CN102289948A - Multi-characteristic fusion multi-vehicle video tracking method under highway scene - Google Patents
Multi-characteristic fusion multi-vehicle video tracking method under highway scene Download PDFInfo
- Publication number
- CN102289948A CN102289948A CN2011102577077A CN201110257707A CN102289948A CN 102289948 A CN102289948 A CN 102289948A CN 2011102577077 A CN2011102577077 A CN 2011102577077A CN 201110257707 A CN201110257707 A CN 201110257707A CN 102289948 A CN102289948 A CN 102289948A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- frame
- video
- tracking
- sequence
- 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
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to a multi-characteristic fusion multi-vehicle video tracking method under a highway scene. The method disclosed by the invention comprises the following steps of: obtaining a video image of a vehicle running in a highway by using a monitoring camera with a fixed view field; modelling on each frame of image in an input video image collected by the monitoring camera, and detecting and locating the vehicle in the image; extracting multiple characteristic information of a moving vehicle, for tracking a video vehicle; and finally, matching detection results of different video frames, judging whether the moving vehicle and the video vehicle belong to the same vehicle; and processing the influence of the tracking effect resulted from mutual shading of vehicles in the image. According to the method disclosed by the invention, a multi-information fusion technology is used; and the system is beneficial to increasing the tracking accuracy and the anti-interference performance through multi-characteristic combined tracking.
Description
Technical field
The invention belongs to image processing field, be specifically related to intelligent transportation field, specially refer to the method that multiple vehicle video is followed the tracks of under the freeway surveillance and control scene.
Background technology
In recent years, because rapid economy development, road traffic develops rapidly, and the recoverable amount of motor vehicles is soaring rapidly, and a large amount of highway communication problems display, and as the life that takes place frequently of traffic congestion, traffic hazard equifrequency, traffic administration has been proposed new challenge.In order to solve the variety of issue that the traffic above-ground fast development is caused, each developed country competitively drops into a large amount of funds and personnel, and beginning is carried out highway communication on a large scale and transported intelligentized research.Critical positions that the research of intelligent transportation system (ITS, Intelligent Traffic System) is referred.Many countries just develop intelligent transportation system and have made longterm planning.After coming into operation, the intelligent transportation system technology that part has been succeeded in developing obtained good effect and income.Moving vehicle detects the core technology of following the tracks of the intelligent transportation system that is based on computer vision, is one of basic problem of computer vision.
Moving vehicle is followed the tracks of on the basis be based upon vehicle detection accurately.At present the detection method based on the moving vehicle of video mainly contains: methods such as background subtraction point-score, time differencing method, gray feature method, optical flow method, Gaussian Background modeling.
Moving vehicle is followed the tracks of and to be equivalent to the corresponding matching problem of creating features relevant such as position-based, speed, shape, texture, color in continuous images interframe, and the tracking of moving target roughly can be divided into method based on feature, based on the method in zone, based on the method for profile, based on four kinds of the methods of model.
Tracking based on feature
Based on the tracking of feature, be from image, to extract the characteristic feature of whole vehicle or local vehicle and between image sequence, mate the tracking of these characteristic features.
Tracking based on the zone
Based on the tracking of zonule is more typically based on the vision track in zone.Its cardinal rule is exactly to utilize the feature of image that every two field picture is divided into zones of different, carries out the zone coupling by these zones in consecutive frame, realizes target following.
Tracking based on profile
Its core concept of tracking based on active contour is to utilize the curved profile of sealing to come the expressive movement target, thereby the distortion that this curved profile reaches direction and direction by various constraint functions gradually with image in real goal adapt retrieval or follow the tracks of target in the complex background.
Tracking based on model
, when following the tracks of, mate and locate and recognition objective by measuring or other computer vision technique is set up the Three-dimension Target model based on the method for typical three-dimensional model in the model by parameters such as model and projections.Because utilized the three-D profile or the surface information of object, these class methods have very strong robustness in itself, block with disturbed condition under can obtain the effect that other method hardly matches.But some shortcomings that oneself is also arranged.
Summary of the invention
The objective of the invention is to be to overcome the deficiencies in the prior art, provide a kind of and carry out vehicle detection, adopt many Feature Fusion then, the tracking of many vehicles that piecemeal modeling, color modeling and position modeling etc. are merged mutually based on improved mixed Gaussian modeling.This method has very strong robustness for the data association of many vehicles on the highway and the difficult point problem of tracking such as blocking.
The objective of the invention is to realize by following steps:
Step 1. is obtained video image with the fixing rig camera of visual field to the vehicle that travels in the highway.
