CN105160326A - Automatic highway parking detection method and device - Google Patents

Automatic highway parking detection method and device Download PDF

Info

Publication number
CN105160326A
CN105160326A CN201510585772.0A CN201510585772A CN105160326A CN 105160326 A CN105160326 A CN 105160326A CN 201510585772 A CN201510585772 A CN 201510585772A CN 105160326 A CN105160326 A CN 105160326A
Authority
CN
China
Prior art keywords
target
module
highway
parking
vehicle
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
Application number
CN201510585772.0A
Other languages
Chinese (zh)
Inventor
石旭刚
张水发
刘嘉
杜雅慧
汤泽胜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
OB TELECOM ELECTRONICS CO Ltd
Original Assignee
OB TELECOM ELECTRONICS CO Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by OB TELECOM ELECTRONICS CO Ltd filed Critical OB TELECOM ELECTRONICS CO Ltd
Priority to CN201510585772.0A priority Critical patent/CN105160326A/en
Publication of CN105160326A publication Critical patent/CN105160326A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30236—Traffic on road, railway or crossing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30241—Trajectory
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/08—Detecting or categorising vehicles

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)

Abstract

The invention provides an automatic highway parking detection method which comprises the following steps: 1) range setting; 2) target detection; 3) target tracking; 4) trajectory analysis; and 5) monitoring scene background update. By analyzing the track of a vehicle target, three kinds of events, which are abnormal parking event, abnormal parking relieving event and normal driving event, are obtained through classification, and thus behaviors of a vehicle can be tracked and analyzed better. Through trajectory analysis of the vehicle target, different background model update strategies and update rates are adopted for happening regions of different kinds of events, thereby improving accuracy of target detection, improving accuracy of abnormal events and reducing false alarm rate.

