CN107705577A - A kind of real-time detection method and system based on lane line demarcation vehicle peccancy lane change - Google Patents

A kind of real-time detection method and system based on lane line demarcation vehicle peccancy lane change Download PDF

Info

Publication number
CN107705577A
CN107705577A CN201711026761.4A CN201711026761A CN107705577A CN 107705577 A CN107705577 A CN 107705577A CN 201711026761 A CN201711026761 A CN 201711026761A CN 107705577 A CN107705577 A CN 107705577A
Authority
CN
China
Prior art keywords
vehicle
tracking
lane line
cos
rectangle
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201711026761.4A
Other languages
Chinese (zh)
Other versions
CN107705577B (en
Inventor
李松斌
杨洁
赵思奇
刘鹏
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.)
Nanhai Research Station Institute Of Acoustics Chinese Academy Of Sciences
Institute of Acoustics CAS
Original Assignee
Institute of Acoustics CAS
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 Institute of Acoustics CAS filed Critical Institute of Acoustics CAS
Priority to CN201711026761.4A priority Critical patent/CN107705577B/en
Publication of CN107705577A publication Critical patent/CN107705577A/en
Application granted granted Critical
Publication of CN107705577B publication Critical patent/CN107705577B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
    • G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • 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
    • G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
    • G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
    • 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)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • Traffic Control Systems (AREA)
  • Image Analysis (AREA)

Abstract

The present invention relates to a kind of real-time detection method based on lane line demarcation vehicle peccancy lane change, specifically include：Step 1) demarcates lane line L, calculates and obtains detection rectangular area D and image I；Step 2) detects image I vehicle set A={ A based on deep layer convolutional neural networks in the detection rectangular area D of step 1)1,A2,...,Ai}；The vehicle set A that step 3) obtains according to step 2), filter out the vehicle set B={ B intersected with lane line L1,B2,...,Bi, wherein,Matched again with tracking vehicle list TL, and update tracking vehicle list TL={ T1,T2,...,Tj}；Step 4) judges the tracking vehicle T in the tracking vehicle list TL after renewaljWhether it is lane change vehicle violating the regulations；If track vehicle TjFor lane change vehicle violating the regulations, then tracking vehicle T is marked outjPosition, and lane change vehicle T violating the regulations is deleted from the tracking vehicle list TL after renewaljInformation.