Step 2. is utilized improved mixed Gaussian modeling for each two field picture in the inputted video image of rig camera collection, carries out vehicle detection and location in image, specifically:
2-1. based on the road area in the video, utilize improved mixed Gaussian modeling, a parameter is introduced in described improved mixed Gaussian modeling
,
Counting of the valid pixel of record present frame and Model Matching makes the context update rate
, wherein
The turnover rate of context parameter in the expression elementary mixing Gauss modeling.
2-2. after the improved mixed Gaussian modeling, the video image of input becomes the image of binaryzation, wherein white pixel is represented as the position of foreground object; By 4 neighborhood merger strategies of connected region, the white connected region that will meet the vehicle size feature is remembered out with the rectangular search collimation mark in original video image, can obtain the vehicle location in the video like this.
Step 3. is extracted a plurality of characteristic informations of moving vehicle, is used for video frequency vehicle and follows the tracks of, specifically: rectangular search frame in vehicle place is carried out the extraction of gray scale, hsv color and positional information.
The testing result of step 4. coupling different video frame judges whether to belong to same vehicle, and the situation that vehicle in the image blocks is mutually handled, specifically:
4-1. vehicle similarity definition: i vehicle detection result of t frame is expressed as
, wherein
Expression vehicle place rectangular search frame center
, the width of rectangular search frame and length are
The hsv color feature of expression vehicle place rectangular search frame.
The gray feature of expression vehicle place rectangular search frame.And j vehicle detection result of t-1 frame is expressed as
, wherein
Expression vehicle place rectangular search frame center
, rectangular search width of frame and length are
,
The hsv color feature of expression vehicle place rectangular search frame,
The gray feature of expression vehicle place rectangular search frame.
Adopt the kernel function modeling of Gaussian distribution, obtain the hsv color similar function respectively
, the position similar function
, the gray scale similar function
, vehicle similarity function then
Wherein
Be respectively
,
With
In weighting function.
4-2. initialization: detect all vehicles in the present frame and constituted vehicle sequence to be matched, if first frame that present frame is improved mixed Gaussian modeling after finishing then is initialized as the vehicle sequence to be matched in first frame existing vehicle sequence of following the tracks of.
4-3. follow the tracks of:
Pre-set threshold value
,
,
4-3-1 mates each vehicle and existing all vehicles of following the tracks of the vehicle sequence of the vehicle sequence to be matched of t frame, finds i vehicle to be matched and existing j vehicle following the tracks of similarity function value maximum in the vehicle sequence in the t frame; If the similarity function value that this is maximum
More than or equal to threshold values
, judge that then the vehicle j of t-1 frame appears at the position of the vehicle i of t frame, upgrade information of vehicles; If this maximum similarity functional value
Less than threshold values
, the vehicle i in the vehicle sequence to be matched of t frame joins the vehicle buffer sequence as the new vehicle that may appear in the video so.
4-3-2 adds up vehicles all in the vehicle buffer sequence, if current vehicle occurs in buffer sequence continuously
Frame.Think that so this vehicle is that new vehicle appears in the video, and this car is joined existing tracking vehicle sequence, and be back to step 4-3-1.If continuously
Frame does not occur, and thinks that so then this vehicle has rolled this video monitoring range away from, deletes in buffering vehicle sequence, and it is deleted from existing the tracking the vehicle sequence.
If the vehicle target in the t frame
Two or two above vehicles in the corresponding t-1 frame
Function of position all greater than a default thresholding
The time, vehicle target is described
Corresponding to two or more auto models in the former frame image
, judge and take place to block mutually between vehicle, adopt the piecemeal tracking to eliminate tracking and block influence.
Described piecemeal tracking, specifically: at vehicle
Adopt in the place rectangular search frame sectional pattern to carry out piecemeal respectively, and extract the hsv color feature of each piecemeal, will divide all distances of the every original vehicle rectangular search frame central point of central point distance on the mark of each good piece then
Wherein
For
Distance on the direction,
For
Distance on the direction,
Quantity for piecemeal.To the vehicle target of each piecemeal at current t frame
Carry out color-match in the rectangular search frame; Selection color-match value maximum
Individual piecemeal (wherein
,
) unite the vehicle target that is blocking at the center of the vehicle of determining former frame
The position of rectangular search frame, adopt the vehicle rectangular search frame of previous frame to obtain the rectangular search frame position of vehicle, separable so a plurality of vehicles that block mutually.
Step 4-1 is obtaining the hsv color similar function
Adopted the method for taking interlacing to extract to current foreground image in the process.
Compared with prior art, the invention has the advantages that:
The present invention is stronger to the anti-interference of illumination variation and picture noise by the mixed Gaussian modeling aspect background modeling, and the vehicle foreground information is extracted in success accurately help preferably.