Description

The automatic testing method that a kind of highway stops and device
Technical field
The present invention relates to monitoring technique field, particularly relate to a kind of highway stops automatic testing method and device.
Background technology
Along with the raising of socioeconomic development, living standards of the people, urban automobile quantity increases fast, and in sustainable growth.The infrastructure such as highway, urban road is also in quick rising passway accordingly, current China highway is more than 100,000 kilometers, but due to the backwardness of administrative skill and supervision method, traffic hazard is taken place frequently, wherein, stopping, driving in the wrong direction and shed the reasons such as thing is the important illegal activities causing accident, once there is these behaviors, and danger signal cannot be transmitted to rear car fast and effectively, reduce the speed of a motor vehicle, easily cause serious traffic hazard, threaten the people's lives and property safety.Therefore, the illegal activities such as the parking on highway how to be detected fast and effectively, drive in the wrong direction, remind in time periphery driving vehicle, take effective risk avoidance measures, have important meaning to what reduce expressway traffic accident rate and prevent second accident.
Parking method for detecting abnormality based on image technique mainly comprises two classes: the method that method and based target based on sectional pattern are followed the tracks of.Method based on sectional pattern detects whether occur parking by the change detecting block image region, exception parking event can be detected fast and accurately in the good scene of light, but highway mileage is long, within 24 hours, do not stop monitoring, faced by weather condition complicated and changeable, cause the method based on sectional pattern to stablize and effective detect exception parking event.
Summary of the invention
The technical matters that first the present invention solves is to provide the automatic testing method that a kind of highway stops, and can solve Problems existing in background technology, and the parking be particularly useful on highway detects.
The present invention solves the problems of the technologies described above adopted technical scheme: the automatic testing method that a kind of highway stops, comprises the following steps:
Step 1: range set, arranges the surveyed area that no parking and parking time of fire alarming threshold values in the freeway surveillance and control scene of specifying;
Step 2: target detection, detects the foreground target in the monitoring scene of specifying;
Step 3: target following, the track of the foreground target detected in obtaining step 2;
Step 4: trajectory analysis, analyzes the track of the foreground target obtained in step 3, and judges according to the state of relation to foreground target of scope set in track and step 1;
Step 5: the background in monitoring scene is upgraded according to the different judged results in step 4.
Adopt said method, make background adapt to scene changes, reduce the interference because illumination gradual change, noise produce target detection.
While employing technique scheme, the present invention can also adopt or combine and adopt following further technical scheme:
Described step 2 specifically comprises the following steps:
Step 2.1: set up Gaussian Background model;
Step 2.2: current frame image is compared with background image according to formula (1), what meet background model distribution then thinks background, otherwise is prospect, thus by current frame image binaryzation;
Wherein, Pt represents the attribute of this point, is namely prospect or background, f tfor grey scale pixel value in present frame, μ, δ are respectively average and the standard deviation of gray-scale value;
Step 2.3, extract the connected region of foreground image, and filtering area is less than the target of certain threshold value.
Described step 3 specifically comprises the following steps:
Step 3.1: extract i-th car (i=1,2 ..., N) center-of-mass coordinate (x gi, y gi) and area A i, according to the center-of-mass coordinate of front cross frame, the possible position of prediction next frame vehicle centroid:
A i = Σ ( x , y ) ∈ R i 1
( x g i , y g i ) = 1 A ( Σ ( x , y ) ∈ R i x , Σ ( x , y ) ∈ R i y )
(x gi,y gi) t+1＝2*(x gi,y gi) t-(x gi,y gi) t-1
Wherein R ifor the foreground area that vehicle i is corresponding, t represents the time;
Step 3.2: in the centroid position certain limit of prediction, calculate present frame vehicle and previous frame vehicle histogram similarity, when histogram similarity is greater than certain threshold value, select the most similar conduct coupling, using the center-of-mass coordinate of coupling vehicle as the position of target following at present frame track, otherwise not think to there is coupling;
Foreground image is the image of difference and background in bianry image, and foreground target is the target formed after denoising, regional connectivity by foreground image, here vehicle and foreground target.
The track of final acquisition is the track of vehicle centroid.
In described step 4, the judgement of foreground target track is divided three classes:
The first kind: target vehicle moves to no-parking zone, and the residence time be greater than time threshold, be judged as exception parking event;
Equations of The Second Kind: the vehicle being judged as exception parking, its speed increases gradually by 0, and sails out of no-parking zone within a certain period of time, is judged as and removes exception parking event;
3rd class: other normally travel event.
In described step 5, when being judged as the first kind, reduce the context update rate that exception parking region occurs, with the study preventing exception parking target very fast in background, all the other regions upgrade background model by normal renewal speed;
When being judged as Equations of The Second Kind, illustrate that target is recovered from exception parking event, at this moment often there is diplopia in exception parking region, therefore, increase the context update rate in exception parking region, to recover from exception parking event, learnt by present image in background model, all the other regions upgrade background model by normal renewal speed;
When being judged as the 3rd class, full figure all upgrades background model by normal renewal speed.
Another technical matters to be solved by this invention is to provide a kind of automatic detection device stopped on highway, this automatic detection device is used for above-mentioned detection method, and comprise image capture module, rule arranges module, module of target detection, target tracking module, trajectory analysis module and model modification module, described image capture module is for gathering image on highway in monitoring scene and change thereof, described rule arranges module for setting detected rule, described module of target detection is used for the foreground target in test and monitoring scene, described target tracking module is used for carrying out track following to the foreground target detected, described trajectory analysis module is used for analyzing the track of foreground target and judges to sort out, described model modification module upgrades existing monitoring scene respectively according to the judgement classification of trajectory analysis module, described image capture module is located at position highway having and widens the vision, described image capture module is connected to described module of target detection, described module of target detection is connected to described target tracking module, described rule arranges model calling to described trajectory analysis module, and described trajectory analysis model calling is to described model modification module.