Description

A kind of real-time detection method and system based on lane line demarcation vehicle peccancy lane change
Technical field
The invention belongs to intelligent transportation system and the technical field of image recognition, and in particular to one kind is demarcated based on lane line The real-time detection method and system of vehicle peccancy lane change.
Background technology
With the progress and development of society, urban automobile is growing day by day, and vehicles number is constantly soaring to cause traffic thing Therefore it is more serious, economic loss caused by road traffic accident and death toll are also constantly rising in recent years.Control traffic thing Therefore the problem of increasingly paying attention to as vehicle supervision department occurs, and it is exactly automobile illegal row to cause the first cause of traffic accident For.As a kind of common traffic violation, lane change violating the regulations not only results in traffic congestion, in some instances it may even be possible to can cause seriously to hand over Interpreter therefore, in order to reduce traffic accident incidence, vehicle supervision department constantly promotes Intelligent traffic management systems, and intelligent transportation The lane change detection technique violating the regulations based on monitor video is more crucial in management system, therefore studies the vehicle based on monitor video and disobey Zhang Biandao detections are necessary.
The existing method to vehicle peccancy lane change detection mainly has two classes：One kind is to be based on space length, as laser is examined Survey method, infrared detection method and ultrasonic Detection Method, such method have that equipment is expensive, space area coverage is small, equipment room The problems such as interfering and circumstance of occlusion can not be handled；Another kind of is to be based on computer vision technique, and such method has The advantages that simple, visual high, accuracy in detection is high, is safeguarded in installation.
At present, the lane change vehicle checking method violating the regulations based on computer vision technique is first extraction moving target set, then Moving vehicle set and positional information are found out, finally judges the movement locus dispersion degree of each car whether beyond setting threshold again Value, vehicle lane change whether violating the regulations is judged with this.Existing method is when road vehicles are less, and detection effect when vehicle movement is very fast Fruit is relatively good；It can however not tackle the lane change violating the regulations when vehicle movement is slow or road vehicles are more.In addition, existing side Method easily by illumination, block, the factor such as shade and video jitter is influenceed；When vehicle is more, the side that is tracked to each car Formula amount of calculation is too big, it is difficult to meet requirement of real time.Therefore, for lane change vehicle detection violating the regulations, it is badly in need of a kind of applied widely, accurate Exactness is high, detection speed is fast, can meet the vehicle peccancy lane change real-time detection method of requirement of real time.
The content of the invention
It is an object of the present invention to drawbacks described above be present for the real-time detection method for solving existing vehicle peccancy lane change, The invention provides a kind of real-time detection method of the vehicle peccancy lane change based on lane line demarcation, this method is applied widely, Testing result is reliable and stable, the degree of accuracy is high, detection speed is fast, can meet the real-time testing requirements of vehicle peccancy lane change.In addition, the party Method can realize the quick accurate identification to lane change vehicle of being broken rules and regulations in video.
To achieve the above object, the invention provides a kind of vehicle peccancy lane change side of detection in real time based on lane line demarcation Method, specifically include：
Step 1) demarcates lane line L, calculates and obtains detection rectangular area D and image I；
Step 2) detects image I vehicle based on deep layer convolutional neural networks in the detection rectangular area D of step 1) Set A={ A1,A2,...,Ai}；
The vehicle set A that step 3) obtains according to step 2), filter out the vehicle set B={ B intersected with lane line L1, B2,...,Bi, wherein,Matched again with tracking vehicle list TL, and update tracking vehicle list TL={ T1, T2,...,Tj}；
Step 4) judges the tracking vehicle T in the tracking vehicle list TL after renewaljWhether it is lane change vehicle violating the regulations；If Track vehicle TjFor lane change vehicle violating the regulations, then tracking vehicle T is marked outjPosition, and from the tracking vehicle list TL after renewal Delete lane change vehicle T violating the regulationsjInformation.
The step 1) specifically includes：
Step 1-1) demarcation lane line L={ L1,L2,...,Lj, every lane line LjPosition include：Starting point pbj= {xbj,ybjAnd terminal pej={ xej,yej, wherein, xbj,ybjThe respectively horizontal stroke of starting point, ordinate, xej,yejRespectively terminal Horizontal, ordinate；
Step 1-2) according to step 1-1) track line position, calculate detection zone rectangle D={ x, y, w, h }, wherein,
X=min (xb1,xe1,xb2,xe2,...,xbj,xej)
Y=min (yb1,ye1,yb2,ye2,...,ybj,yej)
W=max (xb1,xe1,xb2,xe2,...,xbj,xej)-x
H=max (yb1,ye1,yb2,ye2,...,ybj,yej)-y
Step 2) specifically includes：
According to the detection zone D and image I of step 1), the detection zone D in image I is normalized into 576 × 576 works For input picture, the vehicle set A={ A in image I are detected based on deep layer convolutional neural networks1,A2,...,Ai, AiPosition Put and be designated as PAi={ xi,yi,wi,hi, shown in Detection results such as Fig. 2 (b)；
Wherein, the convolutional neural networks include：7 convolutional layers, 4 down-sampling layers, 2 full articulamentums and an output Layer；Scanning boundary filling 0 automatically, enters line activating using Leaky-ReLu function pair neurons in convolutional layer；In down-sampling layer Use maximum pond.
Convolutional layer C1 convolution kernel size is 9 × 9,32 convolution kernels, step-length 2, generation characteristic pattern size is 288 × 288；Down-sampling layer S1 window sizes are 4 × 4, step-length 4, and generation characteristic pattern size is 72 × 72；Convolutional layer C2 convolution kernels are big Small is 1 × Isosorbide-5-Nitrae convolution kernel, and step-length 1, generation characteristic pattern size is 72 × 72；Convolutional layer C3 convolution kernels size is 3 × 3,8 Individual convolution kernel, step-length 1, generation characteristic pattern size are 72 × 72；Down-sampling layer S2 window sizes are 2 × 2, step-length 2, generation Characteristic pattern size is 36 × 36；Convolutional layer C4 convolution kernels size is 1 × 1,8 convolution kernels, step-length 1, generates characteristic pattern size For 36 × 36；Convolutional layer C5 convolution kernels size is 3 × 3,16 convolution kernels, and step-length 1, generation characteristic pattern size is 36 × 36； Down-sampling layer S3 window sizes are 2 × 2, step-length 2, and generation characteristic pattern size is 18 × 18；Convolutional layer C6 convolution kernels size is 3 × 3,32 convolution kernels, step-length 1, generation characteristic pattern size are 18 × 18；Down-sampling layer S4 window sizes are 2 × 2, and step-length is 2, generation characteristic pattern size is 9 × 9；Convolutional layer C7 convolution kernels size is 1 × 1,64 convolution kernels, step-length 1, generates characteristic pattern Size is 9 × 9；Full articulamentum F1 is made up of 256 neurons, enters line activating using Relu function pair neurons；Full articulamentum F2 is made up of 4096 neurons, enters line activating using Leaky-ReLu function pair neurons；Output layer is by 891 neuron structures Into entering line activating using Relu function pair neurons.
