CN106803064A - A kind of traffic lights method for quickly identifying - Google Patents
A kind of traffic lights method for quickly identifying Download PDFInfo
- Publication number
- CN106803064A CN106803064A CN201611214259.1A CN201611214259A CN106803064A CN 106803064 A CN106803064 A CN 106803064A CN 201611214259 A CN201611214259 A CN 201611214259A CN 106803064 A CN106803064 A CN 106803064A
- Authority
- CN
- China
- Prior art keywords
- traffic lights
- image
- region
- traffic
- quickly identifying
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/752—Contour matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a kind of traffic lights method for quickly identifying, including obtain vehicle front road image;Position where quick positioning traffic lights;Denoising is carried out to image;Segmentation image-region;Image is screened;Traffic lights position is confirmed using " cross checking " method;Then further identification traffic lights is circular or arrow.The present invention is by multiple authentication, and degree of accuracy rate is high;Amount of calculation is small, and recognition speed is fast, is capable of identify that circular and arrow-shaped traffic lights, small by such environmental effects, is easy to be applied to pilotless automobile, vehicle DAS (Driver Assistant System).
Description
Technical field
The present invention relates to active safety systems of vehicles field, and in particular to a kind of traffic lights method for quickly identifying.
Background technology
Nowadays, most of existing traffic lights identification technology uses computer graphics disposal technology, computer learning, god
Through technologies such as networks.What is generally used is directly to use the algorithm of template matches or the traffic lights based on SVM SVMs to know
Other algorithm.Because there is the defects such as recognition correct rate is low, recognition rate is slow in the algorithm of template matches;And use SVM SVMs
Traffic lights recognizer, it is necessary to carry out substantial amounts of sample instruction according to different environment (such as daytime, night, the cloudy day, reflective)
Practice, in addition it is also necessary to be trained according to different road traffic crossing positions, operand is big, recognition speed is slow, it is affected by environment compared with
Greatly.
The content of the invention
In order to solve the technical problem existing for prior art, the present invention provides a kind of traffic lights method for quickly identifying, should
Method operand is few, recognition speed is fast, recognition accuracy is high, small by such environmental effects, be easy to be applied to pilotless automobile,
Vehicle DAS (Driver Assistant System).
Traffic lights method for quickly identifying of the invention, comprises the following steps:
One:Obtain vehicle front road information image;
Two:From the road ahead frame for obtaining, quick leapfrog is carried out to traffic lights picture according to traffic lights color
Positioning, obtains traffic lights rough location;
Three:Region division is carried out to the traffic lights rough location that quick leapfrog positioning is obtained, is divided into tri- passages of RGB;Point
The other color to traffic lights is separated, by separating resulting binaryzation;Carry out mean denoising treatment;
Four:Split in continuous image vegetarian refreshments region to image after step 3 treatment;
Five:The image after segmentation is screened using the method for width/height ≈ 1.0, traffic lights position is substantially confirmed, and
Recording traffic lamp position is put in the position of artwork, and wherein width refers to the width of image after segmentation, and height refers to the height of image after segmentation
Degree;
Six:" cross checking " method of utilization further confirms that, detects whether certain region is position where traffic lights;
" cross checking " method by step 5 screen image centered on position, along the center it is upper and lower,
Left and right four direction each extends over out one with its size identical square, and the square to four direction carries out detection fortune
Calculate;If detecting the traffic lights color accounting that some direction is definition reaches 85% or more, judge that center is
Position where traffic lights;
Seven:Detect in region to confirmed traffic lights position, if be circle, if otherwise performing step 8;
Eight:The profile of image is obtained using canny, image HU squares are calculated, is matched with standard arrow HU squares;If with mark
Quasi- arrow HU match by moment is unsuccessful, performs step 9;
Nine:Sending picture to SVM classifier carries out Classification and Identification.
Compared with prior art, the invention has the advantages that and beneficial effect:
1st, the present invention carries out the positioning of suspicious region by quick leapfrog first, greatly reduces the calculating to image procossing
Amount.For example need to carry out binary conversion treatment to image in step s3, needed to image institute if suspicious region positioning is not carried out
Some pixels are calculated, and now need to only carry out processing suspicious region, so as to shorten the calculating time.
2nd, after segmentation figure picture, traffic lights position is confirmed using cross proof method, improves the accuracy rate of algorithm.