The present invention has adopted the technology of many information fusion on the target association of many vehicle trackings, manifold unite to follow the tracks of help improving the accuracy rate and the anti-interference of tracking.
The invention solves and block in the vehicle tracking under the coupling situation, the problem that target is lost easily, vehicle tracking success ratio height and range of application are wider.
Description of drawings
Fig. 1 is the overview flow chart of this method.
Fig. 2 is the process flow diagram of mixed Gaussian modeling.
Fig. 3 is the illustraton of model of position modeling.
Fig. 4 is the sectional pattern figure that piecemeal is followed the tracks of.
Embodiment
The present invention will be further described below in conjunction with the drawings and specific embodiments.
Fig. 1 has provided the techniqueflow chart of vehicle tracking algorithm.
1, video acquisition
At the fixing rig camera of visual field, with Fixed Time Interval road scene is gathered to obtain continuous field scene image, be the accuracy that guarantees vehicle detection and tracking, sampling interval should be less than 0.1 second, promptly greater than 10 frame per seconds.
2, vehicle detection
This method adopts improved mixed Gaussian that background is carried out modeling.The mixed Gaussian modeling is made up of the weighted sum of limited Gaussian function.To each pixel, definition
Individual Gauss model.The flow process of mixed Gaussian modeling as shown in Figure 2.
Have robustness for the color that makes image pixel changes luminance brightness, we adopt the hsv color space that colourity is decomposed from saturation degree, brightness, thereby reduce the influence to illumination variation.
The initialization of mixed Gauss model obtains by average and the variance of calculating one section video sequence image pixel, that is:
（2）
The pixel that needs to utilize each frame input after the initialization is the parameter of new model more, at first check each new pixel value whether with Model Matching, promptly constantly
If the pixel characteristic of input is to the
Individual gaussian component has:
D is for putting the letter parameter, and this algorithm gets 3, if
, so
, think that then the gauss hybrid models of input pixel and background does not match, this pixel is a foreground pixel, corresponding gauss hybrid models parameter is not upgraded.If
Then think
With the gauss hybrid models coupling, then the parameter of corresponding gauss hybrid models is upgraded.The renewal equation of the weights of gauss hybrid models and average and variance is as follows:
Wherein
Be turnover rate,
,
Be the context parameter turnover rate.
In traditional mixed Gauss model, context update speed depends on turnover rate
In order to suppress the noise and the stability that keeps model, turnover rate in the video
Value less usually.Yet turnover rate
Value hour, then the average of background model and variance speed of convergence are slow and need the variation that the long period conforms.Therefore get an adaptive turnover rate
Seem very crucial.At the problem of above-mentioned existence, the present invention improves the renewal process of background modeling:
(1) at first this algorithm is introduced a parameter
,
Counting of the valid pixel of record present frame and certain Model Matching, order
And use
Calculate the context update rate
Be initialized as 1, increase along with the increase of the valid pixel of each renewal coupling.When
Hour,
Bigger, in the renewal process of model, can accelerate the convergence of model parameter.Along with
Increase,
Diminish gradually, the model trend is stable.
(2) in order to solve the problem that the slow object of motion incorporates background, we observe the frame number that certain pixel is continuously background
, when
The time, even this pixel model has been converted into background model because weights rise, think still that this pixel is a foreground pixel this moment, thinks that this pixel is the slow object pixel of motion.If
Then, assert that just this pixel is a background pixel because this pixel model continues to regard as background model.The frame number that promptly is continuously background is greater than at least
Situation under, just can be identified as background.
After improved mixed Gaussian modeling, input picture becomes the image of binaryzation, white pixel point expression foreground object zone.By 4 neighborhood merger strategies of connected region, image is carried out from top to bottom then, the zone of the white pixel that links to each other is discerned in scanning from left to right.The zone that those sizes do not meet the vehicle size bound is rejected in the continuous zone that obtains, eliminate the interference of shade, just obtained the vehicle region location, mark with the rectangular search frame.
3, feature extraction
With hsv color feature, gray feature, the position feature information extraction of vehicle place rectangular search frame and store, be used for the vehicle tracking of video.
(1) extract hsv color:
Color characteristic is set up reliable model, so that follow the tracks of according to color model, this algorithm selects for use the method for Density Estimator that the detected vehicle piecemeal of mixed Gaussian modeling colouring information is carried out modeling, and kernel function adopts gaussian kernel.
Given sample space
, each pixel
Color represent with a tri-vector,
,
,
Expression respectively
,
,
Bandwidth.
(2) gray feature extracts:
Extract gray feature and extract the hsv color feature class seemingly, the operation in HSV space is converted to the gray space operation.
(3) extracting position feature:
Position coordinates is chosen as the point coordinate on the prospect vehicle rectangular search frame axis, and we choose upper, middle and lower three point coordinate, sets up the position similar function by the difference of point coordinate between two frames of front and back, and motion vector is expressed as
, as shown in Figure 3.