The step 3) specifically includes：
Step 3-1) the vehicle set A={ A that are obtained according to step 2)1,A2,...,Ai, therefrom filter out and lane line L Intersecting vehicle set B={ B1,B2,...,Bi, wherein,
Step 3-2) according to step 3-1) obtained vehicle set B, traversal tracking vehicle list TL={ T1,T2,..., Tj, select registration Uij＞ 0.2 vehicle set C,C={ C1,C2,...,Cm}；Calculate Hog features hogjAnd hogk, Its COS distance Cos is calculated againjk；Tracking vehicle T is selected againjWith COS distance Cos maximum in vehicle set CjmIt is and corresponding Vehicle CmIf Cosjm＞ 0.75, then TjAnd CmIt is mutually matched, and enters step 3-3)；If Cosjm≤ 0.75, then vehicle set C In be not present and track vehicle TjThe vehicle to match, and enter step 3-4)；
Step 3-3) according to step 3-2) obtained vehicle Cm, renewal tracking vehicle TjVehicle location be Current vehicle position Put, i.e. TPj=CPm；Renewal tracking vehicle is Current vehicle picture, i.e. TMj=CMm；Continuously with losing number TTj=0, trace point with Lane line L horizontal directed distance TDjIt is updated to the horizontal directed distance of current tracking point and lane line L, i.e. TDj=CDm；This When CmIt has been be matched that, whether Current vehicle is matched mark CFi=true；
Step 3-4) renewal tracking vehicle TjIt is continuous with losing number TTj', wherein, TTj'=TTj+ 1, if TTj' it is continuous With losing number more than 3 times, then vehicle T is deleted from tracking vehicle listj；
Step 3-5) traversal vehicle set B, the vehicle B that will be matchedu={ BPu,BMu,BFu,BQu,BDuAs new Track vehicle Tu={ TPu,TMu,TTu,TODu,TDuBe added in tracking vehicle list TL, vehicle tracking list TL is updated, its In,
TPu=BPu
TMu=BMu
TTu=0
TODu=BDu
TDu=BDu。
The step 3-1) specifically include：
Step 3-1-1) the lane line L and vehicle A that are obtained according to step 1) and step 2)iPosition rectangle PAi, calculate track Line LjWith vehicle AiPosition rectangle PAiThe intersection point on four sides, and lane line LjWith rectangle left edge, right side edge, upper edge, The intersection point of lower edge is respectively plij={ xlij,ylij}、prij={ xrij,yrij}、puij={ xuij,yuij}、pdij={ xdij, ydij}；
Wherein, { xlij,ylijIt is respectively that left side edge intersection point is horizontal, ordinate, { xrij,yrijIt is respectively right side edge intersection point Horizontal, ordinate, { xuij,yuijIt is respectively upper edge intersection point horizontal stroke, ordinate, respectively lower edge intersection point horizontal stroke, ordinate.
Step 3-1-2) according to step 3-1-1) intersection point of obtained lane line and rectangle, if intersection point is on rectangle top and a left side Side and meet following preparatory condition, by vehicle rectangle AiIt is added in vehicle set B；
xi+wi/ 4 ＜ xuij＜ xi+wi
yi+hi/ 2 ＜ ylij＜ yi+hi
Step 3-1-3) according to step 3-1-1) intersection point of obtained lane line and rectangle, if intersection point is on rectangle top and the right side Side and meet following preparatory condition, by vehicle rectangle AiIt is added in vehicle set B；
xi＜ xuij＜ xi+3wi/4
yi+hi/ 2 ＜ yrij＜ yi+hi
Step 3-1-4) according to step 3-1-1) intersection point of obtained lane line and rectangle, if intersection point rectangle top and under Side and meet following preparatory condition, by vehicle rectangle AiIt is added in vehicle set B；
xi+wi/ 4 ＜ xuij＜ xi+3wi/4
xi+wi/ 4 ＜ xdij＜ xi+3wi/4
The step 3-2) specifically include：
Step 3-2-1) according to step 3-1) obtained vehicle set B, information of vehicles is initialized, it includes：Vehicle location BPi={ xi,yi,wi,hi, vehicle pictures BMi, Current vehicle whether be matched mark BFi=false, trace point BQi=(xi+ wi/2,yi+4hi/ 5) and trace point BQiWith every lane line LjHorizontal directed distance Bdij, i.e. vehicle Bi={ BPi,BMi,BFi, BQi,BDi, wherein, BDi={ Bdi1,Bdi2,...,Bdij}；
Step 3-2-2) traversal tracking vehicle list TL={ T1,T2,...,Tj, Tj={ TPj,TMj,TTj,TODj,TDj, Tj∈ TL, wherein, TPjRepresent vehicle location, TMjRepresent vehicle pictures, TTjRepresent continuously with losing number, TODjRepresent starting with The horizontal directed distance of track point and lane line L, TDjRepresent the horizontal directed distance of current tracking point and lane line L；TODj= {TOdj1,TOdj2,...,TOdjl, TDj={ Tdj1,Tdj2,...,Tdjl}；Calculate tracking vehicle TjVehicle location TPjWith car Each truck position BP not being matched in set BiRegistration Uij：
Uij=Scij/Szij
Wherein, SzijFor BPiAnd TPjThe union of rectangular area, ScijFor BPiAnd TPjThe common factor of rectangular area；
Select registration Uij＞ 0.2 vehicle set C,C={ C1,C2,...,Cm}；If C is sky, vehicle collection Close in B and be not present with tracking vehicle TjThe vehicle to match, and enter step 3-4)；If C is not sky, into step 3-2- 3)；
Step 3-2-3) according to step 3-2-2) obtained vehicle set C={ C1,C2,...,Cm, vehicle T will be trackedj's Vehicle pictures TMjWith the picture CM of each car in vehicle set Ck64 × 64 sizes are normalized to, and carry out gray processing, calculate Hog Feature hogjAnd hogk；Wherein, in the calculating, for the window size used for 64 × 64, block size is 16 × 16, and block, which slides, to be increased It is 8 × 8 to measure size, and cell element size is 8 × 8, and the quantity of histogram of gradients is 9 in each born of the same parents' unit；
Step 3-2-4) according to step 3-2-3) obtained Hog features hogjAnd hogk, calculate its COS distance Cosjk；
Step 3-2-5) according to step 3-2-4) obtained COS distance Cosjk, select tracking vehicle TjWith vehicle set C Middle maximum COS distance CosjmAnd its corresponding vehicle CmIf Cosjm＞ 0.75, then TjAnd CmIt is mutually matched, and enters step 3-3)；If Cosjm≤ 0.75, then it is not present and tracks vehicle T in vehicle set CjThe vehicle to match, and enter step 3-4).
The step 4) specifically includes：
Step 4-1) traversal is according to step 3-5) tracking vehicle list TL after obtained renewal, the tracking after traversal renewal Vehicle list TL, calculate TjStarting trace point and lane line LjHorizontal directed distance TOd 'jjWith current tracking point and track Line LjHorizontal directed distance Td 'jjProduct, if TOd 'jj*Td′jj＜ 0, represent that vehicle is successively located at lane line LjBoth sides, Tj For lane change vehicle of breaking rules and regulations；
Step 4-2) according to step 4-1) obtained lane change vehicle T violating the regulationsj, then tracking vehicle T is marked outjPosition, and from Lane change vehicle T violating the regulations is deleted in tracking vehicle list TL after renewaljInformation.
