CN103345618A - Traffic violation detection method based on video technology - Google Patents
Traffic violation detection method based on video technology Download PDFInfo
- Publication number
- CN103345618A CN103345618A CN2013102512067A CN201310251206A CN103345618A CN 103345618 A CN103345618 A CN 103345618A CN 2013102512067 A CN2013102512067 A CN 2013102512067A CN 201310251206 A CN201310251206 A CN 201310251206A CN 103345618 A CN103345618 A CN 103345618A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- image
- detection method
- frame
- regulations
- 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
Abstract
Disclosed is a traffic violation detection method based on a video technology. The traffic violation detection method includes the steps of 1) loading a video image of a current frame and conducting dynamic update on the background, 2) conducting filtering processing on a target image and further conducting background difference to obtain a foreground image, 3) carrying out binaryzation on the foreground image, 4) judging the binaryzation image and whether a vehicle exists or not in a set detection area, if no, finishing processing of the frame and transferring to the next frame, and if yes, calculating the mass center position M of the vehicle, and 5) judging whether the vehicle violates regulations or not according to the mass center position. The traffic violation detection method based on the video technology is excellent in real-time performance and good in accuracy.
Description
Technical field
The present invention relates to a kind of break in traffic rules and regulations detection technique, especially a kind of break in traffic rules and regulations detection method based on video technique.
Background technology
Along with vehicle possess amount increases rapidly, the incidence of traffic hazard is also improving constantly.Therefore, safeguard that road traffic order, the act of violating regulations of correcting vehicle, assurance traffic safety, minimizing traffic hazard are our vital tasks always.The break in traffic rules and regulations behavior is the main factor that traffic hazard takes place.Therefore, the incidence of the reduction traffic hazard that the behavior of minimizing break in traffic rules and regulations can be effective.The break in traffic rules and regulations detection system order that can regulate the traffic effectively provides objective and accurate foundation for traffic department handles act of violating regulations.The violator is punished and educates, improve the consciousness of automobile driver greatly, strengthen traffic safety consciousness, reduce the accident and the traffic disturbance that cause because of the vehicle peccancy behavior, guarantee The coast is clear, save police strength.
The defective that existing break in traffic rules and regulations detection method exists is: real-time is relatively poor, accuracy is relatively poor.
Summary of the invention
The deficiency relatively poor for the real-time that overcomes existing break in traffic rules and regulations detection method, that accuracy is relatively poor the invention provides that a kind of real-time is good, accuracy is preferably based on the break in traffic rules and regulations detection method of video technique.
The technical solution adopted for the present invention to solve the technical problems is:
A kind of break in traffic rules and regulations detection method based on video technique, described detection method may further comprise the steps:
1) is written into the video image of present frame, background is dynamically updated;
2) target image is carried out filtering and handle, and carry out the background subtraction branch, obtain foreground image;
3) binaryzation foreground image;
4) binary image is judged in the detection zone of setting, whether vehicle is arranged; If do not have, then finish the processing to this frame, be transferred to next frame; If vehicle is arranged, then calculate the centroid position M of vehicle;
5) judge the act of violating regulations of vehicle according to centroid position:
(5.1) if the speed of a motor vehicle is greater than preset threshold in the fixed range, overspeed of vehicle then;
(5.2) if barycenter M and 'STOP' line ahead less than setting threshold, and signal lamp is in red light phase, then vehicle makes a dash across the red light;
(5.3) if M and lane line coordinate difference less than threshold value, then vehicle is got over line;
(5.4) if the difference of the ordinate of M less than threshold value, then vehicle drives in the wrong direction;
(5.5) if M in the setting threshold scope, then vehicle peccancy stops;
(5.6) if vehicle is difficult behavior, then finish the processing to this frame.
Further, described detection method is further comprising the steps of: store car picture violating the regulations when 6) judging vehicle peccancy, and this violation information is presented at system interface.