4, vehicle tracking:
The key of the tracking of many vehicles is data associations, the known vehicle tracking sequence of promptly setting up in the former frame how with present frame in the vehicle sequence opening relationships of finding to be matched, confirm that certain vehicle in two sequences is same trace model.
This algorithm has utilized color, gray scale and the positional information feature of target area, and tracking problem can realize by the coupling to each interframe feature of video, determines coupling according to maximum similarity during tracking.
The definition of 4-1 similarity
4-1-1.HSV the color similarity function:
（7）
Kernel function adopts Gaussian distribution:
Each Color Channel to variance is
,
，
，
（9）
4-1-2. the similar function of gray scale
With the hsv color similar function
Similar, the operating space by the HSV space conversion to gray space.
4-1-3. during the extracting position feature, we earlier on the basis of previous frame vehicle position information with this vehicle of kalman filter forecasting in the position of this frame, then the movement position result of this prediction is compared with the position of the vehicle of this frame extraction.When two auto model apart from each others, from less relatively, so the position similar function will play a major role the tracking effect that can produce at adjacent two frame pitches for same car.Vehicle tracked in tracing process need carry out finding the solution of position similar function with the model of previous moment oneself itself, also will carry out finding the solution of position similar function with other auto models simultaneously.
Motion vector is
, three point coordinate of motion vector are separate, so the position similar function
Can be expressed as:
Wherein:
（11）
Wherein
Be big or small decision of move distance by the auto model of two frame interframe.
Similarity function
Wherein
Be the hsv color similar function,
The position similar function,
Be the gray scale similar function.
Be respectively
Weighting function.
4-2. initialization: detect all vehicles in the present frame and constituted vehicle sequence to be matched, if first frame that present frame is improved mixed Gaussian modeling after finishing then is initialized as the vehicle sequence to be matched in first frame existing vehicle sequence of following the tracks of;
4-3. follow the tracks of:
Pre-set threshold value
,
,
4-3-1. each vehicle and existing all vehicles of following the tracks of the vehicle sequence of the vehicle sequence to be matched of t frame are mated, find i vehicle to be matched and existing j vehicle following the tracks of similarity function value maximum in the vehicle sequence in the t frame; If the similarity function value that this is maximum
More than or equal to threshold values
, judge that then the vehicle j of t-1 frame appears at the position of the vehicle i of t frame, upgrade information of vehicles; If this maximum similarity functional value
Less than threshold values
, the vehicle i in the vehicle sequence to be matched of t frame joins the vehicle buffer sequence as the new vehicle that may appear in the video so;
4-3-2. vehicles all in the vehicle buffer sequence is added up, if current vehicle occurs in buffer sequence continuously
Frame.Think that so this vehicle is that new vehicle appears in the video, and this car is joined existing tracking vehicle sequence, and be back to step 4-3-1.If continuously
Frame does not occur, and thinks that so then this vehicle has rolled this video monitoring range away from, deletes in buffering vehicle sequence, and it is deleted from existing the tracking the vehicle sequence.
If the vehicle target in the t frame
Two or two above vehicles in the corresponding t-1 frame
Function of position all greater than a default thresholding
The time, vehicle target is described
Corresponding to two or more auto models in the former frame image
, judge and take place to block mutually between vehicle, adopt the piecemeal tracking to eliminate tracking and block influence.
The piecemeal tracking, specifically: at vehicle
Adopt in the place rectangular search frame sectional pattern (as shown in Figure 4) to carry out piecemeal respectively, and extract the hsv color feature of each piecemeal, will divide all distances of the every original vehicle rectangular search frame central point of central point distance on the mark of each good piece then
Wherein
For
Distance on the direction,
For
Distance on the direction,
Quantity for piecemeal.To the vehicle target of each piecemeal at current t frame
Carry out color-match in the rectangular search frame; Selection color-match value maximum
Individual piecemeal (wherein
,
) unite the vehicle target that is blocking at the center of the vehicle of determining former frame
The position of rectangular search frame, adopt the vehicle rectangular search frame of previous frame to obtain the rectangular search frame position of vehicle, separable so a plurality of vehicles that block mutually.
At this moment, we do the mark that blocks with all vehicles that block in existing auto model sequence, till they separate to certain frame.Judge that the condition of blocking separation is: the motion modeling functional value of several auto models shows and blocks separation greater than a default threshold values in the auto model that mark blocks in former frame and this frame.We use based on the similarity of color model and judge the known trace model of the model of separation corresponding to some storages.Mark is blocked in the cancellation of coupling back.