A kind of real-time detecting system based on lane line demarcation vehicle peccancy lane change, the detecting system is intelligent transportation pipe Reason system, including memory, processor and storage on a memory and the computer program that can run on a processor, it is special Sign is, the step of realizing the detection method during computing device described program.
The advantage of the invention is that：
The present invention is that the vehicle set in image is detected based on deep layer convolutional neural networks, compared to utilizing motion feature Extract vehicle set the method degree of accuracy is higher, the scope of application is wider.In addition, the method for the present invention is filtered out and intersected with lane line Vehicle carry out matching judgment again, it is not necessary to record and judge the movement locus of each car, detection speed is fast, can meet in real time knowledge Do not require.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are the present invention Some embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis These accompanying drawings obtain other accompanying drawings.
Fig. 1 is a kind of flow chart of vehicle peccancy lane change real-time detection method based on lane line demarcation of the present invention；
Fig. 2 (a) is to utilize a kind of vehicle peccancy lane change based on lane line demarcation in the embodiment of the present invention to detect in real time The effect diagram of step 1) the setting lane line of method；
Fig. 2 (b) is to utilize a kind of vehicle peccancy lane change based on lane line demarcation in the embodiment of the present invention to detect in real time The effect diagram of step 2) the detection vehicle set of method；
Fig. 2 (c) is to utilize a kind of vehicle peccancy lane change based on lane line demarcation in the embodiment of the present invention to detect in real time The effect diagram for the vehicle set intersected with lane line that the step 3) of method filters out；
Fig. 2 (d) is to utilize a kind of vehicle peccancy lane change based on lane line demarcation in the embodiment of the present invention to detect in real time The effect diagram of lane change vehicle set violating the regulations in step 5) the mark tracking vehicle list of method；
Fig. 2 (e) is to utilize a kind of vehicle peccancy lane change based on lane line demarcation in the embodiment of the present invention to detect in real time Step 5) the vehicles peccancy of method sails out of the effect diagram of state；
Fig. 3 is the network structure of the vehicle detection model in the embodiment of the present invention.
Embodiment
As shown in figure 1, the invention provides a kind of vehicle peccancy lane change real-time detection method based on lane line demarcation, tool Body includes：
Step 1) demarcates lane line L, calculates and obtains detection rectangular area D and image I；
Step 2) detects image I vehicle based on deep layer convolutional neural networks in the detection rectangular area D of step 1) Set A={ A1,A2,...,Ai}；
The vehicle set A that step 3) obtains according to step 2), filter out the vehicle set B={ B intersected with lane line L1, B2,...,Bi, wherein,Matched again with tracking vehicle list TL, and update tracking vehicle list TL={ T1, T2,...,Tj}；
Step 4) judges the tracking vehicle T in the tracking vehicle list TL after renewaljWhether it is lane change vehicle violating the regulations；If Track vehicle TjFor lane change vehicle violating the regulations, then tracking vehicle T is marked outjPosition, and from the tracking vehicle list TL after renewal Delete lane change vehicle T violating the regulationsjInformation.
The step 1) specifically includes：
Step 1-1) demarcation lane line L={ L1,L2,...,Lj, every lane line LjPosition include：Starting point pbj= {xbj,ybjAnd terminal pej={ xej,yej, as shown in Fig. 2 (a)；Wherein, xbj,ybjThe respectively horizontal stroke of starting point, ordinate, xej, yejThe respectively horizontal stroke of terminal, ordinate；
Step 1-2) according to step 1-1) track line position, calculate detection zone rectangle D={ x, y, w, h }, wherein,
X=min (xb1,xe1,xb2,xe2,...,xbj,xej)
Y=min (yb1,ye1,yb2,ye2,...,ybj,yej)
W=max (xb1,xe1,xb2,xe2,...,xbj,xej)-x
H=max (yb1,ye1,yb2,ye2,...,ybj,yej)-y
Step 2) specifically includes：
According to the detection zone D and image I of step 1), the detection zone D in image I is normalized into 576 × 576 works For input picture, the vehicle set A={ A in image I are detected based on deep layer convolutional neural networks1,A2,...,Ai, AiPosition Put and be designated as PAi={ xi,yi,wi,hi, shown in Detection results such as Fig. 2 (b)；
Wherein, as shown in figure 3, the convolutional neural networks include：7 convolutional layers, 4 down-sampling layers, 2 full articulamentums and One output layer；Scanning boundary filling 0 automatically, enters line activating using Leaky-ReLu function pair neurons in convolutional layer；Under Maximum pond is used in sample level.
Convolutional layer C1 convolution kernel size is 9 × 9,32 convolution kernels, step-length 2, generation characteristic pattern size is 288 × 288；Down-sampling layer S1 window sizes are 4 × 4, step-length 4, and generation characteristic pattern size is 72 × 72；Convolutional layer C2 convolution kernels are big Small is 1 × Isosorbide-5-Nitrae convolution kernel, and step-length 1, generation characteristic pattern size is 72 × 72；Convolutional layer C3 convolution kernels size is 3 × 3,8 Individual convolution kernel, step-length 1, generation characteristic pattern size are 72 × 72；Down-sampling layer S2 window sizes are 2 × 2, step-length 2, generation Characteristic pattern size is 36 × 36；Convolutional layer C4 convolution kernels size is 1 × 1,8 convolution kernels, step-length 1, generates characteristic pattern size For 36 × 36；Convolutional layer C5 convolution kernels size is 3 × 3,16 convolution kernels, and step-length 1, generation characteristic pattern size is 36 × 36； Down-sampling layer S3 window sizes are 2 × 2, step-length 2, and generation characteristic pattern size is 18 × 18；Convolutional layer C6 convolution kernels size is 3 × 3,32 convolution kernels, step-length 1, generation characteristic pattern size are 18 × 18；Down-sampling layer S4 window sizes are 2 × 2, and step-length is 2, generation characteristic pattern size is 9 × 9；Convolutional layer C7 convolution kernels size is 1 × 1,64 convolution kernels, step-length 1, generates characteristic pattern Size is 9 × 9；Full articulamentum F1 is made up of 256 neurons, enters line activating using Relu function pair neurons；Full articulamentum F2 is made up of 4096 neurons, enters line activating using Leaky-ReLu function pair neurons；Output layer is by 891 neuron structures Into entering line activating using Relu function pair neurons.
The step 3) specifically includes：
Step 3-1) the vehicle set A={ A that are obtained according to step 2)1,A2,...,Ai, therefrom filter out and lane line L Intersecting vehicle set B={ B1,B2,...,Bi, wherein,