The invention has the beneficial effects as follows: a kind of highway that the present invention proposes stops automatic testing method and device are integrated solutions, specifically there is following innovation: 1, by analyzing the track of vehicle target, be divided three classes: exception parking event, remove exception parking event and normally travel event, can be followed the trail of and analyze the behavior of vehicle preferably.2, according to the trajectory analysis of vehicle target, for the generation area of dissimilar event, adopt different background model update strategies and turnover rate, improve the accuracy of target detection, thus improve the accuracy rate of anomalous event, reduce wrong report.
Accompanying drawing explanation
Fig. 1 is the automatic testing method schematic flow sheet that a kind of highway of the present invention stops.
Fig. 2 is the automatic detection device structural representation that a kind of highway of the present invention stops.
Embodiment
Embodiment 1, the automatic testing method that highway stops.
With reference to accompanying drawing 1.
Automatic testing method of the present invention as shown in Figure 1, specifically comprises the following steps:
Step 1:
In the freeway surveillance and control scene of specifying, the surveyed area that no parking and parking time of fire alarming threshold value are set; Multiple closed surveyed area can be set, also different parking time of fire alarming threshold values can be set different surveyed areas;
Step 2:
Target detection, detects the foreground target in scene;
First, Gaussian Background model is set up;
Secondly, compared with background image by current frame image, what meet background model distribution then thinks background, otherwise is prospect, thus by current frame image binaryzation;
Wherein, f tfor grey scale pixel value in present frame, μ, δ are respectively average and standard deviation;
Finally, extract the connected region of foreground image, and the target that filtering area is less than 50.
Step 3:
Target following, obtains target trajectory;
First, extraction vehicle i (i=1,2 ..., N) center-of-mass coordinate (x gi, y gi), area A i, according to the center-of-mass coordinate of front cross frame, the possible position of prediction next frame vehicle centroid:
A i = Σ ( x , y ) ∈ R i 1
( x g i , y g i ) = 1 A ( Σ ( x , y ) ∈ R i x , Σ ( x , y ) ∈ R i y )
(x gi,y gi) t+1＝2*(x gi,y gi) t-(x gi,y gi) t-1
Wherein R ifor the foreground area that vehicle i is corresponding.
Secondly, according to the speed limit of highway and the interval of events between frame and frame, within the scope of the centroid position 10px of prediction, find all vehicles within the scope of this, calculate and previous frame vehicle histogram to be matched similarity, weigh by Pasteur's distance, when distance between histogram is less than 0.2, the conduct coupling that chosen distance is minimum, using the center-of-mass coordinate of coupling vehicle as the position of target following at present frame track, otherwise not thinks to there is coupling;
Step 4:
The track of evaluating objects, is divided three classes.The first kind: target vehicle moves to no-parking zone, and the residence time be greater than 2s, be judged as exception parking event; Equations of The Second Kind: the vehicle being judged as exception parking, its speed increases gradually by 0, and sails out of no-parking zone in 2s, is judged as and removes exception parking event; 3rd class: other normally travel event;
Step 5:
When being judged to exception parking event occurs, reducing and the context update rate in exception parking region occur, with prevent exception parking target very fast acquire in background, cause detecting exception parking event, all the other regions upgrade background model by normal renewal speed;
The pixel in exception parking region, if do not mate background model, does not then upgrade; If coupling background model, then as shown in the formula Renewal model parameter:
μ t＝(1-ρ 1)*μ t-1+ρ 1*f t
δ t 2＝(1-ρ 1)*δ t-1 2+ρ 1*(f t-μ t) T(f t-μ t)
Wherein ρ 1for turnover rate, ft is grey scale pixel value, and μ t and δ t represents average and the standard deviation of t grey scale pixel value respectively, and μ t-1 and δ t-1 represents average and the standard deviation of t-1 moment grey scale pixel value respectively, T representing matrix;
When being judged to remove exception parking event, illustrate that target is recovered from exception parking event, at this moment often there is diplopia in exception parking region, therefore, increase the context update rate in exception parking region, to recover from exception parking event, learnt by present image in background model, all the other regions upgrade background model by normal renewal speed;
Remove the pixel in exception parking region, no matter whether mate background model, all force to upgrade by following formula:
μ t＝(1-ρ 2)*μ t-1+ρ 2*f t
δ t 2＝(1-ρ 2)*δ t-1 2+ρ 2*(f t-μ t) T(f t-μ t)
Wherein ρ 2for turnover rate;
When being judged to be the 3rd class, no matter whether mate background model, all force to upgrade by following formula:
μ t＝(1-ρ 3)*μ t-1+ρ 3*f t
δ t 2＝(1-ρ 3)*δ t-1 2+ρ 3*(f t-μ t) T(f t-μ t)
Wherein ρ 3for turnover rate;
Wherein ρ 1=0.0005, ρ 2=0.005, ρ 3=0.001.Turnover rate rule of thumb obtains.δ t is gray standard deviation, μ t gray average, ft are current pixel point gray-scale value, t represents the time.
Embodiment 2, the automatic detection device that highway stops.
With reference to accompanying drawing 2.
Automatic detection device of the present invention is used for above-mentioned detection method, specifically comprise image capture module 1, rule arranges module 2, module of target detection 3, target tracking module 4, trajectory analysis module 5 and model modification module 6, described image capture module 1 is for gathering image on highway in monitoring scene and change thereof, described rule arranges in the step 1 of module 2 for embodiment 1 and sets detected rule, described module of target detection 3 is for the foreground target in test and monitoring scene in the step 2 of embodiment 1, described target tracking module 4 is for carrying out track following to the foreground target detected in the step 3 of embodiment 1, described trajectory analysis module 5 is for analyzing the track of foreground target in the step 4 to embodiment 1 and judging to sort out, described model modification module 6 upgrades existing monitoring scene respectively for the judgement classification according to trajectory analysis module in the step 5 of embodiment 1.
Described image capture module 1 is located at position highway having and widens the vision, described image capture module 1 is connected to described module of target detection 2 and sends to it image information gathered, described module of target detection 2 is connected to described target tracking module 3 and sends to it foreground target information detected, described rule arranges module 2 and is connected to described trajectory analysis module 5 to the surveyed area set by its transmission and time threshold information, and described trajectory analysis module 5 is connected to described model modification module 6 and sends trajectory analysis result to it.