5, improve the operational efficiency of algorithm
Take the method for interlacing extraction to reduce data volume to current foreground image and memory image.In order to take into account precision and calculated amount, we adopt following principle: undistorted to guarantee image if the very little image of resolution just needn't adopt interlacing to extract, the method for having only the bigger image of resolution to adopt interlacing to extract obtains image.Image processing operations that image after interlacing extracted is correlated with again, the rectangle frame that has obtained again will indicating after the tracking results in this frame tracing positional multiply by two times counter the shifting onto on the original image of mode.This method has not only reached original tracking effect but also has improved the operational efficiency of algorithm greatly.
Claims (2)
1. a kind of multiple vehicle video tracking method of many Feature Fusion under the highway scene is characterized in that this method may further comprise the steps:
Step 1. is obtained video image with the fixing rig camera of visual field to the vehicle that travels in the highway;
Step 2. is utilized improved mixed Gaussian modeling for each two field picture in the inputted video image of rig camera collection, carries out vehicle detection and location in image, specifically:
2-1. based on the road area in the video, utilize improved mixed Gaussian modeling, a parameter is introduced in described improved mixed Gaussian modeling
,
Counting of the valid pixel of record present frame and Model Matching makes the context update rate
, wherein
The turnover rate of context parameter in the expression elementary mixing Gauss modeling;
2-2. after the improved mixed Gaussian modeling, the video image of input becomes the image of binaryzation, wherein white pixel is represented as the position of foreground object; By 4 neighborhood merger strategies of connected region, the white connected region that will meet the vehicle size feature is remembered out with the rectangular search collimation mark in original video image, can obtain the vehicle location in the video like this;
Step 3. is extracted a plurality of characteristic informations of moving vehicle, is used for video frequency vehicle and follows the tracks of, specifically: rectangular search frame in vehicle place is carried out the extraction of gray scale, hsv color and positional information;
The testing result of step 4. coupling different video frame judges whether to belong to same vehicle, and the situation that vehicle in the image blocks is mutually handled, specifically:
4-1. vehicle similarity definition: i vehicle detection result of t frame is expressed as
, wherein
Expression vehicle place rectangular search frame center
, the width of rectangular search frame and length are
The hsv color feature of expression vehicle place rectangular search frame;
The gray feature of expression vehicle place rectangular search frame; And j vehicle detection result of t-1 frame is expressed as
, wherein
Expression vehicle place rectangular search frame center
, rectangular search width of frame and length are
,
The hsv color feature of expression vehicle place rectangular search frame,
The gray feature of expression vehicle place rectangular search frame;
Adopt the kernel function modeling of Gaussian distribution, obtain the hsv color similar function respectively
, the position similar function
, the gray scale similar function
, vehicle similarity function then
, wherein
Be respectively
,
With
In weighting function;
4-2. initialization: detect all vehicles in the present frame and constituted vehicle sequence to be matched, if first frame that present frame is improved mixed Gaussian modeling after finishing then is initialized as the vehicle sequence to be matched in first frame existing vehicle sequence of following the tracks of;
4-3. pre-set threshold value
,
,
And follow the tracks of:
4-3-1. each vehicle and existing all vehicles of following the tracks of the vehicle sequence of the vehicle sequence to be matched of t frame are mated, find i vehicle to be matched and existing j vehicle following the tracks of similarity function value maximum in the vehicle sequence in the t frame; If the similarity function value that this is maximum
More than or equal to threshold values
, judge that then the vehicle j of t-1 frame appears at the position of the vehicle i of t frame, upgrade information of vehicles; If this maximum similarity functional value
Less than threshold values
, the vehicle i in the vehicle sequence to be matched of t frame joins the vehicle buffer sequence as the new vehicle that may appear in the video so;
