CN102592454A - Intersection vehicle movement parameter measuring method based on detection of vehicle side face and road intersection line - Google Patents
Intersection vehicle movement parameter measuring method based on detection of vehicle side face and road intersection line Download PDFInfo
- Publication number
- CN102592454A CN102592454A CN2012100495630A CN201210049563A CN102592454A CN 102592454 A CN102592454 A CN 102592454A CN 2012100495630 A CN2012100495630 A CN 2012100495630A CN 201210049563 A CN201210049563 A CN 201210049563A CN 102592454 A CN102592454 A CN 102592454A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- detection
- utilize
- intersection
- image
- 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.)
- Pending
Links
Images
Landscapes
- Traffic Control Systems (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an intersection vehicle movement parameter measuring method based on detection of a vehicle side face and a road intersection line, which comprises the steps of: monitoring an intersection by installing a camera at the intersection; extracting an interested region containing the fewest complete movement objects through such methods as the image processing technology, a series of priori knowledge, background differential, morphologic image processing and outline lookup and the like; reconstructing the background through the improved block processing technology to update the background; eliminating a vehicle shadow through an HSV (hue, saturation and value) shadow detection algorithm of a first order gradient shearing algorithm; detecting the vehicle side face and the road intersection line in the interested region through certain recognition rules; calibrating the camera through a marker and marked line information on the road; creating two-dimensional bottom surface models of different vehicle models, and obtaining the central point position of the vehicle after matching with the intersection line; calculating the speed and acceleration of the vehicle through continuous vehicle position data by vehicle tracking; and finally, performing data filtering processing in a Kalman filtering manner.
Description
Technical field
The invention belongs to intelligent transportation monitoring field, relate in particular to a kind of measuring technique of pure video to the vehicle location under the scene low angle monitoring condition of crossing, car speed, vehicle acceleration.
Background technology
The kinematic parameter of vehicle is meant the displacement of vehicle, speed, parameters such as acceleration.
The real-time and dynamic vehicle movement parameter is the key point that realizes traffic intelligentization.
Practical at present vehicle movement parameter acquisition technique has hardware measuring method and software measurement method.
Hardware approach comprises that velocity radar, toroid winding detecting device, ultrasonic detector, microwave detector, infrared detector etc. carry out vehicle speed measurement.The hardware detection method has disadvantage, and the toroid winding detecting device need break the road and install; Velocity radar, ultrasonic detector, microwave detector, infrared detector mobility are not strong; And the common shortcomings of these technology comprise being only limited to and detect the mean velocity information of vehicle through a certain segment distance, are difficult to realize the tracking of the speed of a motor vehicle; Can not carry out moving object identification; Cost is higher.So be adapted at using on highway or a certain highway section crossing that very difficult application vehicle, pedestrian, bicycle mix.
Software approach is main with video image processing technology, and it is in the majority through the method for the average velocity of coil that the method that main at present employing is provided with the video virtual coil detects vehicle, and this method is mainly used in the measurement of car speed on a certain highway section.
Summary of the invention
The objective of the invention is; Deficiency to prior art; Provide a kind of based on the vehicle movement mathematic(al) parameter measuring method under the crossing scene low angle monitoring condition of Video Detection, the present invention is different from existing video detecting method part and is to have proposed to confirm vehicle location through vehicle bottom surface 2D Wireframe model and the method that vehicular sideview and road surface intersection mate, thereby utilizes two-dimentional photogrammetric technology to calculate vehicle location; Speed, information such as acceleration.Utilize the vehicle tracking technology to realize the continuous kinematics parameters of vehicle at last.
In order to realize above-mentioned purpose of the present invention, can realize through following technical scheme: a kind of crossing vehicle movement parameter measuring method based on car two sides and the detection of road surface intersection may further comprise the steps:
(1) live video stream obtains and transmit: the high definition camera through being installed on the monitoring bar obtains the real-time video flow data, then by Network Transmission to industrial computer.
(2) camera calibration: through on-the-spot identifier marking parameter of road and road photographs camera is demarcated, obtained the transition matrix of image coordinate and world coordinates.This patent is to utilize vehicle ground 2D Wireframe model and vehicular sideview and ground intersection to mate to confirm vehicle location; Because vehicular sideview and ground intersection and vehicle ground 2D Wireframe model projection result are on the ground, so can come the measuring vehicle position with two-dimentional photogrammetric technology.
(3) vehicle detection: to the guarded region of regulation, utilize background subtraction point-score and moving object profile correlativity to carry out interested area of vehicle and detect, vehicle shadow is eliminated, and utilizes the background reconstruction technology in certain time interval, to upgrade background.