Claims (6)

1. automatic testing method highway stopped, is characterized in that: described automatic testing method comprises the following steps:
Step 1: range set, arranges the surveyed area that no parking and parking time of fire alarming threshold values in the freeway surveillance and control scene of specifying;
Step 2: target detection, detects the foreground target in the monitoring scene of specifying;
Step 3: target following, the track of the foreground target detected in obtaining step 2;
Step 4: trajectory analysis, analyzes the track of the foreground target obtained in step 3, and judges according to the state of relation to foreground target of scope set in track and step 1;
Step 5: the background in monitoring scene is upgraded according to the different judged results in step 4.
2. the automatic testing method a kind of highway as claimed in claim 1 stopped, is characterized in that: described step 2 specifically comprises the following steps:
Step 2.1: set up Gaussian Background model;
Step 2.2: current frame image is compared with background image according to formula (1), what meet background model distribution then thinks background, otherwise is prospect, thus by current frame image binaryzation;
Wherein, f tfor grey scale pixel value in present frame, μ, δ are respectively average and standard deviation; Pt represents the attribute of this point, and μ, δ are average and the standard deviation of grey scale pixel value respectively;
Step 2.3, extract the connected region of foreground image, and filtering area is less than the target of certain threshold value.
3. the automatic testing method a kind of highway as claimed in claim 1 stopped, is characterized in that: described step 3 specifically comprises the following steps:
Step 3.1: extract i-th car (i=1,2 ..., N) center-of-mass coordinate (x gi, y gi) and area A i, according to the center-of-mass coordinate of front cross frame, the possible position of prediction next frame vehicle centroid:
A i = Σ ( x , y ) ∈ R i 1
( x g i , y g i ) = 1 A ( Σ ( x , y ) ∈ R i x , Σ ( x , y ) ∈ R i y )
(x gi,y gi) t+1＝2*(x gi,y gi) t-(x gi,y gi) t-1
Wherein R ifor the foreground area that vehicle i is corresponding, t is the time;
Step 3.2: in the centroid position certain limit of prediction, calculate present frame vehicle and previous frame vehicle histogram similarity, when histogram similarity is greater than certain threshold value, select the most similar conduct coupling, using the center-of-mass coordinate of coupling vehicle as the position of target following at present frame track, otherwise not think to there is coupling.
4. the automatic testing method a kind of highway as claimed in claim 1 stopped, is characterized in that: be divided three classes to the judgement of foreground target track in described step 4:
The first kind: target vehicle moves to no-parking zone, and the residence time be greater than time threshold, be judged as exception parking event;
Equations of The Second Kind: the vehicle being judged as exception parking, its speed increases gradually by 0, and sails out of no-parking zone within a certain period of time, is judged as and removes exception parking event;
3rd class: other normally travel event.
5. the automatic testing method a kind of highway as claimed in claim 4 stopped, it is characterized in that: in described step 5, when being judged as the first kind, reduce the context update rate that exception parking region occurs, with the study preventing exception parking target very fast in background, all the other regions upgrade background model by normal renewal speed;
When being judged as Equations of The Second Kind, illustrate that target is recovered from exception parking event, at this moment often there is diplopia in exception parking region, therefore, increase the context update rate in exception parking region, to recover from exception parking event, learnt by present image in background model, all the other regions upgrade background model by normal renewal speed;
When being judged as the 3rd class, full figure all upgrades background model by normal renewal speed.
6. an automatic detection device highway stopped, it is characterized in that: described automatic detection device comprises image capture module, rule arranges module, module of target detection, target tracking module, trajectory analysis module and model modification module, described image capture module is for gathering image on highway in monitoring scene and change thereof, described rule arranges module for setting detected rule, described module of target detection is used for the foreground target in test and monitoring scene, described target tracking module is used for carrying out track following to the foreground target detected, described trajectory analysis module is used for analyzing the track of foreground target and judges to sort out, described model modification module upgrades existing monitoring scene respectively according to the judgement classification of trajectory analysis module, described image capture module is located at position highway having and widens the vision, described image capture module is connected to described module of target detection, described module of target detection is connected to described target tracking module, described rule arranges model calling to described trajectory analysis module, and described trajectory analysis model calling is to described model modification module.
CN201510585772.0A 2015-09-15 2015-09-15 Automatic highway parking detection method and device Pending CN105160326A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510585772.0A CN105160326A (en) 2015-09-15 2015-09-15 Automatic highway parking detection method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510585772.0A CN105160326A (en) 2015-09-15 2015-09-15 Automatic highway parking detection method and device