4-3-2. vehicles all in the vehicle buffer sequence is added up, if current vehicle occurs in buffer sequence continuously
Frame; Think that so this vehicle is that new vehicle appears in the video, and this car is joined existing tracking vehicle sequence, and be back to step 4-3-1; If continuously
Frame does not occur, and thinks that so then this vehicle has rolled this video monitoring range away from, deletes in buffering vehicle sequence, and it is deleted from existing the tracking the vehicle sequence;
If two or two above vehicles in the corresponding t-1 frame of the vehicle target O in the t frame
Function of position all greater than a default thresholding
The time, illustrate that vehicle target O is corresponding to two or more auto models in the former frame image
, judge and take place to block mutually between vehicle, adopt the piecemeal tracking to eliminate tracking and block influence, wherein
Described piecemeal tracking, specifically: at vehicle
Adopt in the place rectangular search frame sectional pattern to carry out piecemeal respectively, and extract the hsv color feature of each piecemeal, will divide all distances of the every original vehicle rectangular search frame central point of central point distance on the mark of each good piece then
Wherein
For
Distance on the direction,
For
Distance on the direction,
Quantity for piecemeal; Each piecemeal is carried out color-match in the vehicle target O of current t frame rectangular search frame; Selection color-match value maximum
Individual piecemeal, the center of uniting the vehicle of determining former frame adopt the rectangular search frame position of the vehicle rectangular search frame acquisition vehicle of previous frame in the position of the rectangular search frame of the vehicle target O that blocks, separable so a plurality of vehicles that block mutually, wherein
,
2. a kind of multiple vehicle video tracking method of many Feature Fusion under the highway scene according to claim 1 is characterized in that: step 4-1 is obtaining the hsv color similar function
Adopted the method for taking interlacing to extract to current foreground image in the process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110257707 CN102289948B (en) | 2011-09-02 | 2011-09-02 | Multi-characteristic fusion multi-vehicle video tracking method under highway scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110257707 CN102289948B (en) | 2011-09-02 | 2011-09-02 | Multi-characteristic fusion multi-vehicle video tracking method under highway scene |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102289948A true CN102289948A (en) | 2011-12-21 |
| CN102289948B CN102289948B (en) | 2013-06-05 |
Family
ID=45336325
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110257707 Expired - Fee Related CN102289948B (en) | 2011-09-02 | 2011-09-02 | Multi-characteristic fusion multi-vehicle video tracking method under highway scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102289948B (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663357A (en) * | 2012-03-28 | 2012-09-12 | 北京工业大学 | Color characteristic-based detection algorithm for stall at parking lot |
| CN102862574A (en) * | 2012-09-21 | 2013-01-09 | 上海永畅信息科技有限公司 | Method for realizing active safety of vehicle on the basis of smart phone |
| CN102915115A (en) * | 2012-09-25 | 2013-02-06 | 上海华勤通讯技术有限公司 | Method for adjusting display frame by eye |
| CN103136935A (en) * | 2013-01-11 | 2013-06-05 | 东南大学 | Method for tracking sheltered vehicles |
| CN103473566A (en) * | 2013-08-27 | 2013-12-25 | 东莞中国科学院云计算产业技术创新与育成中心 | Multi-scale-model-based vehicle detection method |
| CN103679214A (en) * | 2013-12-20 | 2014-03-26 | 华南理工大学 | Vehicle detection method based on online area estimation and multi-feature decision fusion |
| CN104318761A (en) * | 2014-08-29 | 2015-01-28 | 华南理工大学 | Highway-scene-based detection and vehicle detection tracking optimization method |
| CN106446824A (en) * | 2016-09-21 | 2017-02-22 | 防城港市港口区思达电子科技有限公司 | Vehicle detection and tracking method |
| CN106780539A (en) * | 2016-11-30 | 2017-05-31 | 航天科工智能机器人有限责任公司 | Robot vision tracking |
| CN108460968A (en) * | 2017-02-22 | 2018-08-28 | 中兴通讯股份有限公司 | A kind of method and device obtaining traffic information based on car networking |
| CN109359518A (en) * | 2018-09-03 | 2019-02-19 | 惠州学院 | A kind of moving object recognition methods, system and the warning device of infrared video |
| CN109426791A (en) * | 2017-09-01 | 2019-03-05 | 深圳市金溢科技股份有限公司 | A kind of polynary vehicle match method of multi-site, server and system |
| CN109472767A (en) * | 2018-09-07 | 2019-03-15 | 浙江大丰实业股份有限公司 | Stage lamp miss status analysis system |