When the 3rd, being identified to traffic lights, complementary identification fully is carried out using regular geometric figure, when running into circular friendship
During ventilating signal lamp, HU square calculating can not be carried out, without SVM classifier is used, shorten the time of image procossing.
4th, HU squares are introduced to be matched, can effectively recognizes arrowhead-shaped traffic lights;Figure is improve using SVM classifier
As it is unintelligible when discrimination.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of traffic lights method for quickly identifying of the present invention;
Fig. 2 is to carry out the picture after traffic lights color separated;
Fig. 3 is that one of image segmentation and the traffic light signals picture of screening are carried out to Fig. 2;
Fig. 4 is two that image segmentation and the traffic light signals picture of screening are carried out to Fig. 2.
Specific embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described further, but embodiments of the present invention
Not limited to this.
Embodiment
Referring to Fig. 1, a kind of traffic lights method for quickly identifying of the invention is comprised the following steps:
S1:Vehicle front road information image is obtained using camera.
S2:From the road ahead frame for obtaining, quick leapfrog positioning is carried out to traffic lights picture, obtain traffic lights
Rough location.Quick leapfrog is by detecting that traffic lights color, to determine traffic lights suspicious region, can select the side of (5x, 5y)
Formula, i.e., carry out quick leapfrog positioning in units of 5 pixels, reads the quick pre-determined bit traffic lights suspicious region of image information, with
Reach the purpose for reducing treatment image information speed up processing.Wherein, traffic lights color includes red, yellow, and green and black.
S3:Region division is carried out to the traffic lights rough location that quick leapfrog positioning is obtained, and the region is divided into RGB tri-
Individual passage, respectively the color to traffic lights separate, by separating resulting binaryzation, each passage is only existed 0 and 255
Value;Mean denoising treatment is carried out, scattered pixel is disposed.Result is as shown in Figure 2.Color to traffic lights is separated
When, mainly it is compared by tri- calculating of passage of RGB and with default threshold value.
S4:Split in continuous image vegetarian refreshments region to image after step S3 treatment.It is continuous that this step calculates image first
Region, is then divided into an independent small images by each image continuum.There is weight when two image sections are partitioned into
During multiple part, two image sections are merged, take its length and width maximum.
S5:The image after segmentation is screened using the method for width/height ≈ 1.0, it is more independent to remove image border
Pixel, substantially confirm traffic lights position, and recording traffic lamp position is put in the position of artwork, and qualified to each
Image tagged 2R=width, wherein width refer to the width of image after segmentation, and height refers to the height of image after segmentation.No matter hand over
Logical lamp is arrow or circle, can be processed using this step, and the result of arrow traffic lights is as shown in figure 3, circular traffic
The result of lamp is as shown in Figure 4.
S6:" cross checking " method of utilization further confirms that, detects whether certain region is position where traffic lights.
The present invention is mainly when being applied to start and stop, the steering of unmanned motor vehicle on road etc. to the quick of traffic lights
Identification, because the traffic lights for indicating vehicle start and stop, turning to type is typically all more than three, and three traffic lights horizontally sets or
It is vertically arranged, therefore general at least one direction of upper and lower, left and right four direction of a certain traffic lights is the presence of adjacent traffic
Lamp.This step " cross proof method ", is based on this premise to realize.
Due to the picture altitude and wide-band ratio close to 1 screened in step S5:1, " cross checking " method is with region to be detected
Centered on position, along the upper and lower, left and right four direction of the center each extend over out the length of side be 2R square, i.e., in step
The upper and lower, left and right that rapid S5 screens image take one with its size identical region respectively, and the square to four direction enters
Row detection calculations.If (traffic lights is red, yellow, green or black to detect traffic lights color that some direction is definition
Color) accounting reaches 85% or more, then it is assumed that and center is traffic light signals;That is, the square region for extending out
In domain, if traffic lights color accounting reaches more than 85%, position of the center where traffic lights is judged.
S7:Count according to observations, there is white iron ring most traffic panel outside, thus the present invention can also be further
Detection " cross checking " traffic lights position determined by method, if there is white iron ring, so as to finally confirm traffic lights position.
If not calculating the white iron ring outside traffic panel, step S8 is transferred to.
This step is not necessarily.Traffic lights position is substantially confirmed by step S6, around the position
Further detect whether there is white portion, traffic lights position can be further determined that.If there is white portion in surrounding, along white
Color region detection, uses the traffic lights institute that " cross checking " method is detected in place when white portion is closed and surrounds step S6
Then think that white iron ring is present when putting.