Step 3-2) according to step 3-1) obtained vehicle set B, traversal tracking vehicle list TL={ T1,T2,..., Tj, select registration Uij＞ 0.2 vehicle set C,C={ C1,C2,...,Cm}；Calculate Hog features hogjWith hogk, then calculate its COS distance Cosjk；Tracking vehicle T is selected againjWith COS distance Cos maximum in vehicle set CjmIt is and right The vehicle C answeredmIf Cosjm＞ 0.75, then TjAnd CmIt is mutually matched, and enters step 3-3)；If Cosjm≤ 0.75, then vehicle collection Close in C and be not present and track vehicle TjThe vehicle to match, and enter step 3-4)；
Step 3-3) according to step 3-2) obtained vehicle Cm, renewal tracking vehicle TjVehicle location be Current vehicle position Put, i.e. TPj=CPm；Renewal tracking vehicle is Current vehicle picture, i.e. TMj=CMm；Continuously with losing number TTj=0, trace point with Lane line L horizontal directed distance TDjIt is updated to the horizontal directed distance of current tracking point and lane line L, i.e. TDj=CDm；This When CmIt has been be matched that, whether Current vehicle is matched mark CFi=true；
Step 3-4) renewal tracking vehicle TjIt is continuous with losing number TTj', wherein, TTj'=TTj+ 1, if TTj' it is continuous With losing number more than 3 times, then vehicle T is deleted from tracking vehicle listj；
Step 3-5) traversal vehicle set B, the vehicle B that will be matchedu={ BPu,BMu,BFu,BQu,BDuAs new Track vehicle Tu={ TPu,TMu,TTu,TODu,TDuBe added in tracking vehicle list TL, vehicle tracking list TL is updated, its In,
TPu=BPu
TMu=BMu
TTu=0
TODu=BDu
TDu=BDu。
The step 3-1) specifically include：
Step 3-1-1) the lane line L that is obtained according to step 1) and step 2)jWith vehicle AiPosition rectangle PAi, calculate track Line LjWith vehicle AiPosition rectangle PAiThe intersection point on four sides, and lane line LjWith rectangle left edge, right side edge, upper edge, The intersection point of lower edge is respectively plij={ xlij,ylij}、prij={ xrij,yrij}、puij={ xuij,yuij}、pdij={ xdij, ydij}；
Wherein, { xlij,ylijIt is respectively that left side edge intersection point is horizontal, ordinate, { xrij,yrijIt is respectively right side edge intersection point Horizontal, ordinate, { xuij,yuijIt is respectively upper edge intersection point horizontal stroke, ordinate, respectively lower edge intersection point horizontal stroke, ordinate.
Step 3-1-2) according to step 3-1-1) intersection point of obtained lane line and rectangle, if intersection point is on rectangle top and a left side Side and meet following preparatory condition, by vehicle rectangle AiIt is added in vehicle set B；
xi+wi/ 4 ＜ xuij＜ xi+wi
yi+hi/ 2 ＜ ylij＜ yi+hi
Step 3-1-3) according to step 3-1-1) intersection point of obtained lane line and rectangle, if intersection point is on rectangle top and the right side Side and meet following preparatory condition, by vehicle rectangle AiIt is added in vehicle set B；
xi＜ xuij＜ xi+3wi/4
yi+hi/ 2 ＜ yrij＜ yi+hi
Step 3-1-4) according to step 3-1-1) intersection point of obtained lane line and rectangle, if intersection point rectangle top and under Side and meet following preparatory condition, by vehicle rectangle AiIt is added in vehicle set B；
xi+wi/ 4 ＜ xuij＜ xi+3wi/4
xi+wi/ 4 ＜ xdij＜ xi+3wi/4。
The step 3-2) specifically include：
Step 3-2-1) according to step 3-1) obtained vehicle set B, information of vehicles is initialized, it includes：Vehicle location BPi={ xi,yi,wi,hi, vehicle pictures BMi, Current vehicle whether be matched mark BFi=false, trace point BQi=(xi+ wi/2,yi+4hi/ 5) and trace point BQiWith every lane line LjHorizontal directed distance Bdij, i.e. vehicle Bi={ BPi,BMi,BFi, BQi,BDi, wherein, BDi={ Bdi1,Bdi2,...,Bdil}；
Step 3-2-2) traversal tracking vehicle list TL={ T1,T2,...,Tj, Tj={ TPj,TMj,TTj,TODj,TDj, Tj∈ TL, wherein, TPjRepresent vehicle location, TMjRepresent vehicle pictures, TTjRepresent continuously with losing number, TODjRepresent starting with The horizontal directed distance of track point and lane line L, TDjRepresent the horizontal directed distance of current tracking point and lane line L；TODj= {TOdj1,TOdj2,...,TOdjl, TDj={ Tdj1,Tdj2,...,Tdjl}；Calculate tracking vehicle TjVehicle location TPjWith car Each truck position BP not being matched in set BiRegistration Uij：
Uij=Scij/Szij
Wherein, SzijFor BPiAnd TPjThe union of rectangular area, ScijFor BPiAnd TPjThe common factor of rectangular area；
Select registration Uij＞ 0.2 vehicle set C,C={ C1,C2,...,Cm}；If C is sky, vehicle collection Close in B and be not present with tracking vehicle TjThe vehicle to match, and enter step 3-4)；If C is not sky, into step 3-2- 3)；
Step 3-2-3) according to step 3-2-2) obtained vehicle set C={ C1,C2,...,Cm, vehicle T will be trackedj's Vehicle pictures TMjWith the picture CM of each car in vehicle set Ck64 × 64 sizes are normalized to, and carry out gray processing, calculate Hog Feature hogjAnd hogk；Wherein, in the calculating, for the window size used for 64 × 64, block size is 16 × 16, and block, which slides, to be increased It is 8 × 8 to measure size, and cell element size is 8 × 8, and the quantity of histogram of gradients is 9 in each born of the same parents' unit；
Step 3-2-4) according to step 3-2-3) obtained Hog features hogjAnd hogk, calculate its COS distance Cosjk；
Step 3-2-5) according to step 3-2-4) obtained COS distance Cosjk, select tracking vehicle TjWith vehicle set C Middle maximum COS distance CosjmAnd its corresponding vehicle CmIf Cosjm＞ 0.75, then TjAnd CmIt is mutually matched, and enters step 3-3), the match is successful shown in vehicle detection effect such as Fig. 2 (c)；If Cosjm≤ 0.75, then it is not present and tracks in vehicle set C Vehicle TjThe vehicle to match, and enter step 3-4), it fails to match shown in vehicle detection effect such as Fig. 2 (d).
The step 4) specifically includes：
Step 4-1) the tracking vehicle list TL after renewal is traveled through, calculate TjStarting trace point and lane line LjLevel Directed distance TOd 'jjWith current tracking point and lane line LjHorizontal directed distance Td 'jjProduct, if TOd 'jj*Td′jj＜ 0, Represent that vehicle is successively located at lane line LjBoth sides, TjFor lane change vehicle of breaking rules and regulations；
Step 4-2) according to step 4-1) obtained lane change vehicle T violating the regulationsj, then tracking vehicle T is marked outjPosition, such as scheme Shown in 2 (e), and lane change vehicle T violating the regulations is deleted from the tracking vehicle list TL after renewaljInformation.
Wherein, Fig. 2 (a) -2 (e) is the effect detected using the method in the embodiment of the present invention to lane change vehicle of breaking rules and regulations Fruit schematic diagram.
A kind of real-time detecting system based on lane line demarcation vehicle peccancy lane change, the detecting system is intelligent transportation pipe Reason system, including memory, processor and storage on a memory and the computer program that can run on a processor, it is special Sign is, the step of realizing the detection method during computing device described program.
It should be noted last that the above embodiments are merely illustrative of the technical solutions of the present invention and it is unrestricted.Although ginseng The present invention is described in detail according to embodiment, it will be understood by those within the art that, to the technical side of the present invention Case is modified or equivalent substitution, and without departure from the spirit and scope of technical solution of the present invention, it all should cover in the present invention Right among.