| CN110264493A (en) * | 2019-06-17 | 2019-09-20 | 北京影谱科技股份有限公司 | A kind of multiple target object tracking method and device under motion state |
| CN110542908A (en) * | 2019-09-09 | 2019-12-06 | 阿尔法巴人工智能(深圳)有限公司 | laser radar dynamic object perception method applied to intelligent driving vehicle |
| CN111260929A (en) * | 2018-11-30 | 2020-06-09 | 西安宇视信息科技有限公司 | Vehicle tracking abnormity detection method and device |
| CN111597871A (en) * | 2020-03-27 | 2020-08-28 | 广州杰赛科技股份有限公司 | Vehicle tracking method and device, terminal equipment and computer storage medium |
| CN111882582A (en) * | 2020-07-24 | 2020-11-03 | 广州云从博衍智能科技有限公司 | Image tracking correlation method, system, device and medium |
| CN112382104A (en) * | 2020-11-13 | 2021-02-19 | 重庆盘古美天物联网科技有限公司 | Roadside parking management method based on vehicle track analysis |
| CN112532938A (en) * | 2020-11-26 | 2021-03-19 | 武汉宏数信息技术有限责任公司 | Video monitoring system based on big data technology |
| CN112597830A (en) * | 2020-12-11 | 2021-04-02 | 国汽(北京)智能网联汽车研究院有限公司 | Vehicle tracking method, device, equipment and computer storage medium |
| CN114241398A (en) * | 2022-02-23 | 2022-03-25 | 深圳壹账通科技服务有限公司 | Vehicle damage assessment method, device, equipment and storage medium based on artificial intelligence |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090309966A1 (en) * | 2008-06-16 | 2009-12-17 | Chao-Ho Chen | Method of detecting moving objects |
| CN101800890A (en) * | 2010-04-08 | 2010-08-11 | 北京航空航天大学 | Multiple vehicle video tracking method in expressway monitoring scene |
| CN101976504A (en) * | 2010-10-13 | 2011-02-16 | 北京航空航天大学 | Multi-vehicle video tracking method based on color space information |
-
2011
- 2011-09-02 CN CN 201110257707 patent/CN102289948B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090309966A1 (en) * | 2008-06-16 | 2009-12-17 | Chao-Ho Chen | Method of detecting moving objects |
| CN101800890A (en) * | 2010-04-08 | 2010-08-11 | 北京航空航天大学 | Multiple vehicle video tracking method in expressway monitoring scene |
| CN101976504A (en) * | 2010-10-13 | 2011-02-16 | 北京航空航天大学 | Multi-vehicle video tracking method based on color space information |
Cited By (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663357A (en) * | 2012-03-28 | 2012-09-12 | 北京工业大学 | Color characteristic-based detection algorithm for stall at parking lot |
| CN102862574A (en) * | 2012-09-21 | 2013-01-09 | 上海永畅信息科技有限公司 | Method for realizing active safety of vehicle on the basis of smart phone |
| CN102862574B (en) * | 2012-09-21 | 2015-08-19 | 上海永畅信息科技有限公司 | The method of vehicle active safety is realized based on smart mobile phone |
| CN102915115A (en) * | 2012-09-25 | 2013-02-06 | 上海华勤通讯技术有限公司 | Method for adjusting display frame by eye |
| CN103136935B (en) * | 2013-01-11 | 2015-04-15 | 东南大学 | Method for tracking sheltered vehicles |
| CN103136935A (en) * | 2013-01-11 | 2013-06-05 | 东南大学 | Method for tracking sheltered vehicles |
| CN103473566A (en) * | 2013-08-27 | 2013-12-25 | 东莞中国科学院云计算产业技术创新与育成中心 | Multi-scale-model-based vehicle detection method |
| CN103473566B (en) * | 2013-08-27 | 2016-09-14 | 东莞中国科学院云计算产业技术创新与育成中心 | A kind of vehicle checking method based on multiple dimensioned model |
| CN103679214A (en) * | 2013-12-20 | 2014-03-26 | 华南理工大学 | Vehicle detection method based on online area estimation and multi-feature decision fusion |
| CN103679214B (en) * | 2013-12-20 | 2017-10-20 | 华南理工大学 | Vehicle checking method based on online Class area estimation and multiple features Decision fusion |
| CN104318761A (en) * | 2014-08-29 | 2015-01-28 | 华南理工大学 | Highway-scene-based detection and vehicle detection tracking optimization method |
| CN106446824A (en) * | 2016-09-21 | 2017-02-22 | 防城港市港口区思达电子科技有限公司 | Vehicle detection and tracking method |
| CN106780539A (en) * | 2016-11-30 | 2017-05-31 | 航天科工智能机器人有限责任公司 | Robot vision tracking |
| CN106780539B (en) * | 2016-11-30 | 2019-08-20 | 航天科工智能机器人有限责任公司 | Robot vision tracking |
| CN108460968A (en) * | 2017-02-22 | 2018-08-28 | 中兴通讯股份有限公司 | A kind of method and device obtaining traffic information based on car networking |
| CN109426791A (en) * | 2017-09-01 | 2019-03-05 | 深圳市金溢科技股份有限公司 | A kind of polynary vehicle match method of multi-site, server and system |
| CN109426791B (en) * | 2017-09-01 | 2022-09-16 | 深圳市金溢科技股份有限公司 | Multi-site and multi-vehicle matching method, server and system |
| CN109359518A (en) * | 2018-09-03 | 2019-02-19 | 惠州学院 | A kind of moving object recognition methods, system and the warning device of infrared video |
| CN109472767A (en) * | 2018-09-07 | 2019-03-15 | 浙江大丰实业股份有限公司 | Stage lamp miss status analysis system |
| CN109472767B (en) * | 2018-09-07 | 2022-02-08 | 浙江大丰实业股份有限公司 | Stage lamp missing state analysis system |
| CN111260929A (en) * | 2018-11-30 | 2020-06-09 | 西安宇视信息科技有限公司 | Vehicle tracking abnormity detection method and device |
| CN110264493B (en) * | 2019-06-17 | 2021-06-18 | 北京影谱科技股份有限公司 | Method and device for tracking multiple target objects in motion state |
| CN110264493A (en) * | 2019-06-17 | 2019-09-20 | 北京影谱科技股份有限公司 | A kind of multiple target object tracking method and device under motion state |
| CN110542908B (en) * | 2019-09-09 | 2023-04-25 | 深圳市海梁科技有限公司 | Laser radar dynamic object sensing method applied to intelligent driving vehicle |
| CN110542908A (en) * | 2019-09-09 | 2019-12-06 | 阿尔法巴人工智能(深圳)有限公司 | laser radar dynamic object perception method applied to intelligent driving vehicle |
| CN111597871A (en) * | 2020-03-27 | 2020-08-28 | 广州杰赛科技股份有限公司 | Vehicle tracking method and device, terminal equipment and computer storage medium |
| CN111882582B (en) * | 2020-07-24 | 2021-10-08 | 广州云从博衍智能科技有限公司 | Image tracking correlation method, system, device and medium |
| CN111882582A (en) * | 2020-07-24 | 2020-11-03 | 广州云从博衍智能科技有限公司 | Image tracking correlation method, system, device and medium |
| CN112382104A (en) * | 2020-11-13 | 2021-02-19 | 重庆盘古美天物联网科技有限公司 | Roadside parking management method based on vehicle track analysis |
| CN112532938B (en) * | 2020-11-26 | 2021-08-31 | 武汉宏数信息技术有限责任公司 | Video monitoring system based on big data technology |
| CN112532938A (en) * | 2020-11-26 | 2021-03-19 | 武汉宏数信息技术有限责任公司 | Video monitoring system based on big data technology |
| CN112597830A (en) * | 2020-12-11 | 2021-04-02 | 国汽(北京)智能网联汽车研究院有限公司 | Vehicle tracking method, device, equipment and computer storage medium |
| CN114241398A (en) * | 2022-02-23 | 2022-03-25 | 深圳壹账通科技服务有限公司 | Vehicle damage assessment method, device, equipment and storage medium based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102289948B (en) | 2013-06-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102289948B (en) | Multi-characteristic fusion multi-vehicle video tracking method under highway scene | |
| CN108320510B (en) | Traffic information statistical method and system based on aerial video shot by unmanned aerial vehicle | |
| CN102222346B (en) | Vehicle detecting and tracking method | |
| CN104978567B (en) | Vehicle checking method based on scene classification | |
| CN105931295B (en) | A kind of geologic map Extracting Thematic Information method | |
| CN103325126B (en) | A kind of video target tracking method under dimensional variation and circumstance of occlusion | |
| CN100595792C (en) | Vehicle detecting and tracing method based on video technique | |
| CN102214291B (en) | Method for quickly and accurately detecting and tracking human face based on video sequence | |
| CN102542289B (en) | Pedestrian volume statistical method based on plurality of Gaussian counting models | |
| CN103150559B (en) | Head recognition and tracking method based on Kinect three-dimensional depth image | |
| CN110175576A (en) | A kind of driving vehicle visible detection method of combination laser point cloud data | |
| CN109145708B (en) | Pedestrian flow statistical method based on RGB and D information fusion | |
| CN103310444B (en) | A kind of method of the monitoring people counting based on overhead camera head | |
| CN102044151A (en) | Night vehicle video detection method based on illumination visibility identification | |
| CN105528794A (en) | Moving object detection method based on Gaussian mixture model and superpixel segmentation | |
| CN107563310B (en) | Illegal lane change detection method | |
| CN101976504B (en) | Multi-vehicle video tracking method based on color space information | |
| CN106128121B (en) | Vehicle queue length fast algorithm of detecting based on Local Features Analysis | |
| CN104091348A (en) | Multi-target tracking method integrating obvious characteristics and block division templates | |
| CN103413120A (en) | Tracking method based on integral and partial recognition of object | |
| CN104134222A (en) | Traffic flow monitoring image detecting and tracking system and method based on multi-feature fusion | |
| CN108198201A (en) | A kind of multi-object tracking method, terminal device and storage medium | |
| CN104318263A (en) | Real-time high-precision people stream counting method | |
| CN106127812B (en) | A kind of passenger flow statistical method of the non-gate area in passenger station based on video monitoring | |
| CN101916446A (en) | Gray level target tracking algorithm based on marginal information and mean shift |
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: 20130605 Termination date: 20140902 |
|
| EXPY | Termination of patent right or utility model |