before: 510620, No. two, No. 2, Tianhe South Road, Guangzhou, Guangdong, Tianhe District Patentee before: GUANGZHOU POWER SUPPLY Co.,Ltd. Patentee before: SHENZHEN LAUNCH DIGITAL TECHNOLOGY Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |