CN103425764A - Vehicle matching method based on videos - Google Patents
Vehicle matching method based on videos Download PDFInfo
- Publication number
- CN103425764A CN103425764A CN2013103371386A CN201310337138A CN103425764A CN 103425764 A CN103425764 A CN 103425764A CN 2013103371386 A CN2013103371386 A CN 2013103371386A CN 201310337138 A CN201310337138 A CN 201310337138A CN 103425764 A CN103425764 A CN 103425764A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- search
- window
- video
- matching
- 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 Processing (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method for matching moving vehicles in videos. According to the method, firstly vehicle monitoring videos are obtained, the moving vehicles in each frame of each image is parted off through a background frame differential method; secondly, vehicle positions in each frame are recorded to establish a vehicle position database; thirdly, searching windows are placed in adjacent image frames with the vehicle needing matching as the center; then target vehicles are searched in the searching windows, adaptive adjustment is carried out on the sizes of the windows according to searching results till the vehicles are found, and the vehicle positions are recorded; finally, calculation of variation of displacement and variation of speed is carried out on found targets, and the optimal target is selected to achieve the matching. According to the vehicle matching method based on the videos, the defect that no effective method for carrying out vehicle matching after the detecting test of vehicles exists is overcome, proposed matching characteristics are simple in calculation, low in complexity and strong in instantaneity, verification modules can improve matching accuracy, and the vehicle matching method based on the videos has broad application prospect.
Description
Technical field
The present invention relates to the traffic video detection technical field, is a kind of vehicle matching process based on video, is mainly used in intelligent transportation system and extracts Traffic Information.
Background technology
Vehicle coupling based on video refers in a series of picture frame, and same the car that is in diverse location identified.At present a lot of systems (comprising intelligent transportation system) all need to extract the vehicle flowrate of road.A lot of algorithms all successfully detect moving vehicle, but the result detected can only mean the number of vehicles in single still image.If need statistical vehicle flowrate, need each car is mated.The perfect traffic intelligence that contributes to of vehicle matching process.Simultaneously by vehicle mate can registration of vehicle driving trace.When the unsafe acts such as illegal lane change occur when vehicle, the method can record and report to vehicle supervision department, improves traffic safety.
Traditional vehicle matching process mainly is based on the mean-shift method, and this method is not considered the practical engineering application background, when search, does not have the Optimizing Search scope, makes the calculated amount of algorithm larger, and real-time is poor; Secondly, the search box size of mean-shift is fixed, can not the self-adaptation adjustment, lack practicality.
Defect and deficiency due to this traditional matching process, be difficult to have the product of ripe vehicle Flow Detection to be released always, while is along with the intelligent transportation system development is perfect, universal and the traffic information collection hardware cost based on video of video capture device reduces gradually, and the vehicle matching technique based on video has a wide range of applications.
Summary of the invention
Defect or deficiency for existing vehicle matching technique, the invention provides a kind of vehicle matching process based on video.The method can be mated fast and accurately to the driving vehicle in range of video.
In order to realize above-mentioned task, the present invention takes following technical solution:
A kind of vehicle matching process based on video is characterized in that following steps:
Step 1: obtain the vehicle monitoring video, utilize the background frame difference method to be partitioned into the moving vehicle in each two field picture.
Obtain the vehicle monitoring video from video capture device, utilize the background frame difference method to be partitioned into the moving vehicle in each two field picture.
Step 2: record the position of the vehicle in each two field picture, set up the vehicle position data storehouse.
Record the vehicle location in each two field picture, obtain the minimum rectangle outline of vehicle, find the geometric center position of minimum rectangle outline, the unique point using this as vehicle also deposits database in.
Step 3: centered by the vehicle location of needs coupling, in the adjacent image frame, place search window.
The vehicle of select to need following the trail of, centered by the geometric center of the minimum rectangle outline of this vehicle, place a rectangular search window on the consecutive frame of following the trail of at needs, this search window defines the scope of vehicle search.The size of this rectangular window is 2*T*V* α, wherein T means the real time spacing between consecutive frame, V means the maximum limit speed per hour of this road, and α means the mapping relations of actual range and imaging plane in scene, i.e. actual range and ratio at the video middle distance.
Step 4: search for target vehicle in search window, and adjust window size until search vehicle, the registration of vehicle position according to the Search Results self-adaptation.
In the scope limited at search window, search vehicle from the vehicle characteristics point data base.If do not search vehicle, automatically increase window size, the step-length of increase is 0.2*T*V* α, until search vehicle; When search box size exceeds monitoring range, search stops; Perhaps, when searching plain window and be greater than T*405* α, search stops.Record all vehicle characteristics points that search, deposit in a set to be screened.
Step 5: the target searched is carried out to the calculating of displacement variable and velocity variable, select optimal objective, complete coupling.
Calculate displacement variable Δ S and the velocity variable Δ V of vehicle from set to be screened, according to both quadratic sum D=Δ S
2+ Δ V
2Weigh both motion state variable quantities.Choosing D is mated as optimal objective.
The present invention is based on the video frequency vehicle matching process, compared with prior art, can be mated all vehicles in range of video.The method computation complexity is little, and match time is short, and accuracy is higher, has broad application prospects.
The accompanying drawing explanation
The process flow diagram that Fig. 1 is the inventive method.
Fig. 2 is the process flow diagram that the window self-adaptation is adjusted.
The scheme of installation that Fig. 3 is video capture device.
Below in conjunction with drawings and Examples, the present invention is described in further detail.
Embodiment
Of the present invention based on the video frequency vehicle matching process, the principle of employing is: the grayscale image sequence that is the certain format size by the Video processing that collects.Vehicle is along with the diverse location of time change profile on image.The time interval between adjacent image is smaller, and the displacement that vehicle moves within this time interval is equally very little.So can be mated vehicle by the displacement of calculating vehicle.Because the time of travelling is very short, direction and the size variation of the speed of vehicle are also very little.So speed also can be used as a feature of coupling vehicle.Find in the process of being mated, do not need all vehicles in searching image to be mated, can directly get rid of those distant vehicles.Angle for optimized algorithm is considered, the present invention proposes a search window as hunting zone.In the actual match process, the size of vehicle and speed are all different.If it is excessive that window is set, can search a lot of jamming targets, increase the calculated amount in follow-up matching process.If it is too small that window is set, just be difficult to guarantee to search vehicle, can not complete search mission.Size how to set window becomes a difficult problem to be solved.The present invention proposes a kind of self-adaptation and adjust the method for window size, when having guaranteed to search vehicle, can get rid of most jamming target again.
The concrete following steps that adopt realize:
The first step: be partitioned into moving vehicle
Shown in Fig. 3, video capture device by hypothesis on traffic route obtains the real-time traffic video, after this video is carried out to the poor processing of background frames, is partitioned into the vehicle in all travelling, and, by the vehicle binaryzation split, record the position coordinates that each car covers in image.
Second step: determine vehicle location
The fundamental purpose of this step is to determine the geometric center position of vehicle.At first to determine the minimum rectangle outline of vehicle.Obtained the binary map after vehicles segmentation in the first step.So only need to determine the connected domain minimum rectangle outline in binary map.At first it is a set { (x that pixel coordinate vehicle covered arranges
1, y
1), (x
2, y
2), (x
3, y
3) ... (x
i, y
i), x wherein
i, y
iRepresent respectively the coordinate position at i pixel place.The minimum rectangle outline can be by the x in coordinate set, the maximal value of y and minimum value.Next pair set is sorted, and finds respectively minimum x coordinate x
min, maximum x coordinate x
max, minimum y coordinate y
min, maximum y coordinate y
max.Can determine four angle point (x of minimum rectangle outline
min, y
min), (x
min, y
max), (x
max, y
min), (x
max, y
max).Determining the geometric center (x of minimum outline according to four angle points
mid, y
mid), x wherein
mid=(x
min+ x
max)/2, y
mid=(y
min+ y
max)/2.
Step 3: place search window
The upper placement search window of the target of mating at needs, to search for this vehicle periphery minor increment target.Centered by the geometric center of minimum rectangle outline, place a rectangular search window on consecutive frame.The size of rectangular window is 2*T*V* α, and wherein T means the real time spacing between consecutive frame, and V means the maximum limit speed per hour of this road, and α means the mapping relations of actual range and imaging plane in scene, i.e. actual range and ratio at the video middle distance.In actual field, the image capture device that we choose is per second 30 frames, so T=1/108000h, and the Maximum speed limit of road regulation is 60km/h, and α is according to the calibration result of image capture device and difference.The present invention chooses α=12.3p/m, and wherein p/m means the number of pixels that the 1m in actual range occupies on imaging plane.After result is rounded, 14 pixels of the length of rectangular window and wide difference.
Step 4: window self-adaptation adjustment size is until search vehicle.
If the geometric center of i the vehicle detected is (x
i, y
i), for the vector representation of the geometric center in the vehicle of memory search window, be Vector. (x
m, y
m), Vector.x
mThe horizontal ordinate that means m car of storage, and Vector.y
mThe ordinate that means m car.Geometric center (x with tracked vehicle
0, y
0) centered by, in length and width, be a, in the rectangular window of b, searched for.If search in this window, moving vehicle is arranged, the geometric center of this vehicle is stored in to vectorial Vector. (x
m, y
m) in.Then to being stored in Vector. (x
m, y
m) in vehicle carry out the tracking computing of position-based and speed.If after having searched for, the number of vehicles of searching in plain window is 0, and the length and width of search window are respectively with δ, and ε is that step-length is increased, i.e. a=a+ δ, and b=b+ ε, until search vehicle.
Step 5, carry out the calculating of displacement variable and velocity variable to the target searched, select optimal objective, completes coupling.
Calculate displacement variable Δ S and the velocity variable Δ V of vehicle, wherein by the Euclidean that calculates the two vehicles geometric center, apart from obtaining, and velocity variable Δ V need to calculate displacement variable poor of same car to Δ S.Then according to D=Δ S
2+ Δ V
2Calculate the displacement variable of vehicle and the quadratic sum of velocity variable.From all object searches, find have least square and target, can complete tracing process.
Claims (6)
1. the vehicle matching process based on video is characterized in that following steps:
Step 1: obtain the vehicle monitoring video, utilize the background frame difference method to be partitioned into the moving vehicle in each two field picture.
Obtain the vehicle monitoring video from video capture device, utilize the background frame difference method to be partitioned into the moving vehicle in each two field picture.
Step 2: record the position of vehicle in each two field picture, set up the vehicle position data storehouse.
At first record the vehicle location in each two field picture, obtain the minimum rectangle outline of vehicle; Then find the geometric center position of minimum rectangle outline, the unique point using this as vehicle; Then deposit the unique point of all vehicles in database again.
Step 3: centered by the vehicle characteristics point of needs coupling, in the adjacent image frame, place search window.
The vehicle of select to need following the trail of, the geometric center of minimum rectangle outline of this vehicle of take is window center, places a rectangular search window on consecutive frame, this search window defines the scope of vehicle search.The length of side of this rectangular window is T*V* α, and wherein T means the time interval between consecutive frame, and V means the maximum limit speed per hour of this road, and α means the mapping relations of actual range and imaging plane in scene, the i.e. ratio of actual range and video middle distance.
Step 4: search for target vehicle in search window, and adjust window size until search vehicle, the registration of vehicle position according to the Search Results self-adaptation.
In the scope limited at search window, search vehicle from the set of vehicle characteristics point data.If do not search vehicle, automatically increase window size, the step-length of increase is 0.2*T*V* α, until search vehicle; When search window exceeds monitoring range, search stops; Perhaps, when searching plain window and be greater than T*405* α, search stops.Record all vehicle characteristics points that search, deposit in a set to be screened.
Step 5: calculate the displacement variable and the velocity variable that search target, select optimal objective, complete coupling.
Calculate displacement variable Δ S and the velocity variable Δ V of vehicle from set to be screened, according to both quadratic sum D=Δ S
2+ Δ V
2Weigh both motion state variable quantities, then choose D and mated as optimal objective.
2. according to the described vehicle fast matching method based on video of claim 1, it is characterized in that: the geometric center that the vehicle characteristics for coupling in step 2 is the vehicle minimum outline.
3. according to the described vehicle fast matching method based on video of claim 1, it is characterized in that: the window for search in step 3 for rectangle, the initial size of rectangular window is T*V* α, wherein T means the real time spacing between consecutive frame, V means the maximum limit speed per hour of this road, α means the mapping relations of actual range and imaging plane in scene, i.e. actual range and ratio at the video middle distance.
4. according to the described vehicle fast matching method based on video of claim 1, it is characterized in that: the size of the search window in step 4 is self-adaptation adjustment.When not searching vehicle, the step-length that window size increases is 0.2*T*V* α, and wherein the increase of window has two upper limits: (1). exceed the detection coverage; (2) exceed T*405* α.
5. according to the described vehicle fast matching method based on video of claim 1, it is characterized in that: the displacement variable that the velocity variable computing method in step 5 are the vehicle front cross frame poor: Δ V=Δ S
1-Δ S
2, Δ S wherein
1The displacement variable that means vehicle minimum rectangle outline geometric center previous frame and real-time frame, Δ S
2Mean the displacement variable that the vehicle previous frame calculates.
6. according to the described vehicle fast matching method based on video of claim 1, it is characterized in that: displacement variable and the quadratic sum of velocity variable, i.e. D=Δ S that the index of weighing the state of motion of vehicle variable quantity in step 5 is vehicle minimum rectangle profile geometric center
2+ Δ V
2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310337138.6A CN103425764B (en) | 2013-07-30 | 2013-07-30 | Vehicle matching method based on videos |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310337138.6A CN103425764B (en) | 2013-07-30 | 2013-07-30 | Vehicle matching method based on videos |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103425764A true CN103425764A (en) | 2013-12-04 |
| CN103425764B CN103425764B (en) | 2017-04-12 |
Family
ID=49650503
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310337138.6A Expired - Fee Related CN103425764B (en) | 2013-07-30 | 2013-07-30 | Vehicle matching method based on videos |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103425764B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103839259A (en) * | 2014-02-13 | 2014-06-04 | 西安交通大学 | Optimal matching block searching method and device for image |
| CN106022307A (en) * | 2016-06-08 | 2016-10-12 | 中国科学院自动化研究所 | Remote sensing image vessel detection method based on vessel rotation rectangular space |
| CN108108680A (en) * | 2017-12-13 | 2018-06-01 | 长安大学 | A kind of front vehicle identification and distance measuring method based on binocular vision |
| WO2018205120A1 (en) * | 2017-05-09 | 2018-11-15 | 深圳市速腾聚创科技有限公司 | Target tracking method, smart device and storage medium |
| CN109255803A (en) * | 2018-08-24 | 2019-01-22 | 长安大学 | A kind of displacement calculation method for the moving target soundd out based on displacement |
| CN111108342A (en) * | 2016-12-30 | 2020-05-05 | 迪普迈普有限公司 | Visual ranging and pairwise alignment for high definition map creation |
| CN111354191A (en) * | 2020-03-09 | 2020-06-30 | 深圳大学 | Lane driving condition determining method, device and equipment and storage medium |
| CN111523385A (en) * | 2020-03-20 | 2020-08-11 | 北京航空航天大学合肥创新研究院 | Stationary vehicle detection method and system based on frame difference method |
| CN112836631A (en) * | 2021-02-01 | 2021-05-25 | 南京云计趟信息技术有限公司 | Vehicle axle number determining method and device, electronic equipment and storage medium |
| CN116884236A (en) * | 2023-06-26 | 2023-10-13 | 中关村科学城城市大脑股份有限公司 | Traffic flow collection device and traffic flow collection method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100013935A1 (en) * | 2006-06-14 | 2010-01-21 | Honeywell International Inc. | Multiple target tracking system incorporating merge, split and reacquisition hypotheses |
| CN102222346B (en) * | 2011-05-23 | 2013-03-13 | 北京云加速信息技术有限公司 | Vehicle detecting and tracking method |
-
2013
- 2013-07-30 CN CN201310337138.6A patent/CN103425764B/en not_active Expired - Fee Related
Non-Patent Citations (1)
| Title |
|---|
| 陈泓屺等: "基于二次帧差背景提取的车辆追踪方法", 《广东工业大学学报》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103839259B (en) * | 2014-02-13 | 2016-11-23 | 西安交通大学 | A kind of image searches blocks and optimal matching blocks method and device |
| CN103839259A (en) * | 2014-02-13 | 2014-06-04 | 西安交通大学 | Optimal matching block searching method and device for image |
| CN106022307A (en) * | 2016-06-08 | 2016-10-12 | 中国科学院自动化研究所 | Remote sensing image vessel detection method based on vessel rotation rectangular space |
| CN106022307B (en) * | 2016-06-08 | 2019-09-27 | 中国科学院自动化研究所 | Remote sensing images ship detection method based on ship rotation rectangular space |
| CN111108342A (en) * | 2016-12-30 | 2020-05-05 | 迪普迈普有限公司 | Visual ranging and pairwise alignment for high definition map creation |
| CN111108342B (en) * | 2016-12-30 | 2023-08-15 | 辉达公司 | Visual range method and pair alignment for high definition map creation |
| WO2018205120A1 (en) * | 2017-05-09 | 2018-11-15 | 深圳市速腾聚创科技有限公司 | Target tracking method, smart device and storage medium |
| CN108108680A (en) * | 2017-12-13 | 2018-06-01 | 长安大学 | A kind of front vehicle identification and distance measuring method based on binocular vision |
| CN109255803B (en) * | 2018-08-24 | 2022-04-12 | 长安大学 | Displacement calculation method of moving target based on displacement heuristic |
| CN109255803A (en) * | 2018-08-24 | 2019-01-22 | 长安大学 | A kind of displacement calculation method for the moving target soundd out based on displacement |
| CN111354191A (en) * | 2020-03-09 | 2020-06-30 | 深圳大学 | Lane driving condition determining method, device and equipment and storage medium |
| CN111354191B (en) * | 2020-03-09 | 2022-05-20 | 深圳大学 | Lane driving condition determining method, device and equipment and storage medium |
| CN111523385A (en) * | 2020-03-20 | 2020-08-11 | 北京航空航天大学合肥创新研究院 | Stationary vehicle detection method and system based on frame difference method |
| CN112836631A (en) * | 2021-02-01 | 2021-05-25 | 南京云计趟信息技术有限公司 | Vehicle axle number determining method and device, electronic equipment and storage medium |
| CN116884236A (en) * | 2023-06-26 | 2023-10-13 | 中关村科学城城市大脑股份有限公司 | Traffic flow collection device and traffic flow collection method |
| CN116884236B (en) * | 2023-06-26 | 2024-04-16 | 中关村科学城城市大脑股份有限公司 | Traffic flow collection device and traffic flow collection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103425764B (en) | 2017-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103425764A (en) | Vehicle matching method based on videos | |
| CN106951879B (en) | Multi-feature fusion vehicle detection method based on camera and millimeter wave radar | |
| WO2022141914A1 (en) | Multi-target vehicle detection and re-identification method based on radar and video fusion | |
| WO2018177026A1 (en) | Device and method for determining road edge | |
| CN104700414A (en) | Rapid distance-measuring method for pedestrian on road ahead on the basis of on-board binocular camera | |
| CN103226834B (en) | A kind of image motion target signature point method for fast searching | |
| CN110379168B (en) | Traffic vehicle information acquisition method based on Mask R-CNN | |
| Huang et al. | Lane detection based on inverse perspective transformation and Kalman filter | |
| Liu et al. | A survey of vision-based vehicle detection and tracking techniques in ITS | |
| CN103903019A (en) | Automatic generating method for multi-lane vehicle track space-time diagram | |
| CN104183127A (en) | Traffic surveillance video detection method and device | |
| CN115049700A (en) | Target detection method and device | |
| CN114049382B (en) | Target fusion tracking method, system and medium in intelligent network connection environment | |
| Zhang et al. | A real-time curb detection and tracking method for UGVs by using a 3D-LIDAR sensor | |
| CN102034355A (en) | Feature point matching-based vehicle detecting and tracking method | |
| CN103886325A (en) | Cyclic matrix video tracking method with partition | |
| CN103150908A (en) | Average vehicle speed detecting method based on video | |
| Zhang et al. | Multiple vehicle-like target tracking based on the velodyne lidar | |
| CN103150550B (en) | A kind of road pedestrian event detection method based on gripper path analysis | |
| Gomaa et al. | Real-time algorithm for simultaneous vehicle detection and tracking in aerial view videos | |
| Liu et al. | Extending the detection range for low-channel roadside LiDAR by static background construction | |
| CN116503818A (en) | Multi-lane vehicle speed detection method and system | |
| Rasmussen et al. | On-vehicle and aerial texture analysis for vision-based desert road following | |
| Qing et al. | A novel particle filter implementation for a multiple-vehicle detection and tracking system using tail light segmentation | |
| Wang et al. | Geometry constraints-based visual rail track extraction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170412 Termination date: 20190730 |