(4) vehicle location: in the vehicle interesting district, detect the straight-line segment in the vehicle binary image through rim detection and Hough straight-line detection.Thereby from the pedestrian, bicycle is distinguished vehicle in the motor vehicle.Utilize vehicular sideview base straight line decision rule, confirm vehicular sideview base straight-line segment.And utilize vehicle bottom surface two dimensional model and vehicle bottom mating, iris out the vehicle base surface area, thereby confirm the vehicle center position.
(5) vehicle tracking: through the vehicle size that vehicle detection and vehicle location obtain, information such as vehicle location in conjunction with the tracking of kalman filter method realization vehicle, with the generation of track of vehicle, in conjunction with transition matrix, are calculated the position of vehicle in world coordinate system.Through vehicle position information, and the calculating of the time interval between video image car speed, acceleration.
(6) data processing: utilize Kalman filtering that kinematic parameter is handled, improve the precision and the stability of testing result.。
With the prior art ratio, the present invention has following characteristics:
1, the present invention is made up of a high definition camera and industrial computer, does not need other vehicle equipment; Vehicle in each two field picture detects, and matees to confirm vehicle location through vehicle bottom surface 2D Wireframe model and vehicular sideview and road surface intersection, and preserves vehicle location in real time; And vehicle pixel center point; Size and sequence number are utilized space time information, and coupling and prediction scheduling algorithm are realized the tracking of vehicle.Each car position of last output in real time, speed, acceleration, the present invention is simple in structure, practicality.
2, since vehicular sideview and road surface intersection near ground, so the present invention can adopt the two-dimensional measurement method to come the measuring vehicle position.
The major advantage that adopts two-dimentional photogrammetric survey method to detect is following:
(1) equipment is simple, adopts a camera (monocular vision) images acquired;
(2) need not to obtain the three-dimensional coordinate of measurement point, can utilize on-the-spot identifier marking that image sequence is demarcated easily.Need not video camera (camera) inner parameter is demarcated specially, simplified calibration process.
3, because utilizing two-dimentional photogrammetric survey method to carry out vehicle location measures, improved the vehicle location real-time.
4, the camera fixed position belongs to the low angle monitoring greater than about about 7 meters apart from the road surface, needn't road pavement vertically take, and has reduced the installation difficulty of camera to a great extent.
5, safeguard the aspect, removable, need not break the road and install.
Description of drawings
Fig. 1 is the crossing vehicle movement parameter detection method figure based on Video Detection;
Fig. 2 is vehicular sideview and road surface intersection synoptic diagram
Fig. 3 is a vehicle movement parameter detection system block diagram
Fig. 4 is a vehicle detection practical implementation step
Fig. 5 is vehicle location and vehicle tracking synoptic diagram
Fig. 6 is the vehicle location parametric plot of the present invention's output
Fig. 7 is the car speed parametric plot of the present invention's output
Fig. 8 is the vehicle acceleration parametric plot of the present invention's output
Embodiment
The present invention detects vehicular sideview and road surface intersection to the vehicle in each two field picture of video flowing, and utilizes vehicle bottom surface 2D Wireframe model and intersection coupling, thereby confirms the vehicle center position, has guaranteed the accuracy of vehicle position information under the monitoring condition of low latitude.Through real-time detection and tracking, realize the measurement of vehicle location again through two-dimentional photogrammetric technology, and then calculate the instantaneous speed of a motor vehicle and the acceleration of vehicle the vehicle center position; With Kalman filtering data are carried out filtering at last, realized the measurement of vehicle movement parameter, guaranteed measuring accuracy; And low latitude, crossing monitoring, no signal lamp crossing have been solved; The pedestrian, bicycle, the vehicle movement parameter problems of measurement of vehicle and the complicated traffic scene deposited.
The crossing vehicle movement mathematic(al) parameter measuring method that the present invention is based on Video Detection comprises the steps:
(1) live video stream obtains and transmit: the high definition camera through being installed on the monitoring bar obtains the real-time video flow data, then by Network Transmission to industrial computer.
(2) camera calibration: through on-the-spot identifier marking parameter of road and road photographs camera is demarcated, obtained the transition matrix of image coordinate and world coordinates.
(3) vehicle detection: through the area-of-interest of regulation guarded region, utilize background subtraction point-score and moving object profile correlativity to carry out interested area of vehicle and detect, vehicle shadow is eliminated, and utilizes the background reconstruction technology in certain time interval, to upgrade background.