Publications (1)

Publication Number Publication Date
CN105160326A true CN105160326A (en) 2015-12-16

Family

ID=54801178

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510585772.0A Pending CN105160326A (en) 2015-09-15 2015-09-15 Automatic highway parking detection method and device

Country Status (1)

Country Link
CN (1) CN105160326A (en)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105513371A (en) * 2016-01-15 2016-04-20 昆明理工大学 Expressway illegal parking detection method based on kernel density estimation
CN106652465A (en) * 2016-11-15 2017-05-10 成都通甲优博科技有限责任公司 Method and system for identifying abnormal driving behavior on road
CN107025044A (en) * 2017-03-30 2017-08-08 宇龙计算机通信科技(深圳)有限公司 A kind of timing method and its equipment
CN107886726A (en) * 2017-07-06 2018-04-06 杭州盛棠信息科技有限公司 Road occupying/parking behavior detection method and device
CN108091142A (en) * 2017-12-12 2018-05-29 公安部交通管理科学研究所 For vehicle illegal activities Tracking Recognition under highway large scene and the method captured automatically
CN108230687A (en) * 2017-12-18 2018-06-29 北京工业大学 Become more meticulous vehicle behavior management method and the system in a kind of traffic conflict region
CN108280815A (en) * 2018-02-26 2018-07-13 安徽新闻出版职业技术学院 A kind of geometric correction method towards monitoring scene structure
CN110187334A (en) * 2019-05-28 2019-08-30 深圳大学 A kind of target monitoring method, apparatus and computer readable storage medium
CN110335467A (en) * 2019-07-24 2019-10-15 山东交通学院 A method of vehicle on highway behavioral value is realized using computer vision
CN112633228A (en) * 2020-12-31 2021-04-09 北京市商汤科技开发有限公司 Parking detection method, device, equipment and storage medium
CN112687125A (en) * 2020-12-09 2021-04-20 北京中交兴路车联网科技有限公司 Early warning method and device after illegal parking on road, storage medium and terminal
CN113409587A (en) * 2021-06-16 2021-09-17 北京字跳网络技术有限公司 Abnormal vehicle detection method, device, equipment and storage medium
CN114530043A (en) * 2022-03-03 2022-05-24 上海闪马智能科技有限公司 Event detection method and device, storage medium and electronic device

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102568206A (en) * 2012-01-13 2012-07-11 大连民族学院 Video monitoring-based method for detecting cars parking against regulations
CN103136514A (en) * 2013-02-05 2013-06-05 长安大学 Parking event detecting method based on double tracking system
CN103617410A (en) * 2013-08-30 2014-03-05 重庆大学 Highway tunnel parking detection method based on video detection technology

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102568206A (en) * 2012-01-13 2012-07-11 大连民族学院 Video monitoring-based method for detecting cars parking against regulations
CN103136514A (en) * 2013-02-05 2013-06-05 长安大学 Parking event detecting method based on double tracking system
CN103617410A (en) * 2013-08-30 2014-03-05 重庆大学 Highway tunnel parking detection method based on video detection technology