Claims (8)

1. a kind of real-time detection method based on lane line demarcation vehicle peccancy lane change, is specifically included：
Step 1) demarcates lane line L, calculates and obtains detection rectangular area D and image I；
Step 2) detects image I vehicle set based on deep layer convolutional neural networks in the detection rectangular area D of step 1) A={ A1,A2,...,Ai}；
The vehicle set A that step 3) obtains according to step 2), filter out the vehicle set B={ B intersected with lane line L1, B2,...,Bi, wherein,Matched again with tracking vehicle list TL, and update tracking vehicle list TL={ T1, T2,...,Tj}；
Step 4) judges the tracking vehicle T in the tracking vehicle list TL after renewaljWhether it is lane change vehicle violating the regulations；If tracking Vehicle TjFor lane change vehicle violating the regulations, then tracking vehicle T is marked outjPosition, and deleted from the tracking vehicle list TL after renewal Lane change vehicle T violating the regulationsjInformation.
2. detection method according to claim 1, it is characterised in that the step 1) specifically includes：
Step 1-1) demarcation lane line L={ L1,L2,...,Lj, every lane line LjPosition include：Starting point pbj={ xbj, ybjAnd terminal pej={ xej,yej}；Wherein, xbj,ybjThe respectively horizontal stroke of starting point, ordinate, xej,yejRespectively the horizontal stroke of terminal, Ordinate；
Step 1-2) according to step 1-1) track line position, calculate detection zone rectangle D={ x, y, w, h }, wherein,
X=min (xb1,xe1,xb2,xe2,...,xbj,xej)
Y=min (yb1,ye1,yb2,ye2,...,ybj,yej)
W=max (xb1,xe1,xb2,xe2,...,xbj,xej)-x
H=max (yb1,ye1,yb2,ye2,...,ybj,yej)-y。
3. detection method according to claim 1, it is characterised in that step 2) specifically includes：
According to the detection zone D and image I of step 1), the detection zone D in image I is normalized to 576 × 576 as defeated Enter image, the vehicle set A={ A in image I are detected based on deep layer convolutional neural networks1,A2,...,Ai, AiPosition note For PAi={ xi,yi,wi,hi}；
Wherein, the convolutional neural networks include：7 convolutional layers, 4 down-sampling layers, 2 full articulamentums and an output layer；Volume Scanning boundary filling 0 automatically, enters line activating using Leaky-ReLu function pair neurons in lamination；Used in down-sampling layer Maximum pond；
Convolutional layer C1 convolution kernel size is 9 × 9,32 convolution kernels, and step-length 2, generation characteristic pattern size is 288 × 288；Under Sample level S1 window sizes are 4 × 4, step-length 4, and generation characteristic pattern size is 72 × 72；
Convolutional layer C2 convolution kernels size is 1 × Isosorbide-5-Nitrae convolution kernel, and step-length 1, generation characteristic pattern size is 72 × 72；
Convolutional layer C3 convolution kernels size is 3 × 3,8 convolution kernels, and step-length 1, generation characteristic pattern size is 72 × 72；Down-sampling Layer S2 window sizes are 2 × 2, step-length 2, and generation characteristic pattern size is 36 × 36；
Convolutional layer C4 convolution kernels size is 1 × 1,8 convolution kernels, and step-length 1, generation characteristic pattern size is 36 × 36；
Convolutional layer C5 convolution kernels size is 3 × 3,16 convolution kernels, and step-length 1, generation characteristic pattern size is 36 × 36；Down-sampling Layer S3 window sizes are 2 × 2, step-length 2, and generation characteristic pattern size is 18 × 18；
Convolutional layer C6 convolution kernels size is 3 × 3,32 convolution kernels, and step-length 1, generation characteristic pattern size is 18 × 18；Down-sampling Layer S4 window sizes are 2 × 2, step-length 2, and generation characteristic pattern size is 9 × 9；
Convolutional layer C7 convolution kernels size is 1 × 1,64 convolution kernels, and step-length 1, generation characteristic pattern size is 9 × 9；
Full articulamentum F1 is made up of 256 neurons, enters line activating using Relu function pair neurons；Full articulamentum F2 is by 4096 Individual neuron is formed, and enters line activating using Leaky-ReLu function pair neurons；Output layer is made up of 891 neurons, is used Relu function pair neurons enter line activating.
4. detection method according to claim 1, it is characterised in that the step 3) specifically includes：
Step 3-1) the vehicle set A={ A that are obtained according to step 2)1,A2,...,Ai, therefrom filter out and intersect with lane line L Vehicle set B={ B1,B2,...,Bi, wherein,
Step 3-2) according to step 3-1) obtained vehicle set B, traversal tracking vehicle list TL={ T1,T2,...,Tj, choosing Go out registration Uij＞ 0.2 vehicle set C,C={ C1,C2,...,Cm}；Calculate Hog features hogjAnd hogk, then count Calculate its COS distance Cosjk；Tracking vehicle T is selected againjWith COS distance Cos maximum in vehicle set CjmAnd corresponding vehicle CmIf Cosjm＞ 0.75, then TjAnd CmIt is mutually matched, and enters step 3-3)；If Cosjm≤ 0.75, then in vehicle set C not In the presence of and tracking vehicle TjThe vehicle to match, and enter step 3-4)；
Step 3-3) according to step 3-2) obtained vehicle Cm, renewal tracking vehicle TjVehicle location be current vehicle location, i.e., TPj=CPm；Renewal tracking vehicle is Current vehicle picture, i.e. TMj=CMm；Continuously with losing number TTj=0, trace point and track Line L horizontal directed distance TDjIt is updated to the horizontal directed distance of current tracking point and lane line L, i.e. TDj=CDm；Now Cm It has been be matched that, whether Current vehicle is matched mark CFi=true；
Step 3-4) renewal tracking vehicle TjIt is continuous with losing number TTj', wherein, TTj'=TTj+ 1, if TTj' it is continuous with losing Number is more than 3 times, then deletes vehicle T from tracking vehicle listj；