Further, in the described step 1), it is as follows that background is carried out dynamic updating process: utilize weight coefficient, regulate the speed of context update
dst(x,y)=α·src(x,y)+(1-α)·dst(x,y)ifmask(x,y)≠0, (1)
Wherein, src (x, y) expression source image pixels, dst (x, the background image pixel after y) expression is upgraded; α is weight parameter, is used for adjusting the speed of context update, and photo current is updated in the background; (x y) is mask plate in the renewal process to mask, directly the prospect result who extracts is calculated as mask plate at this place;
Utilize wavelet transformation that background image is further handled.Two-dimensional discrete wavelet conversion can resolve into source images 1 ll channel, with level, vertical 3 the details subgraphs in diagonal angle that reach, carry out after the one-level wavelet transformation again, to level, vertical, diagonal angle 3 details subgraphs mean filter, carry out wavelet reconstruction with low frequency part again, obtain new background image.
Further again, in the described step 4), the process of the centroid position M of calculating vehicle is as follows:
4.1) in foreground image, set surveyed area, adopt edge detection method to detect the rectangular profile of moving vehicle, the definition area compares:
r=A
rect/A
roi, (2)
Wherein, A
RectAnd A
RoiThe area of representing the surveyed area of the area of detected rectangular profile and setting respectively is as r〉during predetermined threshold value, this rectangular profile is exactly the initial detecting frame;
4.2) calculate the color histogram of tracking target, namely calculate this image at the histogram of the H in HSV space component, with the every bit that obtains (x, the probability distribution I of color y) in object module (x, y);
4.3) calculate moving target barycenter M (X, Y)
Wherein, M
00, M
10, M
01Be respectively 0 rank distance and the 1 rank distance of search box, (X is Y) for calculating the target barycenter that obtains for M, repeat transfer to follow the tracks of whole window center to M (X, Y), up to window displacement less than threshold value T, obtain size and the position of new rectangle frame, as the initial detecting frame of next frame.
Beneficial effect of the present invention mainly shows: real-time is good, accuracy is better.
Description of drawings
Fig. 1 is dynamic trace flow figure.
Fig. 2 is the measuring ability frame diagram.
Fig. 3 is that function realizes simple and easy process flow diagram.
Fig. 4 is testing process figure violating the regulations.
Fig. 5 is different update speed design sketch.
Fig. 6 is the vehicle tracking design sketch.
Fig. 7 is the vehicle design sketch that makes a dash across the red light.
Embodiment
Below in conjunction with accompanying drawing the present invention is further described.
With reference to Fig. 1~Fig. 7, a kind of break in traffic rules and regulations detection method based on video technique, described detection method may further comprise the steps:
1) is written into the video image of present frame, background is dynamically updated;
2) target image is carried out filtering and handle, and carry out the background subtraction branch, obtain foreground image;
3) binaryzation foreground image;
4) binary image is judged in the detection zone of setting, whether vehicle is arranged; If do not have, then finish the processing to this frame, be transferred to next frame; If vehicle is arranged, then calculate the centroid position M of vehicle;
5) judge the act of violating regulations of vehicle according to centroid position:
(5.1) if the speed of a motor vehicle is greater than preset threshold in the fixed range, overspeed of vehicle then;
(5.2) if barycenter M and 'STOP' line ahead less than setting threshold, and signal lamp is in red light phase, then vehicle makes a dash across the red light;
(5.3) if M and lane line coordinate difference less than threshold value, then vehicle is got over line;
(5.4) if the difference of the ordinate of M less than threshold value, then vehicle drives in the wrong direction;
(5.5) if M in the setting threshold scope, then vehicle peccancy stops;
(5.6) if vehicle is difficult behavior, then finish the processing to this frame.
Further, described detection method is further comprising the steps of: store car picture violating the regulations when 6) judging vehicle peccancy, and this violation information is presented at system interface.
In the present embodiment, adopt the background subtraction point-score based on wavelet transformation, in conjunction with the tracing based on feature, realize the detection of moving vehicle.Mainly be divided into following four steps:
1) at first filtered video image is carried out the background subtraction point-score, can in time detect moving target efficiently.Because the background subtraction point-score is subjected to the influence of change of background easily, so the background that need upgrade in time.This paper has adopted a kind of dynamic background update algorithm, can be in real time background image updating effectively.This algorithm utilizes weight coefficient, regulates the speed of context update
dst(x,y)=α·src(x,y)+(1-α)·dst(x,y)ifmask(x,y)≠0, (1)
Wherein, src (x, y) expression source image pixels, dst (x, the background image pixel after y) expression is upgraded; α is weight parameter, is used for adjusting the speed of context update, and photo current is updated in the background; (x y) is mask plate in the renewal process to mask, directly the prospect result who extracts is calculated as mask plate at this place, has avoided the interference of the prospect in the context process of upgrading.