(4) vehicle location: in the vehicle interesting district, detect the straight-line segment in the vehicle binary image through rim detection and Hough straight-line detection.Thereby from the pedestrian, bicycle is distinguished vehicle in the motor vehicle.Utilize vehicular sideview base straight line decision rule, confirm vehicular sideview base straight-line segment.And utilize vehicle bottom surface two dimensional model and vehicle bottom mating, iris out the vehicle base surface area, thereby confirm the vehicle center position.
(5) vehicle tracking: through the vehicle size that vehicle detection and vehicle location obtain, information such as vehicle location in conjunction with the tracking of kalman filter method realization vehicle, with the generation of track of vehicle, in conjunction with transition matrix, are calculated the position of vehicle in world coordinate system.Through vehicle position information, and the calculating of the time interval between video image car speed, acceleration.
(6) data processing: utilize Kalman filtering that kinematic parameter is handled, improve the precision and the stability of testing result.。
Below in conjunction with accompanying drawing and specific embodiment the present invention is further specified:
Fig. 1 for the installation of this method, synoptic diagram is set, mainly comprise CCD high definition camera 1, monitoring bar 2, netting twine, 3, switch 4, industrial computer 5.Wherein CCD high definition camera is aimed at the crossing, makes camera can clap four zebra crossings of full crossing, selects for use suitable resolution of video camera and camera lens to guarantee that video image is clear as far as possible; The height of monitoring bar will be monitored bar and be installed in any one jiao of crossing generally more than 7 meters.
This method only needs on the monitoring bar, to set up a CCD high definition camera in practical application, does not need other vehicle equipments; Only need utilize markings parameter and photographs on the intersection pavement to come camera is demarcated.Utilize Video processing and pattern-recognition related algorithm then, in real-time video flowing, accomplish vehicle body base straight line (approximate vehicular sideview and the road surface intersection of being regarded as) and detect, and export the vehicle location of high confidence level, car speed, the information that vehicle quickens.The present invention is simple in structure, and layout setting is portable, and the vehicle movement parameter accuracy of detection is high.
Fig. 2 is vehicle body side and road surface intersection synoptic diagram.The present invention proposes the vehicle positioning method of a kind of vehicle bottom surface two dimensional model and vehicular sideview and road surface intersection coupling, and the definition of vehicular sideview and road surface intersection is: on the intersection on vehicle body side and road surface, and the line segment of this from the headstock to the tailstock.Black line segment as shown in Figure 2.So because this line can utilize two-dimentional photogrammetric survey method detection line fragment position information on the road surface.
Fig. 3 is the system architecture diagram of this method.This structured flowchart also is a main flow chart of the present invention, and this method mainly is divided into camera calibration module, the background reconstruction module; The region of interest detection module, shadow cancellation module, intersection detection module; The vehicle location module, the vehicle tracking module, vehicle movement mathematic(al) parameter measurement module is formed.Background reconstruction module wherein, the region of interest detection module, the shadow cancellation module, the intersection detection module has been formed the vehicle detection module.The camera calibration module utilizes the image parameter of crossing background image and markings and on-site parameters that camera is demarcated, and obtains the transformation matrix of plane of delineation coordinate system and on-the-spot plane world coordinate system.The vehicle detection module is to utilize the background reconstruction technology, the region of interest detection technique, and the shadow technology for eliminating, intersection detects, and Model Matching waits and realizes vehicle detection.The effect of vehicle location module is a field position coordinate of confirming vehicle to be detected.The vehicle tracking module realizes the real-time follow-up of vehicle to be detected, thereby generates the vehicle movement track.Vehicle movement mathematic(al) parameter measurement module is used for calculating the field position route and the speed of vehicle, acceleration.
Fig. 4 is the practical implementation step of vehicle detection module.
At first the present invention utilizes averaging method to extract initial background, promptly utilizes the video sequence of some to calculate the background image of guarded region.
The incremental form of averaging method is:
B wherein
j(x, y) expression j frame pixel (x, the background of y) locating, I
j(x y) then representes j frame pixel (x, gray-scale value y) or color vector.Because there is incremental form in averaging method, therefore be fit to real-time Video processing.It can also be seen that for averaging method, the reconstruct of background and renewal are same processes, this simple characteristic helps Project Realization.This also is the reason that the present invention selects to ask in this way initial background.
The present invention utilizes the background subtraction point-score to carry out foreground detection.The background image of video image and initial background image or reconstruct after the difference image binary conversion treatment, can obtain foreground image after making difference.If f
k(x, the y) two field picture of expression input video sequence, B (x, y) expression initial background or reconstruction background image, F (then foreground detection can be expressed as for x, y) expression prospect mask (mask) image:
(x is a binary image y) to F, and 1 represents foreground area, 0 representative background area.Th represents binary-state threshold, and the difference image pixel value is thought foreground area greater than the point of Th, otherwise is the background area.
Area-of-interest detects concrete implementation method: the prospect mask is through behind the morphologic filtering; Through searching the moving object outline; And utilize the minimum rectangle frame to comprise outline one by one; With the region of interest of rectangular shaped rim position as vehicle; Owing to the situation that a moving object is split into several portions in the foreground detection process, may occur, so the present invention adopts the rectangular area that meets a straightforward principle is merged, the consistance rule has overlapping for rectangle frame or comprises characteristics.This merging, object sometimes that the position is approaching or adhesion merges to a zone.This situation will be in the intersection testing process be separated the moving object of the same area.
Background reconstruction module practical implementation method is: utilize improved treatment technology to obtain new background.Through setting certain time interval, intercepting is used for the image sequence that background is rebuild from video image, and label is 1,2..N.
After the area-of-interest detection, change region of interest into white area, non-region of interest is an original image.
The sub-piece that above-mentioned image is divided into 8*8.
Through judging that the white pixel statistic is judged in every sub-block, this sub-block is prospect or background.
The sub-piece of used background in the N frame is combined into new background, will produces blocky effect because of the brightness difference of sub-piece like this, and have the sub-piece of failing to rebuild fully.To this problem, can adopt qualified a plurality of adjacent sub-blocks are averaged after, add estimated background again.For the sub-piece of failing to rebuild fully, piece corresponding in the desirable average scene splices.
Shadow is eliminated concrete implementation method: the bright utilization of this law finishes the HSV shadow Detection algorithm that unification ladder degree is wiped out algorithm, and this method realizes simple, and robustness is good, can detect vehicle shadow accurately.
Osculatory detects concrete implementation method: after the binary image in the area-of-interest is carried out the Sobel rim detection; Utilize hough transform to detect the vehicle body straight line; Often pedestrian and bicycle do not possess tangible linear feature, so can get rid of bicycle and pedestrian's interference thus.But there are many straight lines on the vehicle binaryzation edge image usually, therefore need formulation vehicular sideview and road surface intersection judgment rule to discern intersection.This rule comprises the length of straight line, the position of straight line, rectilineal interval, the correlativity of the whole angle of straight line angle and vehicle region.This method can be identified in two the above vehicles vehicular sideview and road surface intersection separately of same area-of-interest equally, thus the difference vehicle.We after the identification intersection, all vehicular sideviews that need to detect on the statistics monitoring image and the length of road surface intersection.Detect the length distribution situation of straight line according to zones of different, confirm zones of different respectively, the blanket length of different automobile types vehicular sideview and road surface intersection, and the vehicular sideview and the road surface intersection of vehicle carry out the normalization processing separately to utilize this length.
Fig. 5 is vehicle location and vehicle tracking synoptic diagram, has shown vehicle location result and tracking effect.The vehicle location module is to realize through Model Matching; Concrete implementation method is: utilize the relation of image coordinate, on-the-spot road surface coordinate, vehicle bottom surface model coordinate to set up the vehicle bottom surface 2D Wireframe model of a fixed size respectively for different automobile types; Through two dimensional model and vehicular sideview and road surface intersection coupling, with the 2D Wireframe model projection to image, like dark border among Fig. 5; Thereby utilize the model coordinate; On-the-spot coordinate, the image coordinate transforming relationship calculates 2D Wireframe model point midway, i.e. the position of vehicle on the road surface.Utilize information such as vehicle size position that the vehicle in the guarded region of regulation is followed the tracks of then, Fig. 5 (a) be vehicle 6 beginnings by first two field picture on following the tracks of, the last frame image that Fig. 5 (b) is followed the tracks of for vehicle 6 quilts.
Fig. 6 is the curve map of location parameter after through the Kalman filtering smoothing processing that utilizes vehicle among Fig. 5 of the inventive method output.
Fig. 7 calculates and filtered speed data curve map through the vehicle movement parameter measurement module for vehicle among Fig. 5 of the present invention's output; Wherein Fig. 7 (1) is the rate curve of vehicle in the on-the-spot world coordinate system X-direction of monitoring, and Fig. 7 (2) is the rate curve of vehicle in the on-the-spot world coordinate system X-direction of monitoring.The present invention is followed the tracks of back the 2nd frame from vehicle and is begun to calculate the speed of a motor vehicle, because Kalman filter is in learning state when handling former frame data, error is bigger than normal, becomes accurately, stablizes since the 4th frame filtering data.
Fig. 8 calculates and filtered acceleration information curve map through the vehicle movement parameter measurement module for vehicle among Fig. 5 of the bright output of this law; Wherein Fig. 8 (1) is the accelerating curve of vehicle in the on-the-spot world coordinate system X-direction of monitoring, and Fig. 8 (2) is the accelerating curve of vehicle in the on-the-spot world coordinate system X-direction of monitoring.The present invention is followed the tracks of back the 3rd frame from vehicle and is begun to calculate the speed of a motor vehicle, because Kalman filter is in learning state when handling former frame data, error is bigger than normal, becomes accurately, stablizes since the 5th frame filtering data.
Claims (5)
1. the crossing vehicle movement parameter measuring method based on vehicular sideview and the detection of road surface intersection is characterized in that, comprises the steps:
(1) live video stream obtains and transmit: the high definition camera through being installed on the monitoring bar obtains the real-time video flow data, then by Network Transmission to industrial computer.
(2) camera calibration: through on-the-spot identifier marking parameter of road and road photographs camera is demarcated, obtained the transition matrix of image coordinate and world coordinates.This patent is to utilize vehicular sideview and ground intersection to confirm vehicle location, because vehicular sideview and ground intersection are near ground, so can come the measuring vehicle position with two-dimentional photogrammetric technology.
(3) vehicle detection: at first confirm the regulation guarded region, be used for judging that at tracking phase vehicle to be detected whether in this zone, determines whether following the tracks of.Utilize background subtraction point-score and moving object profile correlativity to carry out interested area of vehicle and detect, vehicle shadow is eliminated, and utilizes the background reconstruction technology in certain time interval, to upgrade background.
(4) vehicle location: in the vehicle interesting district, detect the straight-line segment in the vehicle binary image through rim detection and Hough straight-line detection.Thereby from the pedestrian, bicycle is distinguished vehicle in the motor vehicle.Utilize vehicular sideview base straight line decision rule, confirm vehicular sideview base straight-line segment.And utilize vehicle bottom surface two dimensional model and vehicle bottom mating, iris out the vehicle base surface area, thereby confirm the vehicle center position.
(5) vehicle tracking: through the vehicle size of vehicle detection and vehicle location acquisition; Information such as vehicle location are in conjunction with the tracking of kalman filter method realization vehicle, with the generation of track of vehicle; In conjunction with transition matrix, calculate the vehicle position in the world coordinate system at the scene.Through vehicle position information, and the calculating of the time interval between video image car speed, acceleration.
(6) data processing: utilize Kalman filtering that kinematic parameter is handled, improve the precision and the stability of testing result.
2. according to the said a kind of crossing vehicle movement parameter measuring method of claim 1, it is characterized in that said step (2) is specially based on Video Detection:
(a) real-time video flowing.
(b) utilize the real time video image of some to adopt averaging method to obtain the background image of crossing.
(c) background image is preserved.
(d) according to the on-the-spot world coordinate system of setting up in the crossing; Measure and calculate 4 straight-line equations of zebra crossing edge under world coordinate system; Background image comprises four the complete zebra crossings in crossing; Pick up 2 points on the zebra crossing edge line on the background image through mouse; Obtain 4 straight-line equations under the image coordinate system, utilize the straight-line equation parameter of 4 groups of world coordinate systems and image coordinate system that camera is demarcated, calculate the transition matrix of image coordinate system and world coordinate system.
(e) preserve transition matrix.
3. according to the said a kind of crossing vehicle movement parameter measuring method of claim 1, it is characterized in that said step (3) is specially based on Video Detection:
(a) on background image, select useful zone in the guarded region, non-road area is got rid of.
(b) utilize background subtraction to assign to detect moving object, and to detected image process binaryzation, smothing filtering and morphologic filtering are handled.
(c) in step (b) result, search outline, utilize minimum boundary rectangle frame to iris out, and the rectangle frame that satisfies the consistance rule is merged.Be that a rectangle frame is corresponding with one or more complete moving objects.Thereby confirm the area-of-interest of vehicle.
(d) the HSV shadow Detection algorithm that in area-of-interest, utilizes the single order gradient to wipe out algorithm is eliminated the shade of vehicle, improves the accuracy of back vehicle location.
(e) in the certain hour interval, utilize improved treatment technology, rebuild background.
4. according to the said a kind of crossing vehicle movement parameter measuring method of claim 1, it is characterized in that said step (4) is specially based on Video Detection:
(a) in the vehicle interesting district, the binary image behind the morphologic filtering is carried out the Sobel rim detection.
(b) result to step (a) carries out the Hough straight-line detection, and utilizes vehicular sideview and road surface intersection decision rule identification intersection.
(c) utilize the statistics of vehicular sideview and road surface intersection length in the guarded region, use vehicular sideview and the road surface intersection normalization of regular length respectively same model.
(d) different automobile types is set up the vehicle ground two dimensional model of different sizes, and, utilized model coordinate systems model and vehicular sideview and road surface intersection coupling; Site coordinate system; Transforming relationship between the image coordinate system, the on-the-spot coordinate of model mid point, i.e. vehicle location after confirming to mate.
5. according to the said a kind of crossing vehicle movement parameter measuring method of claim 1, it is characterized in that said step (5) is specially based on Video Detection:
(a) confirm the size of detected vehicle in image, information such as position.
(b) utilize Kalman filtering method to follow the tracks of vehicle to be detected, and the field position coordinate of registration of vehicle.
(c) utilize the vehicle field position to calculate car speed, and acceleration information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100495630A CN102592454A (en) | 2012-02-29 | 2012-02-29 | Intersection vehicle movement parameter measuring method based on detection of vehicle side face and road intersection line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012100495630A CN102592454A (en) | 2012-02-29 | 2012-02-29 | Intersection vehicle movement parameter measuring method based on detection of vehicle side face and road intersection line |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102592454A true CN102592454A (en) | 2012-07-18 |
Family
ID=46481028
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012100495630A Pending CN102592454A (en) | 2012-02-29 | 2012-02-29 | Intersection vehicle movement parameter measuring method based on detection of vehicle side face and road intersection line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102592454A (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246872A (en) * | 2013-04-28 | 2013-08-14 | 北京农业智能装备技术研究中心 | Broad spectrum insect situation automatic forecasting method based on computer vision technology |
| CN103325255A (en) * | 2013-06-29 | 2013-09-25 | 佘若凡 | Regional traffic condition detection method based on photogrammetric technology |
| CN103854320A (en) * | 2012-12-05 | 2014-06-11 | 上海海事大学 | Automatic vehicle type identification device and method based on laser radar |
| CN103903278A (en) * | 2012-12-28 | 2014-07-02 | 重庆凯泽科技有限公司 | Moving target detection and tracking system |
| CN104166841A (en) * | 2014-07-24 | 2014-11-26 | 浙江大学 | Rapid detection identification method for specified pedestrian or vehicle in video monitoring network |
| CN104240222A (en) * | 2013-06-19 | 2014-12-24 | 贺江涛 | Intelligent detecting method and device for firefighting access blockage |
| CN104504912A (en) * | 2014-12-31 | 2015-04-08 | 长安大学 | Vehicle speed detection method based on three-dimensional horizontal and reverse projection plane |
| CN104915672A (en) * | 2014-03-13 | 2015-09-16 | 北京大学 | Rectangular building extraction method and system based on high-resolution remote sensing image |
| CN106056926A (en) * | 2016-07-18 | 2016-10-26 | 华南理工大学 | Video vehicle speed detection method based on dynamic virtual coil |
| CN106156745A (en) * | 2016-07-06 | 2016-11-23 | 宁波大学 | Pedestrian crossing traffic feature extraction method and device based on space-time track |
| CN106446790A (en) * | 2016-08-30 | 2017-02-22 | 上海交通大学 | Method for tracking and analyzing traffic video flow of fixed camera |
| CN107194386A (en) * | 2017-07-20 | 2017-09-22 | 东南大学 | A kind of intersection electric bicycle travel speed acquisition methods based on video |
| CN107392144A (en) * | 2017-07-20 | 2017-11-24 | 东南大学 | A kind of intersection electric bicycle driving trace acquisition methods based on video |
| CN107644528A (en) * | 2017-08-02 | 2018-01-30 | 浙江工业大学 | A kind of vehicle queue length detection method based on vehicle tracking |
| CN107845290A (en) * | 2016-09-21 | 2018-03-27 | 意法半导体股份有限公司 | Junction alarm method, processing system, junction alarm system and vehicle |
| CN108154146A (en) * | 2017-12-25 | 2018-06-12 | 陈飞 | A kind of car tracing method based on image identification |
| CN108257410A (en) * | 2017-12-25 | 2018-07-06 | 陈飞 | The accurate air navigation aid in parking stall that a kind of on-site supervision cooperates with navigation system |