Step 3-5) traversal vehicle set B, the vehicle B that will be matchedu={ BPu,BMu,BFu,BQu,BDuAs new tracking Vehicle Tu={ TPu,TMu,TTu,TODu,TDuBe added in tracking vehicle list TL, vehicle tracking list TL is updated, wherein,
TPu=BPu
TMu=BMu
TTu=0
TODu=BDu
TDu=BDu。
5. detection method according to claim 4, it is characterised in that the step 3-1) specifically include：
Step 3-1-1) the lane line L and vehicle A that are obtained according to step 1) and step 2)iPosition rectangle PAi, calculate lane line Lj With vehicle AiPosition rectangle PAiThe intersection point on four sides, and lane line LjWith rectangle left edge, right side edge, upper edge, downside The intersection point at edge is respectively plij={ xlij,ylij}、prij={ xrij,yrij}、puij={ xuij,yuij}、pdij={ xdij,ydij}；
Wherein, { xlij,ylijIt is respectively that left side edge intersection point is horizontal, ordinate, { xrij,yrijBe respectively right side edge intersection point it is horizontal, Ordinate, { xuij,yuijIt is respectively upper edge intersection point horizontal stroke, ordinate, respectively lower edge intersection point horizontal stroke, ordinate；
Step 3-1-2) according to step 3-1-1) intersection point of obtained lane line and rectangle, if intersection point on rectangle top and the left side and Meet following preparatory condition, by vehicle rectangle AiIt is added in vehicle set B；
xi+wi/ 4 ＜ xuij＜ xi+wi
yi+hi/ 2 ＜ ylij＜ yi+hi；
Step 3-1-3) according to step 3-1-1) intersection point of obtained lane line and rectangle, if intersection point on rectangle top and the right and Meet following preparatory condition, by vehicle rectangle AiIt is added in vehicle set B；
xi＜ xuij＜ xi+3wi/4
yi+hi/ 2 ＜ yrij＜ yi+hi；
Step 3-1-4) according to step 3-1-1) intersection point of obtained lane line and rectangle, if intersection point in rectangle bottom and upper segment and Meet following preparatory condition, by vehicle rectangle AiIt is added in vehicle set B；
xi+wi/ 4 ＜ xuij＜ xi+3wi/4
xi+wi/ 4 ＜ xdij＜ xi+3wi/4。
6. detection method according to claim 4, it is characterised in that the step 3-2) specifically include：
Step 3-2-1) according to step 3-1) obtained vehicle set B, information of vehicles is initialized, including：Vehicle location BPi ={ xi,yi,wi,hi, vehicle pictures BMi, Current vehicle whether be matched mark BFi=false, trace point BQi=(xi+wi/ 2,yi+4hi/ 5) and trace point BQiWith every lane line LjHorizontal directed distance Bdij, that is, track vehicle Bi={ BPi,BMi, BFi,BQi,BDi, wherein, BDi={ Bdi1,Bdi2,...,Bdil}；
Step 3-2-2) traversal tracking vehicle list TL={ T1,T2,...,Tj, Tj={ TPj,TMj,TTj,TODj,TDj, Tj∈ TL, wherein, TPjRepresent vehicle location, TMjRepresent vehicle pictures, TTjRepresent continuously with losing number, TODjRepresent starting trace point With lane line L horizontal directed distance, TDjRepresent the horizontal directed distance of current tracking point and lane line L；TODj={ TOdj1, TOdj2,...,TOdjl, TDj={ Tdj1,Tdj2,...,Tdjl}；Calculate tracking vehicle TjVehicle location TPjWith vehicle set B In each truck position BP for not being matchediRegistration Uij：
Uij=Scij/Szij
Wherein, SzijFor BPiAnd TPjThe union of rectangular area, ScijFor BPiAnd TPjThe common factor of rectangular area；
Select registration Uij＞ 0.2 vehicle set C,C={ C1,C2,...,Cm}；If C is sky, in vehicle set B In the absence of with tracking vehicle TjThe vehicle to match, and enter step 3-4)；If C is not sky, into step 3-2-3)；
Step 3-2-3) according to step 3-2-2) obtained vehicle set C={ C1,C2,...,Cm, vehicle T will be trackedjVehicle Picture TMjWith the picture CM of each car in vehicle set Ck64 × 64 sizes are normalized to, and carry out gray processing, calculate Hog features hogjAnd hogk；Wherein, in the calculating, the window size used is 64 × 64, and block size is 16 × 16, and it is big that block slides increment Small is 8 × 8, and cell element size is 8 × 8, and the quantity of histogram of gradients is 9 in each born of the same parents' unit；
Step 3-2-4) according to step 3-2-3) obtained Hog features hogjAnd hogk, calculate its COS distance Cosjk；
Step 3-2-5) according to step 3-2-4) obtained COS distance Cosjk, select tracking vehicle TjWith in vehicle set C most Big COS distance CosjmAnd its corresponding vehicle CmIf Cosjm＞ 0.75, then TjAnd CmIt is mutually matched, and enters step 3- 3)；If Cosjm≤ 0.75, then it is not present and tracks vehicle T in vehicle set CjThe vehicle to match, and enter step 3-4).
7. detection method according to claim 1, it is characterised in that the step 4) specifically includes：
Step 4-1) the tracking vehicle list TL after renewal is traveled through, calculate TjStarting trace point and lane line LjIt is horizontal oriented Distance TOd 'jjWith current tracking point and lane line LjHorizontal directed distance Td 'jjProduct, if TOd 'jj*Td′jj＜ 0, represent Vehicle is successively located at lane line LjBoth sides, TjFor lane change vehicle of breaking rules and regulations；
Step 4-2) according to step 4-1) obtained lane change vehicle T violating the regulationsj, then tracking vehicle T is marked outjPosition, and from renewal Lane change vehicle T violating the regulations is deleted in tracking vehicle list TL afterwardsjInformation.
8. a kind of real-time detecting system based on lane line demarcation vehicle peccancy lane change, the detecting system is intelligent traffic administration system System, including memory, processor and storage on a memory and the computer program that can run on a processor, its feature It is, the step of claim 1~7 detection method is realized during the computing device described program.
CN201711026761.4A 2017-10-27 2017-10-27 Real-time detection method and system for calibrating illegal lane change of vehicle based on lane line Active CN107705577B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711026761.4A CN107705577B (en) 2017-10-27 2017-10-27 Real-time detection method and system for calibrating illegal lane change of vehicle based on lane line