α chooses the direct accuracy that influences vehicle detection.Renewal speed is big, and serious cavity appears in vehicle; Renewal speed is slow, and the movement locus of vehicle has a strong impact on the detection effect.Therefore, utilize wavelet transformation that background image is further handled.Two-dimensional discrete wavelet conversion can resolve into source images 1 ll channel, level, vertically reach 3 the details subgraphs in diagonal angle, after carrying out the one-level wavelet transformation, to level, vertical, diagonal angle 3 details subgraphs mean filter, carry out wavelet reconstruction with low frequency part again, obtain new background image, efficiently solve the retention problems of track of vehicle in the background.
2) set surveyed area in foreground image, adopt edge detection method to detect the rectangular profile of moving vehicle, when the area of this profile reached a certain threshold value, the vehicle ' s contour in just should the zone was as the initial rectangular frame of vehicle tracking.Definition area ratio
r=A
rect/A
roi, (2)
Wherein, A
RectAnd A
RoiThe area of representing the surveyed area of the area of detected rectangular profile and setting respectively.As r〉1/2 the time, this rectangular profile is exactly the initial detecting frame.
3) calculate the color histogram of tracking target, namely calculate this image at the histogram of the H in HSV space component, with the every bit that obtains (x, the probability distribution I of color y) in object module (x, y).
4) calculate moving target barycenter M (X, Y)
Wherein, M
00, M
10, M
01Be respectively 0 rank distance and the 1 rank distance of search box, (X is Y) for calculating the target barycenter that obtains for M, repeat adjust to follow the tracks of window center to M (X, Y), up to window displacement less than threshold value T, obtain size and the position of new rectangle frame, as the initial ranging frame of next frame.So circulation has just reached the effect of following the tracks of.This process as shown in Figure 1.
The function of the detection system of present embodiment mainly comprises three modules: video image load-on module, vehicles peccancy detection module, violation information memory module.The vehicles peccancy detection module mainly is divided into: image pre-service, vehicles peccancy detect, context update, as shown in Figure 2.
The used software of system development has Visual Studio2010, OpenCV2.3.1, and successfully disposes according to the development environment configuration file.
The video load-on module is realized by the function of the DrawToHDC () among the MFC among the OpenCV.Because this function in the CvvImage class, needs manually to add such header file and code file.
Comprised the function that various images are handled in the OpenCV storehouse, as gaussian filtering function G aussianBlur (), image space mapping function cvtColor (), background difference function absdiff (), threshold function table threshold (), rim detection Canny operator function canny (), histogram calculation function calcHist common mathematical functions such as ().Fig. 3 is the simple and easy process flow diagram of system.
Wherein,
OnTimer (): be mainly used in loading video;
Processing (): be used for the image pre-service, do work such as filtering, background difference, extraction prospect;
LineDetect (): detect stop line and lane line;
LightDetect (): detect lights state;
Tracking (): do Image Edge-Detection, set the effective coverage, and in this zone, detect moving vehicle, return initial block information;
VehicleInfo (): show vehicle peccancy information;
TrackDetect (): pursuit movement vehicle, the motion barycenter of return movement vehicle;
DrawTrack (): draw pursuit path;
ShowDetectImg (): storage also shows the vehicles peccancy picture;
PlayVideo (): display video.
Testing process figure violating the regulations as shown in Figure 4.As seen from Figure 4, the function of this system comprises image processing, vehicle tracking, identification violating the regulations and information preservation.Wherein the gordian technique in the image processing process is exactly the context update problem.This paper adopts adaptive algorithm to realize the real-time update of background, improves the vehicle detection effect.The gordian technique of this system is exactly vehicle tracking, namely according to the vehicle centroid position that obtains, and movement locus, set corresponding threshold value, judge which kind of act of violating regulations vehicle belongs to.As gained vehicle center-of-mass coordinate (x
M, y
M) ordinate not at preset threshold y
TIn the scope, namely work as y
MY
TThe time, and when detecting the signal lamp state and being red light phase, the expression vehicle makes a dash across the red light.After detecting vehicle peccancy, system will preserve picture violating the regulations automatically, and is presented on system's panel, provides the user to consult real-time violation information.