| CN108922196A (en) * | 2018-06-10 | 2018-11-30 | 昆明创培知识产权服务有限公司 | A kind of intersection collision avoidance system |
| CN109917359A (en) * | 2019-03-19 | 2019-06-21 | 福州大学 | Robust vehicle distances estimation method based on vehicle-mounted monocular vision |
| US10769813B2 (en) | 2018-08-28 | 2020-09-08 | Bendix Commercial Vehicle Systems, Llc | Apparatus and method for calibrating surround-view camera systems |
| CN111753797A (en) * | 2020-07-02 | 2020-10-09 | 浙江工业大学 | Vehicle speed measuring method based on video analysis |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0626654A2 (en) * | 1993-05-25 | 1994-11-30 | Matsushita Electric Industrial Co., Ltd. | Apparatus for measuring intervehicle distance by stereo vision |
| CN101251927A (en) * | 2008-04-01 | 2008-08-27 | 东南大学 | Vehicle detecting and tracing method based on video technique |
| CN102074113A (en) * | 2010-09-17 | 2011-05-25 | 浙江大华技术股份有限公司 | License tag recognizing and vehicle speed measuring method based on videos |
-
2012
- 2012-02-29 CN CN2012100495630A patent/CN102592454A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0626654A2 (en) * | 1993-05-25 | 1994-11-30 | Matsushita Electric Industrial Co., Ltd. | Apparatus for measuring intervehicle distance by stereo vision |
| CN101251927A (en) * | 2008-04-01 | 2008-08-27 | 东南大学 | Vehicle detecting and tracing method based on video technique |
| CN102074113A (en) * | 2010-09-17 | 2011-05-25 | 浙江大华技术股份有限公司 | License tag recognizing and vehicle speed measuring method based on videos |
Non-Patent Citations (3)
| Title |
|---|
| 张凤静等: "双目立体视觉测距系统目标车辆检测算法研究", 《重庆科技学院学报（自然科技版）》 * |
| 路红等: "基于背景重构和阴影消除的运动目标分割", 《中国图象图形学报》 * |
| 鲁光泉等: "基于视频的交叉口车辆运动参数检测技术http://www.paper.edu.cn/releasepaper/content/20106-351", 《中国科技论文在线》 * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103854320A (en) * | 2012-12-05 | 2014-06-11 | 上海海事大学 | Automatic vehicle type identification device and method based on laser radar |
| CN103854320B (en) * | 2012-12-05 | 2016-06-22 | 上海海事大学 | Vehicle automatic identifying method based on laser radar |
| CN103903278A (en) * | 2012-12-28 | 2014-07-02 | 重庆凯泽科技有限公司 | Moving target detection and tracking system |
| CN103246872A (en) * | 2013-04-28 | 2013-08-14 | 北京农业智能装备技术研究中心 | Broad spectrum insect situation automatic forecasting method based on computer vision technology |
| CN103246872B (en) * | 2013-04-28 | 2016-02-10 | 北京农业智能装备技术研究中心 | A kind of wide spectrum insect pest situation automatic telemetry method based on computer vision technique |
| CN104240222A (en) * | 2013-06-19 | 2014-12-24 | 贺江涛 | Intelligent detecting method and device for firefighting access blockage |
| CN103325255B (en) * | 2013-06-29 | 2016-01-20 | 佘若凡 | The method of region transportation situation detection is carried out based on photogrammetric technology |
| CN103325255A (en) * | 2013-06-29 | 2013-09-25 | 佘若凡 | Regional traffic condition detection method based on photogrammetric technology |
| CN104915672B (en) * | 2014-03-13 | 2018-08-03 | 北京大学 | A kind of Rectangle building extracting method and system based on high-resolution remote sensing image |
| CN104915672A (en) * | 2014-03-13 | 2015-09-16 | 北京大学 | Rectangular building extraction method and system based on high-resolution remote sensing image |
| CN104166841B (en) * | 2014-07-24 | 2017-06-23 | 浙江大学 | The quick detection recognition methods of pedestrian or vehicle is specified in a kind of video surveillance network |
| CN104166841A (en) * | 2014-07-24 | 2014-11-26 | 浙江大学 | Rapid detection identification method for specified pedestrian or vehicle in video monitoring network |
| CN104504912A (en) * | 2014-12-31 | 2015-04-08 | 长安大学 | Vehicle speed detection method based on three-dimensional horizontal and reverse projection plane |
| CN106156745A (en) * | 2016-07-06 | 2016-11-23 | 宁波大学 | Pedestrian crossing traffic feature extraction method and device based on space-time track |
| CN106056926A (en) * | 2016-07-18 | 2016-10-26 | 华南理工大学 | Video vehicle speed detection method based on dynamic virtual coil |
| CN106056926B (en) * | 2016-07-18 | 2021-07-30 | 华南理工大学 | Video vehicle speed detection method based on dynamic virtual coil |
| CN106446790A (en) * | 2016-08-30 | 2017-02-22 | 上海交通大学 | Method for tracking and analyzing traffic video flow of fixed camera |