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105513371A (en) * 2016-01-15 2016-04-20 昆明理工大学 Expressway illegal parking detection method based on kernel density estimation
CN106652465A (en) * 2016-11-15 2017-05-10 成都通甲优博科技有限责任公司 Method and system for identifying abnormal driving behavior on road
CN107025044B (en) * 2017-03-30 2021-02-23 宇龙计算机通信科技(深圳)有限公司 Timing method and device
CN107025044A (en) * 2017-03-30 2017-08-08 宇龙计算机通信科技(深圳)有限公司 A kind of timing method and its equipment
CN107886726A (en) * 2017-07-06 2018-04-06 杭州盛棠信息科技有限公司 Road occupying/parking behavior detection method and device
CN107886726B (en) * 2017-07-06 2024-02-23 杭州盛棠信息科技有限公司 Method and device for detecting road occupation/parking behavior
CN108091142A (en) * 2017-12-12 2018-05-29 公安部交通管理科学研究所 For vehicle illegal activities Tracking Recognition under highway large scene and the method captured automatically
CN108230687A (en) * 2017-12-18 2018-06-29 北京工业大学 Become more meticulous vehicle behavior management method and the system in a kind of traffic conflict region
CN108230687B (en) * 2017-12-18 2021-04-13 北京工业大学 Refined vehicle behavior management method and system for traffic conflict area
CN108280815A (en) * 2018-02-26 2018-07-13 安徽新闻出版职业技术学院 A kind of geometric correction method towards monitoring scene structure
CN108280815B (en) * 2018-02-26 2021-10-22 安徽新闻出版职业技术学院 Geometric correction method for monitoring scene structure
CN110187334A (en) * 2019-05-28 2019-08-30 深圳大学 A kind of target monitoring method, apparatus and computer readable storage medium
CN110187334B (en) * 2019-05-28 2021-06-08 深圳大学 Target monitoring method and device and computer readable storage medium
CN110335467A (en) * 2019-07-24 2019-10-15 山东交通学院 A method of vehicle on highway behavioral value is realized using computer vision
CN110335467B (en) * 2019-07-24 2021-08-27 山东交通学院 Method for realizing highway vehicle behavior detection by using computer vision
CN112687125A (en) * 2020-12-09 2021-04-20 北京中交兴路车联网科技有限公司 Early warning method and device after illegal parking on road, storage medium and terminal
CN112633228A (en) * 2020-12-31 2021-04-09 北京市商汤科技开发有限公司 Parking detection method, device, equipment and storage medium
CN113409587B (en) * 2021-06-16 2022-11-22 北京字跳网络技术有限公司 Abnormal vehicle detection method, device, equipment and storage medium
CN113409587A (en) * 2021-06-16 2021-09-17 北京字跳网络技术有限公司 Abnormal vehicle detection method, device, equipment and storage medium
CN114530043A (en) * 2022-03-03 2022-05-24 上海闪马智能科技有限公司 Event detection method and device, storage medium and electronic device

Similar Documents

Publication Publication Date Title
CN105160326A (en) Automatic highway parking detection method and device
CN100440269C (en) Intelligent detecting prewarning method for expressway automobile running and prewaring system thereof
CN105744232B (en) A kind of method of the transmission line of electricity video external force damage prevention of Behavior-based control analytical technology
KR101095528B1 (en) An outomatic sensing system for traffic accident and method thereof
CN101984478B (en) Abnormal S-type driving warning method based on binocular vision lane marking detection
Zhang et al. Pedestrian safety analysis in mixed traffic conditions using video data
CN110766982B (en) Vehicle collision detection system based on vehicle-mounted sensor
CN102164270A (en) Intelligent video monitoring method and system capable of exploring abnormal events
CN102568206B (en) Video monitoring-based method for detecting cars parking against regulations
CN103770733A (en) Method and device for detecting safety driving states of driver
CN102717765A (en) Fatigue driving detection method and anti-fatigue driving auxiliary device
CN104021370A (en) Driver state monitoring method based on vision information fusion and driver state monitoring system based on vision information fusion
CN110335467B (en) Method for realizing highway vehicle behavior detection by using computer vision
CN105513349A (en) Double-perspective learning-based mountainous area highway vehicle event detection method
CN102897086A (en) Back car driving information detecting and prompting method and system
CN106254827A (en) A kind of group's mist Intelligent Recognition method for early warning and device thereof
CN105574502A (en) Automatic detection method for violation behaviors of self-service card sender
CN109615880A (en) A kind of wagon flow measuring method based on radar image processing
CN103152558B (en) Based on the intrusion detection method of scene Recognition
CN106428003A (en) Lane departure forewarning device and method for vehicle on highway under adverse weather
CN115782905A (en) Automatic driving vehicle driving safety degree quantification system
CN113781471B (en) Automatic driving test field system and method
CN113706901B (en) Intelligent accident prevention and control and early warning system for entrance section of expressway tunnel
CN113850995B (en) Event detection method, device and system based on tunnel radar vision data fusion
CN116001800B (en) Vehicle driving risk information acquisition method and device, electronic equipment and medium

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: 20151216