Context update speed directly influences the motion target detection accuracy, and testing used video frame rate was 25 frame/seconds, and this paper compares at different context update speed, effect as shown in Figure 5, renewal rate is respectively α=0.01,0.03,0.05,0.1.As seen from Figure 5, effect is carried out when renewal rate reaches α=0.03.
Realize behind the vehicle tracking effect as shown in Figure 6, Fig. 7 has shown vehicle peccancy information (because video resource is limited, this situation is to serve as to detect foundation with the left-side signal lamp, judges whether vehicle makes a dash across the red light).
Claims (4)
1. break in traffic rules and regulations detection method based on video technique, it is characterized in that: described detection method may further comprise the steps:
1) is written into the video image of present frame, background is dynamically updated;
2) target image is carried out filtering and handle, and carry out the background subtraction branch, obtain foreground image;
3) binaryzation foreground image;
4) binary image is judged in the detection zone of setting, whether vehicle is arranged; If do not have, then finish the processing to this frame, be transferred to next frame; If vehicle is arranged, then calculate the centroid position M of vehicle;
5) judge the act of violating regulations of vehicle according to centroid position:
(5.1) if the speed of a motor vehicle is greater than preset threshold in the fixed range, overspeed of vehicle then;
(5.2) if barycenter M and 'STOP' line ahead less than setting threshold, and signal lamp is in red light phase, then vehicle makes a dash across the red light;
(5.3) if M and lane line coordinate difference less than threshold value, then vehicle is got over line;
(5.4) if the difference of the ordinate of M less than threshold value, then vehicle drives in the wrong direction;
(5.5) if M in the setting threshold scope, then vehicle peccancy stops;
(5.6) if vehicle is difficult behavior, then finish the processing to this frame.
2. a kind of break in traffic rules and regulations detection method based on video technique as claimed in claim 1, it is characterized in that: described detection method is further comprising the steps of:
Store car picture violating the regulations when 6) judging vehicle peccancy, and this violation information is presented at system interface.
3. a kind of break in traffic rules and regulations detection method based on video technique as claimed in claim 1 or 2 is characterized in that: in the described step 1), it is as follows that background is carried out dynamic updating process: utilize weight coefficient, regulate the speed of context update
dst(x,y)=α·src(x,y)+(1-α)·dst(x,y)ifmask(x,y)≠0, (1)
Wherein, src (x, y) expression source image pixels, dst (x, the background image pixel after y) expression is upgraded; α is weight parameter, is used for adjusting the speed of context update, and photo current is updated in the background; (x y) is mask plate in the renewal process to mask, directly the prospect result who extracts is calculated as mask plate at this place;
Utilize wavelet transformation that background image is further handled.Two-dimensional discrete wavelet conversion can resolve into source images 1 ll channel, with level, vertical 3 the details subgraphs in diagonal angle that reach, carry out after the one-level wavelet transformation again, to level, vertical, diagonal angle 3 details subgraphs mean filter, carry out wavelet reconstruction with low frequency part again, obtain new background image.