S8:Detect in region to confirmed traffic lights position, if be circle.The detection of this step confirms traffic lights
When whether the region of position is circle, method therefor is:Calculate graphics area S, calculate image girth C, due to circular area,
Girth has following characteristics:
S/C=(π R2The π R of)/(2)=R/2
Think that this is shaped as circle if now S/C=(width+height)/8.Turn if testing result is for circle
Enter step S11, otherwise perform step S9.
S9:The profile of image is obtained using canny, image HU squares are calculated, is matched with standard arrow HU squares;If with mark
Quasi- arrow HU match by moment success is then transferred to step S11, otherwise performs step S10.
In step s 9, calculating the formula used by the HU squares of figure is:
I1=y20+y02
I2=(y20+y02)2+4y11 2
I3=(y30+3y)2+(3y21-y03)2
I4=(y30+y12)2+(y21+y03)2
I5=(y30-y12)(y30-y12)[(y30+y12)2-3(y21+y03)2]+(3y21-y03)(y21+y30)[3(y30+y12)2-
(y21+y03)2]
I6=(y20-y02)[(y30+y12)2-(y21+y03)2]+4y11(y30+y12)(y21+y03)
I7=(3y21+y03)(y30+y12)[(y30+y12)2-3(y21+y03)2]+9y30-3y12)(y21+y30)[3(y30+y12)2-
(y21+y03)2]
Wherein, IkFor not bending moment, ypqFor (p+q) rank normalizes central moment, k, q and p are integer, and 1≤k≤7,0≤p
≤ 3,0≤q≤3.Above-mentioned seven not bending moment be to be constructed by second order and third central moment, no matter which resulting arrow is
Direction, or the arrow cause not of uniform size for detecting, its HU squares all without changing, it is possible to reaching quick detection
Purpose.
S10:Sending picture to SVM classifier carries out Classification and Identification.Wherein, SVM classifier need to be trained.Training
When, the feature of picture is extracted using SVM classifier, multi-C vector feature is formed, then handmarking's sample is left and right, advance
Manual identified result is stored into feature database by arrow, last SVM classifier.
S11:Operation result is sent by serial ports.
Because step S8, step S9, step S10 successively enter, thus most result can operand compared with
Few step S8, step S9 is completed, and reduces recognition time.
Above-described embodiment is the present invention preferably implementation method, but embodiments of the present invention are not by above-described embodiment
Limitation, it is other it is any without departing from Spirit Essence of the invention and the change, modification, replacement made under principle, combine, simplification,
Equivalent substitute mode is should be, is included within protection scope of the present invention.
Claims (4)
1. a kind of traffic lights method for quickly identifying, it is characterised in that comprise the following steps:
One:Obtain vehicle front road information image;
Two:From the road ahead frame for obtaining, quick leapfrog positioning is carried out to traffic lights picture according to traffic lights color,
Obtain traffic lights rough location;
Three:Region division is carried out to the traffic lights rough location that quick leapfrog positioning is obtained, is divided into tri- passages of RGB;It is right respectively
The color of traffic lights is separated, by separating resulting binaryzation;Carry out mean denoising treatment;
Four:Split in continuous image vegetarian refreshments region to image after step 3 treatment;
Five:The image after segmentation is screened using the method for width/height ≈ 1.0, traffic lights position is substantially confirmed, and record
In the position of artwork, wherein width refers to the width of image after segmentation for traffic lights position, and height refers to the height of image after segmentation;
Six:" cross checking " method of utilization further confirms that, detects whether certain region is position where traffic lights;
" cross checking " method by step 5 screen image centered on position, along the upper and lower, left and right of the center
Four direction each extends over out one with its size identical square, and the square to four direction carries out detection calculations;Such as
Fruit detects the traffic lights color accounting that some direction is definition and reaches 85% or more, then judge that center is traffic lights
The position at place;
Seven:Detect in region to confirmed traffic lights position, if be circle, if otherwise performing step 8;
Eight:The profile of image is obtained using canny, image HU squares are calculated, is matched with standard arrow HU squares;If with standard arrow
Head HU match by moment is unsuccessful, performs step 9;
Nine:Sending picture to SVM classifier carries out Classification and Identification.