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711026761.4A CN107705577B (en) 2017-10-27 2017-10-27 Real-time detection method and system for calibrating illegal lane change of vehicle based on lane line

Publications (2)

Publication Number Publication Date
CN107705577A true CN107705577A (en) 2018-02-16
CN107705577B CN107705577B (en) 2020-05-26

Family

ID=61176340

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711026761.4A Active CN107705577B (en) 2017-10-27 2017-10-27 Real-time detection method and system for calibrating illegal lane change of vehicle based on lane line

Country Status (1)

Country Link
CN (1) CN107705577B (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109271942A (en) * 2018-09-26 2019-01-25 上海七牛信息技术有限公司 A kind of stream of people's statistical method and system
CN109284752A (en) * 2018-08-06 2019-01-29 中国科学院声学研究所 A kind of rapid detection method of vehicle
CN110348273A (en) * 2018-04-04 2019-10-18 北京四维图新科技股份有限公司 Neural network model training method, system and Lane detection method, system
CN112712703A (en) * 2020-12-09 2021-04-27 上海眼控科技股份有限公司 Vehicle video processing method and device, computer equipment and storage medium
CN114120625A (en) * 2020-08-31 2022-03-01 上汽通用汽车有限公司 Vehicle information integration system, method, and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102254429A (en) * 2011-05-13 2011-11-23 东南大学 Video identification-based detection apparatus and method of vehicles against regulations
CN105632186A (en) * 2016-03-11 2016-06-01 博康智能信息技术有限公司 Method and device for detecting vehicle queue jumping behavior
CN105702048A (en) * 2016-03-23 2016-06-22 武汉理工大学 Automobile-data-recorder-based illegal lane occupation identification system and method for automobile on highway
WO2016141553A1 (en) * 2015-03-10 2016-09-15 冯旋宇 System for preventing lane change on solid-line road and method for preventing lane change by using system
CN106297314A (en) * 2016-11-03 2017-01-04 北京文安智能技术股份有限公司 A kind of drive in the wrong direction or the detection method of line ball vehicle behavior, device and a kind of ball machine

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102254429A (en) * 2011-05-13 2011-11-23 东南大学 Video identification-based detection apparatus and method of vehicles against regulations
WO2016141553A1 (en) * 2015-03-10 2016-09-15 冯旋宇 System for preventing lane change on solid-line road and method for preventing lane change by using system
CN105632186A (en) * 2016-03-11 2016-06-01 博康智能信息技术有限公司 Method and device for detecting vehicle queue jumping behavior
CN105702048A (en) * 2016-03-23 2016-06-22 武汉理工大学 Automobile-data-recorder-based illegal lane occupation identification system and method for automobile on highway
CN106297314A (en) * 2016-11-03 2017-01-04 北京文安智能技术股份有限公司 A kind of drive in the wrong direction or the detection method of line ball vehicle behavior, device and a kind of ball machine

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110348273A (en) * 2018-04-04 2019-10-18 北京四维图新科技股份有限公司 Neural network model training method, system and Lane detection method, system
CN109284752A (en) * 2018-08-06 2019-01-29 中国科学院声学研究所 A kind of rapid detection method of vehicle
CN109271942A (en) * 2018-09-26 2019-01-25 上海七牛信息技术有限公司 A kind of stream of people's statistical method and system
CN114120625A (en) * 2020-08-31 2022-03-01 上汽通用汽车有限公司 Vehicle information integration system, method, and storage medium
CN114120625B (en) * 2020-08-31 2023-02-21 上汽通用汽车有限公司 Vehicle information integration system, method, and storage medium
CN112712703A (en) * 2020-12-09 2021-04-27 上海眼控科技股份有限公司 Vehicle video processing method and device, computer equipment and storage medium

Also Published As

Publication number Publication date
CN107705577B (en) 2020-05-26

Similar Documents

Publication Publication Date Title
CN107705577A (en) A kind of real-time detection method and system based on lane line demarcation vehicle peccancy lane change
Chen et al. Lane departure warning systems and lane line detection methods based on image processing and semantic segmentation: A review
CN110738857B (en) Vehicle violation evidence obtaining method, device and equipment
CN108335489A (en) Vehicle on highway behavior semantic analysis and abnormal behaviour monitoring system and method
CN109460709A (en) The method of RTG dysopia analyte detection based on the fusion of RGB and D information
CN104537841A (en) Unlicensed vehicle violation detection method and detection system thereof
CN103778786A (en) Traffic violation detection method based on significant vehicle part model
CN104574993B (en) A kind of method of road monitoring and device
CN106446150A (en) Method and device for precise vehicle retrieval
CN107886055A (en) A kind of retrograde detection method judged for direction of vehicle movement
CN103632427B (en) A kind of gate cracking protection method and gate control system
CN107644206A (en) A kind of road abnormal behaviour action detection device
CN114898296A (en) Bus lane occupation detection method based on millimeter wave radar and vision fusion
CN111311766A (en) Roadside parking intelligent charging system and method based on license plate recognition and tracking technology
Dow et al. A crosswalk pedestrian recognition system by using deep learning and zebra‐crossing recognition techniques
CN106778484A (en) Moving vehicle tracking under traffic scene
CN113903008A (en) Ramp exit vehicle violation identification method based on deep learning and trajectory tracking
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
CN113011331A (en) Method and device for detecting whether motor vehicle gives way to pedestrians, electronic equipment and medium
Tao et al. Smoky vehicle detection based on range filtering on three orthogonal planes and motion orientation histogram
CN116912517B (en) Method and device for detecting camera view field boundary
CN106960193A (en) A kind of lane detection apparatus and method
CN116311166A (en) Traffic obstacle recognition method and device and electronic equipment
CN114693722A (en) Vehicle driving behavior detection method, detection device and detection equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20220729

Address after: 100190, No. 21 West Fourth Ring Road, Beijing, Haidian District

Patentee after: INSTITUTE OF ACOUSTICS, CHINESE ACADEMY OF SCIENCES

Patentee after: NANHAI RESEARCH STATION, INSTITUTE OF ACOUSTICS, CHINESE ACADEMY OF SCIENCES

Address before: 100190, No. 21 West Fourth Ring Road, Beijing, Haidian District

Patentee before: INSTITUTE OF ACOUSTICS, CHINESE ACADEMY OF SCIENCES

TR01 Transfer of patent right