4. a kind of break in traffic rules and regulations detection method based on video technique as claimed in claim 1 or 2 is characterized in that: in the described step 4), the process of centroid position M of calculating vehicle is as follows:
4.1) in foreground image, set surveyed area, adopt edge detection method to detect the rectangular profile of moving vehicle, the definition area compares:
r=A
rect/A
roi, (2)
Wherein, A
RectAnd A
RoiThe area of representing the surveyed area of the area of detected rectangular profile and setting respectively is as r〉during predetermined threshold value, this rectangular profile is exactly the initial detecting frame;
4.2) calculate the color histogram of tracking target, namely calculate this image at the histogram of the H in HSV space component, with the every bit that obtains (x, the probability distribution I of color y) in object module (x, y);
4.3) calculate moving target barycenter M (X, Y)
Wherein, M
00, M
10, M
01Be respectively 0 rank distance and the 1 rank distance of search box, (X is Y) for calculating the target barycenter that obtains for M, repeat transfer to follow the tracks of whole window center to M (X, Y), up to window displacement less than threshold value T, obtain size and the position of new rectangle frame, as the initial detecting frame of next frame.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013102512067A CN103345618A (en) | 2013-06-21 | 2013-06-21 | Traffic violation detection method based on video technology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013102512067A CN103345618A (en) | 2013-06-21 | 2013-06-21 | Traffic violation detection method based on video technology |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103345618A true CN103345618A (en) | 2013-10-09 |
Family
ID=49280413
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2013102512067A Pending CN103345618A (en) | 2013-06-21 | 2013-06-21 | Traffic violation detection method based on video technology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103345618A (en) |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103955705A (en) * | 2014-04-29 | 2014-07-30 | 银江股份有限公司 | Traffic signal lamp positioning, recognizing and classifying method based on video analysis |
CN104809887A (en) * | 2015-04-23 | 2015-07-29 | 杭州中威电子股份有限公司 | Detecting method for vehicle converse running on expressway and automatic alarm device |
CN105513371A (en) * | 2016-01-15 | 2016-04-20 | 昆明理工大学 | Expressway illegal parking detection method based on kernel density estimation |
CN106022243A (en) * | 2016-05-13 | 2016-10-12 | 浙江大学 | Method for recognizing converse vehicle driving in vehicle lanes on the basis of image processing |
CN107067734A (en) * | 2017-04-11 | 2017-08-18 | 山东大学 | A kind of urban signal controlling intersection vehicles are detained peccancy detection method |
CN107292239A (en) * | 2017-05-24 | 2017-10-24 | 南京邮电大学 | A kind of parking offense detection method based on triple context updates |
CN107293116A (en) * | 2017-06-27 | 2017-10-24 | 浙江浩腾电子科技股份有限公司 | A kind of traffic incident detecting system based on video analysis |
CN107578489A (en) * | 2016-07-05 | 2018-01-12 | 上海小享网络科技有限公司 | A kind of drive recorder for motor vehicles |
CN107844772A (en) * | 2017-11-09 | 2018-03-27 | 汕头职业技术学院 | A kind of motor vehicle automatic testing method based on movable object tracking |
CN109285341A (en) * | 2018-10-31 | 2019-01-29 | 中电科新型智慧城市研究院有限公司 | A kind of urban road vehicle exception stagnation of movement detection method based on real-time video |
CN109544907A (en) * | 2018-10-23 | 2019-03-29 | 大唐软件技术股份有限公司 | A kind of vehicle count method, device |
CN109636832A (en) * | 2018-12-14 | 2019-04-16 | 清华大学苏州汽车研究院(吴江) | Stop detection method, device, electronic equipment and storage medium |
CN109949579A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of illegal automatic auditing method that makes a dash across the red light based on deep learning |
CN110008932A (en) * | 2019-04-17 | 2019-07-12 | 四川九洲视讯科技有限责任公司 | A kind of vehicle violation crimping detection method based on computer vision |
CN110633609A (en) * | 2019-04-29 | 2019-12-31 | 韩明 | Positioning platform based on face object position analysis |
CN110660225A (en) * | 2019-10-28 | 2020-01-07 | 上海眼控科技股份有限公司 | Red light running behavior detection method, device and equipment |