2. traffic lights method for quickly identifying according to claim 1, it is characterised in that the step 8 calculates figure HU squares
Formula used is:
I1=y20+y02;
I2=(y20+y02)2+4y11 2;
I3=(y30+3y)2+(3y21-y03)2;
I4=(y30+y12)2+(y21+y03)2;
I5=(y30-y12)(y30-y12)[(y30+y12)2-3(y21+y03)2]+(3y21-y03)(y21+y30)[3(y30+y12)2-(y21+
y03)2];
I6=(y20-y02)[(y30+y12)2-(y21+y03)2]+4y11(y30+y12)(y21+y03);
I7=(3y21+y03)(y30+y12)[(y30+y12)2-3(y21+y03)2]+9y30-3y12)(y21+y30)[3(y30+y12)2-(y21+
y03)2];
Wherein, IkFor not bending moment, ypqIt is (p+q) rank normalization central moment, k, q and p are integer, and 1≤k≤7,0≤p≤3,
0≤q≤3。
3. traffic lights method for quickly identifying according to claim 1, it is characterised in that the step 9 SVM classifier is needed
It is trained;During training, the feature of picture is extracted using SVM classifier, form multi-C vector feature, then handmarking's sample
It is left and right, forward arrow, manual identified result is stored into feature database by last SVM classifier.
4. traffic lights method for quickly identifying according to claim 1, it is characterised in that between the step 6 and step 7
Also perform:
Detection whether there is white portion around " cross checking " traffic lights position determined by method;If there is white in surrounding
Region, then detect, along white portion when white portion closes and surround the traffic that use " cross checking " method is detected
Then think that white iron ring is present during lamp position.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611214259.1A CN106803064B (en) | 2016-12-26 | 2016-12-26 | Traffic light rapid identification method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611214259.1A CN106803064B (en) | 2016-12-26 | 2016-12-26 | Traffic light rapid identification method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106803064A true CN106803064A (en) | 2017-06-06 |
CN106803064B CN106803064B (en) | 2020-05-19 |
Family
ID=58984236
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611214259.1A Active CN106803064B (en) | 2016-12-26 | 2016-12-26 | Traffic light rapid identification method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106803064B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107563301A (en) * | 2017-08-09 | 2018-01-09 | 上海炬宏信息技术有限公司 | Red signal detection method based on image processing techniques |
CN109635640A (en) * | 2018-10-31 | 2019-04-16 | 百度在线网络技术(北京)有限公司 | Traffic light recognition method, device, equipment and storage medium based on cloud |
CN110021176A (en) * | 2018-12-21 | 2019-07-16 | 文远知行有限公司 | Traffic lights decision-making technique, device, computer equipment and storage medium |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011216051A (en) * | 2010-04-02 | 2011-10-27 | Institute Of National Colleges Of Technology Japan | Program and device for discriminating traffic light |
CN102819263A (en) * | 2012-07-30 | 2012-12-12 | 中国航天科工集团第三研究院第八三五七研究所 | Multi-camera visual perception system for UGV (Unmanned Ground Vehicle) |
CN103996017A (en) * | 2014-02-24 | 2014-08-20 | 航天恒星科技有限公司 | Ship detection method based on Hu invariant moment and support vector machine |
CN104021378A (en) * | 2014-06-07 | 2014-09-03 | 北京联合大学 | Real-time traffic light recognition method based on space-time correlation and priori knowledge |
CN104574960A (en) * | 2014-12-25 | 2015-04-29 | 宁波中国科学院信息技术应用研究院 | Traffic light recognition method |
CN104766046A (en) * | 2015-02-06 | 2015-07-08 | 哈尔滨工业大学深圳研究生院 | Detection and recognition algorithm conducted by means of traffic sign color and shape features |
CN104791113A (en) * | 2015-03-20 | 2015-07-22 | 武汉理工大学 | Automatic engine start and stop intelligent trigger method and system based on driving road condition |
CN104851288A (en) * | 2015-04-16 | 2015-08-19 | 宁波中国科学院信息技术应用研究院 | Traffic light positioning method |
US9442487B1 (en) * | 2014-08-15 | 2016-09-13 | Google Inc. | Classifier hierarchies for traffic light and traffic indicator detection |
CN106023623A (en) * | 2016-07-28 | 2016-10-12 | 南京理工大学 | Recognition and early warning method of vehicle-borne traffic signal and symbol based on machine vision |
-
2016
- 2016-12-26 CN CN201611214259.1A patent/CN106803064B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011216051A (en) * | 2010-04-02 | 2011-10-27 | Institute Of National Colleges Of Technology Japan | Program and device for discriminating traffic light |