| CN107845290A (en) * | 2016-09-21 | 2018-03-27 | 意法半导体股份有限公司 | Junction alarm method, processing system, junction alarm system and vehicle |
| CN107845290B (en) * | 2016-09-21 | 2021-07-27 | 意法半导体股份有限公司 | Intersection warning method, processing system, intersection warning system and vehicle |
| CN107392144A (en) * | 2017-07-20 | 2017-11-24 | 东南大学 | A kind of intersection electric bicycle driving trace acquisition methods based on video |
| CN107194386A (en) * | 2017-07-20 | 2017-09-22 | 东南大学 | A kind of intersection electric bicycle travel speed acquisition methods based on video |
| CN107644528A (en) * | 2017-08-02 | 2018-01-30 | 浙江工业大学 | A kind of vehicle queue length detection method based on vehicle tracking |
| CN108154146A (en) * | 2017-12-25 | 2018-06-12 | 陈飞 | A kind of car tracing method based on image identification |
| CN108257410A (en) * | 2017-12-25 | 2018-07-06 | 陈飞 | The accurate air navigation aid in parking stall that a kind of on-site supervision cooperates with navigation system |
| CN108922196A (en) * | 2018-06-10 | 2018-11-30 | 昆明创培知识产权服务有限公司 | A kind of intersection collision avoidance system |
| US10769813B2 (en) | 2018-08-28 | 2020-09-08 | Bendix Commercial Vehicle Systems, Llc | Apparatus and method for calibrating surround-view camera systems |
| CN109917359A (en) * | 2019-03-19 | 2019-06-21 | 福州大学 | Robust vehicle distances estimation method based on vehicle-mounted monocular vision |
| CN109917359B (en) * | 2019-03-19 | 2022-10-14 | 福州大学 | Robust vehicle distance estimation method based on vehicle-mounted monocular vision |
| CN111753797A (en) * | 2020-07-02 | 2020-10-09 | 浙江工业大学 | Vehicle speed measuring method based on video analysis |
| CN111753797B (en) * | 2020-07-02 | 2022-02-22 | 浙江工业大学 | Vehicle speed measuring method based on video analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102592454A (en) | Intersection vehicle movement parameter measuring method based on detection of vehicle side face and road intersection line | |
| CN108960183B (en) | Curve target identification system and method based on multi-sensor fusion | |
| CN103176185B (en) | Method and system for detecting road barrier | |
| US9652980B2 (en) | Enhanced clear path detection in the presence of traffic infrastructure indicator | |
| CN101075376B (en) | Intelligent video traffic monitoring system based on multi-viewpoints and its method | |
| Poggenhans et al. | Precise localization in high-definition road maps for urban regions | |
| CN111801711A (en) | Image annotation | |
| Arróspide et al. | Homography-based ground plane detection using a single on-board camera | |
| CN110379168B (en) | Traffic vehicle information acquisition method based on Mask R-CNN | |
| US20210341303A1 (en) | Clustering event information for vehicle navigation | |
| CN105676253A (en) | Longitudinal positioning system and method based on city road marking map in automatic driving | |
| CN103206957B (en) | The lane detection and tracking method of vehicular autonomous navigation | |
| CN105260699A (en) | Lane line data processing method and lane line data processing device | |
| CN111006655A (en) | Multi-scene autonomous navigation positioning method for airport inspection robot | |
| Fernández et al. | Free space and speed humps detection using lidar and vision for urban autonomous navigation | |
| Benedek et al. | Positioning and perception in LIDAR point clouds | |
| CN200990147Y (en) | Intelligent video traffic monitoring system based on multi-view point | |
| Revilloud et al. | An improved approach for robust road marking detection and tracking applied to multi-lane estimation | |
| CN103794050A (en) | Real-time transport vehicle detecting and tracking method | |
| CN111160132B (en) | Method and device for determining lane where obstacle is located, electronic equipment and storage medium | |
| Xu et al. | Road lane modeling based on RANSAC algorithm and hyperbolic model | |
| Cheng et al. | Semantic segmentation of road profiles for efficient sensing in autonomous driving | |
| Guerrieri et al. | Traffic flow variables estimation: An automated procedure based on moving observer method. potential application for autonomous vehicles | |
| Xuan et al. | Robust lane-mark extraction for autonomous driving under complex real conditions | |
| Chunmei et al. | Obstacles detection based on millimetre-wave radar and image fusion techniques |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120718 |