CN110853342A (en) * | 2018-07-24 | 2020-02-28 | 东旭科技集团有限公司 | Method and device for determining driving state, street lamp and readable storage medium |
CN110853094A (en) * | 2019-10-11 | 2020-02-28 | 北京文香信息技术有限公司 | Method, device and equipment for determining threshold of target position and storage medium |
CN111383458A (en) * | 2018-12-30 | 2020-07-07 | 浙江宇视科技有限公司 | Vehicle violation detection method, device, equipment and storage medium |
CN111968378A (en) * | 2020-07-07 | 2020-11-20 | 浙江大华技术股份有限公司 | Motor vehicle red light running snapshot method and device, computer equipment and storage medium |
CN112507993A (en) * | 2021-02-05 | 2021-03-16 | 上海闪马智能科技有限公司 | Abnormal driving-away detection method and system, electronic equipment and storage medium |
CN113298060A (en) * | 2021-07-27 | 2021-08-24 | 支付宝(杭州)信息技术有限公司 | Privacy-protecting biometric feature recognition method and device |
CN113591683A (en) * | 2021-07-28 | 2021-11-02 | 北京百度网讯科技有限公司 | Attitude estimation method, attitude estimation device, electronic equipment and storage medium |
CN114419106A (en) * | 2022-03-30 | 2022-04-29 | 深圳市海清视讯科技有限公司 | Vehicle violation detection method, device and storage medium |
-
2013
- 2013-06-21 CN CN2013102512067A patent/CN103345618A/en active Pending
Non-Patent Citations (3)
Title |
---|
何巧萍: "基于视频图像处理技术的运动车辆检测算法研究及实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
佟守愚: "基于视频技术的交通违章检测与识别理论及方法研究", 《中国优秀博硕士学位论文全文数据库(博士)工程科技Ⅱ辑》 * |
杨聪俐: "基于视频流的车辆检测和跟踪算法的研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (31)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103955705B (en) * | 2014-04-29 | 2017-11-28 | 银江股份有限公司 | Traffic lights positioning, identification and sorting technique based on video analysis |
CN103955705A (en) * | 2014-04-29 | 2014-07-30 | 银江股份有限公司 | Traffic signal lamp positioning, recognizing and classifying method based on video analysis |
CN104809887A (en) * | 2015-04-23 | 2015-07-29 | 杭州中威电子股份有限公司 | Detecting method for vehicle converse running on expressway and automatic alarm device |
CN105513371A (en) * | 2016-01-15 | 2016-04-20 | 昆明理工大学 | Expressway illegal parking detection method based on kernel density estimation |
CN106022243A (en) * | 2016-05-13 | 2016-10-12 | 浙江大学 | Method for recognizing converse vehicle driving in vehicle lanes on the basis of image processing |
CN106022243B (en) * | 2016-05-13 | 2019-02-26 | 浙江大学 | A kind of retrograde recognition methods of the car lane vehicle based on image procossing |
CN107578489A (en) * | 2016-07-05 | 2018-01-12 | 上海小享网络科技有限公司 | A kind of drive recorder for motor vehicles |
CN107067734A (en) * | 2017-04-11 | 2017-08-18 | 山东大学 | A kind of urban signal controlling intersection vehicles are detained peccancy detection method |
CN107292239A (en) * | 2017-05-24 | 2017-10-24 | 南京邮电大学 | A kind of parking offense detection method based on triple context updates |
CN107293116A (en) * | 2017-06-27 | 2017-10-24 | 浙江浩腾电子科技股份有限公司 | A kind of traffic incident detecting system based on video analysis |
CN107844772A (en) * | 2017-11-09 | 2018-03-27 | 汕头职业技术学院 | A kind of motor vehicle automatic testing method based on movable object tracking |
CN110853342A (en) * | 2018-07-24 | 2020-02-28 | 东旭科技集团有限公司 | Method and device for determining driving state, street lamp and readable storage medium |
CN109544907A (en) * | 2018-10-23 | 2019-03-29 | 大唐软件技术股份有限公司 | A kind of vehicle count method, device |
CN109285341A (en) * | 2018-10-31 | 2019-01-29 | 中电科新型智慧城市研究院有限公司 | A kind of urban road vehicle exception stagnation of movement detection method based on real-time video |
CN109285341B (en) * | 2018-10-31 | 2021-08-31 | 中电科新型智慧城市研究院有限公司 | Urban road vehicle abnormal stop detection method based on real-time video |
CN109636832A (en) * | 2018-12-14 | 2019-04-16 | 清华大学苏州汽车研究院(吴江) | Stop detection method, device, electronic equipment and storage medium |
CN111383458B (en) * | 2018-12-30 | 2021-07-27 | 浙江宇视科技有限公司 | Vehicle violation detection method, device, equipment and storage medium |
CN111383458A (en) * | 2018-12-30 | 2020-07-07 | 浙江宇视科技有限公司 | Vehicle violation detection method, device, equipment and storage medium |
CN109949579A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of illegal automatic auditing method that makes a dash across the red light based on deep learning |