CN102819263A (en) * | 2012-07-30 | 2012-12-12 | 中国航天科工集团第三研究院第八三五七研究所 | Multi-camera visual perception system for UGV (Unmanned Ground Vehicle) |
CN103996017A (en) * | 2014-02-24 | 2014-08-20 | 航天恒星科技有限公司 | Ship detection method based on Hu invariant moment and support vector machine |
CN104021378A (en) * | 2014-06-07 | 2014-09-03 | 北京联合大学 | Real-time traffic light recognition method based on space-time correlation and priori knowledge |
US9442487B1 (en) * | 2014-08-15 | 2016-09-13 | Google Inc. | Classifier hierarchies for traffic light and traffic indicator detection |
CN104574960A (en) * | 2014-12-25 | 2015-04-29 | 宁波中国科学院信息技术应用研究院 | Traffic light recognition method |
CN104766046A (en) * | 2015-02-06 | 2015-07-08 | 哈尔滨工业大学深圳研究生院 | Detection and recognition algorithm conducted by means of traffic sign color and shape features |
CN104791113A (en) * | 2015-03-20 | 2015-07-22 | 武汉理工大学 | Automatic engine start and stop intelligent trigger method and system based on driving road condition |
CN104851288A (en) * | 2015-04-16 | 2015-08-19 | 宁波中国科学院信息技术应用研究院 | Traffic light positioning method |
CN106023623A (en) * | 2016-07-28 | 2016-10-12 | 南京理工大学 | Recognition and early warning method of vehicle-borne traffic signal and symbol based on machine vision |
Non-Patent Citations (2)
Title |
---|
SATHIYA 等: "Real time recognition of traffic light and their signal count-down timings", 《INTERNATIONAL CONFERENCE ON INFORMATION COMMUNICATION & EMBEDDED SYSTEMS IEEE》 * |
李广亮 等: "智能车交通灯识别", 《杭州电子科技大学学报》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107563301A (en) * | 2017-08-09 | 2018-01-09 | 上海炬宏信息技术有限公司 | Red signal detection method based on image processing techniques |
CN109635640A (en) * | 2018-10-31 | 2019-04-16 | 百度在线网络技术(北京)有限公司 | Traffic light recognition method, device, equipment and storage medium based on cloud |
CN110021176A (en) * | 2018-12-21 | 2019-07-16 | 文远知行有限公司 | Traffic lights decision-making technique, device, computer equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN106803064B (en) | 2020-05-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107798335B (en) | Vehicle logo identification method fusing sliding window and Faster R-CNN convolutional neural network | |
CN103984950B (en) | A kind of moving vehicle brake light status recognition methods for adapting to detection on daytime | |
CN103971128B (en) | A kind of traffic sign recognition method towards automatic driving car | |
CN104021378B (en) | Traffic lights real-time identification method based on space time correlation Yu priori | |
CN109190523B (en) | Vehicle detection tracking early warning method based on vision | |
CN107590470B (en) | Lane line detection method and device | |
KR102227843B1 (en) | Operating method of lane departure warning system | |
CN109299674B (en) | Tunnel illegal lane change detection method based on car lamp | |
CN102298693B (en) | Expressway bend detection method based on computer vision | |
CN107316486A (en) | Pilotless automobile visual identifying system based on dual camera | |
CN104966049B (en) | Lorry detection method based on image | |
CN109344704B (en) | Vehicle lane change behavior detection method based on included angle between driving direction and lane line | |
CN106169244A (en) | The guidance information utilizing crossing recognition result provides device and method | |
CN103927548B (en) | Novel vehicle collision avoiding brake behavior detection method | |
CN103345766A (en) | Method and device for identifying signal light | |
CN109740484A (en) | The method, apparatus and system of road barrier identification | |
CN104700072A (en) | Lane line historical frame recognition method | |
CN103984917A (en) | Multi-feature nighttime vehicle detection method based on machine vision | |
CN108446668A (en) | Traffic lights detection recognition method and system based on unmanned platform | |
CN106803064A (en) | A kind of traffic lights method for quickly identifying | |
CN101369312B (en) | Method and equipment for detecting intersection in image | |
CN106407951A (en) | Monocular vision-based nighttime front vehicle detection method | |
CN106022242B (en) | Driver's making and receiving calls recognition methods in intelligent transportation system | |
CN107025796A (en) | Automobile assistant driving vision early warning system and its method for early warning | |
Devane et al. | Lane detection techniques using image processing |
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 |