CN110008932A (en) * | 2019-04-17 | 2019-07-12 | 四川九洲视讯科技有限责任公司 | A kind of vehicle violation crimping detection method based on computer vision |
CN110008932B (en) * | 2019-04-17 | 2022-11-22 | 四川九洲视讯科技有限责任公司 | Vehicle violation line-pressing detection method based on computer vision |
CN110633609A (en) * | 2019-04-29 | 2019-12-31 | 韩明 | Positioning platform based on face object position analysis |
CN110853094A (en) * | 2019-10-11 | 2020-02-28 | 北京文香信息技术有限公司 | Method, device and equipment for determining threshold of target position and storage medium |
CN110853094B (en) * | 2019-10-11 | 2020-12-29 | 北京文香信息技术有限公司 | Method, device and equipment for determining threshold of target position and storage medium |
CN110660225A (en) * | 2019-10-28 | 2020-01-07 | 上海眼控科技股份有限公司 | Red light running behavior detection method, device and equipment |
CN111968378A (en) * | 2020-07-07 | 2020-11-20 | 浙江大华技术股份有限公司 | Motor vehicle red light running snapshot method and device, computer equipment and storage medium |
CN112507993A (en) * | 2021-02-05 | 2021-03-16 | 上海闪马智能科技有限公司 | Abnormal driving-away detection method and system, electronic equipment and storage medium |
CN113298060A (en) * | 2021-07-27 | 2021-08-24 | 支付宝(杭州)信息技术有限公司 | Privacy-protecting biometric feature recognition method and device |
CN113591683A (en) * | 2021-07-28 | 2021-11-02 | 北京百度网讯科技有限公司 | Attitude estimation method, attitude estimation device, electronic equipment and storage medium |
CN114419106A (en) * | 2022-03-30 | 2022-04-29 | 深圳市海清视讯科技有限公司 | Vehicle violation detection method, device and storage medium |
CN114419106B (en) * | 2022-03-30 | 2022-07-22 | 深圳市海清视讯科技有限公司 | Vehicle violation detection method, device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103345618A (en) | Traffic violation detection method based on video technology | |
CN110443225B (en) | Virtual and real lane line identification method and device based on feature pixel statistics | |
CN205992300U (en) | For providing the electronic installation of the guidance information using crossing recognition result | |
US9292750B2 (en) | Method and apparatus for detecting traffic monitoring video | |
CN104008645B (en) | One is applicable to the prediction of urban road lane line and method for early warning | |
CN103324930B (en) | A kind of registration number character dividing method based on grey level histogram binaryzation | |
CN103522970B (en) | Based on vehicle driving safety method of inspection and the system of machine vision | |
CN101739829B (en) | Video-based vehicle overspeed monitoring method and system | |
CN106647776B (en) | Method and device for judging lane changing trend of vehicle and computer storage medium | |
CN104129389A (en) | Method for effectively judging and recognizing vehicle travelling conditions and device thereof | |
CN105426864A (en) | Multiple lane line detecting method based on isometric peripheral point matching | |
CN104866823A (en) | Vehicle detection and tracking method based on monocular vision | |
CN111563412B (en) | Rapid lane line detection method based on parameter space voting and Bessel fitting | |
CN104318258A (en) | Time domain fuzzy and kalman filter-based lane detection method | |
CN105844222A (en) | System and method for front vehicle collision early warning based on visual sense | |
CN109948552B (en) | Method for detecting lane line in complex traffic environment | |
CN102419820A (en) | Method for rapidly detecting car logo in videos and images | |
CN104237881A (en) | FMCW anti-collision radar multi-target detecting and tracking system and method | |
CN202134079U (en) | Unmanned vehicle lane marker line identification and alarm device | |
CN104077756A (en) | Direction filtering method based on lane line confidence | |
CN103488975A (en) | Zebra crossing real-time detection method based in intelligent driving | |
CN105260701A (en) | Front vehicle detection method applied to complex scene | |
CN105426868A (en) | Lane detection method based on adaptive region of interest | |
CN110263714B (en) | Lane line detection method, lane line detection device, electronic device, and storage medium | |
CN110379168A (en) | A kind of vehicular traffic information acquisition method based on Mask R-CNN |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20131009 |