CN105488485A - Lane line automatic extraction method based on vehicle trajectory - Google Patents
Lane line automatic extraction method based on vehicle trajectory Download PDFInfo
- Publication number
- CN105488485A CN105488485A CN201510890424.4A CN201510890424A CN105488485A CN 105488485 A CN105488485 A CN 105488485A CN 201510890424 A CN201510890424 A CN 201510890424A CN 105488485 A CN105488485 A CN 105488485A
- Authority
- CN
- China
- Prior art keywords
- track
- vehicle
- lane line
- image
- extraction method
- 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
- 238000000605 extraction Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 claims abstract description 34
- 239000000284 extract Substances 0.000 claims abstract description 22
- 230000003628 erosive effect Effects 0.000 claims abstract description 7
- 230000000877 morphologic effect Effects 0.000 claims abstract description 6
- 238000005260 corrosion Methods 0.000 claims description 7
- 230000007797 corrosion Effects 0.000 claims description 7
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000010339 dilation Effects 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000012634 fragment Substances 0.000 description 3
- 239000003518 caustics Substances 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 238000007598 dipping method Methods 0.000 description 2
- 239000003550 marker Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 238000010276 construction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012946 outsourcing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
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
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a lane line automatic extraction method based on a vehicle trajectory. The method comprises the following steps: step 1, extracting the vehicle trajectory; step 2, creating a vehicle trajectory image; step 3, eliminating trajectory adhesion between different lanes based on morphological close operation and erosion; and step 4, extracting a lane line by a minimum enclosing rectangle method. The lane line automatic extraction method based on the vehicle trajectory provided by the invention does not depend on a road marking, therefore, the lane line automatic extraction method is also applicable to road areas without markings or the markings are fuzzy; the invention can extract the lane lines of multiple lanes simultaneously and the efficiency is high; besides, an unmanned aerial vehicle has an advantage of being mobile, therefore, the unmanned aerial vehicle can be used for performing lane line extraction on road sections without traffic monitoring equipment.
Description
Technical field
The invention belongs to technical field of image processing, relate to a kind of method automatically extracting lane line based on track of vehicle, the present invention is intended to introduce a kind of blanket lane line extracting method, pavement marker is not existed or lane line under ambiguity to extract problem all applicable.
Background technology
Lane line builds the important composition composition of road conditions, and extracting lane line the traffic surveillance and control system of robotization is important function.Lane line extracts traffic study, as driving behavior detect, the method for the road network of classification (change, deviation etc.) and bicycle road level necessity of extracting that to be important be also.
At present, the lane line extracting method based on video is that such mode has some deficiency: 1) do not exist or fuzzy section roadmarking, cannot extract lane line by identifying that roadmarking completes lane line and extracts; 2) because the viewing angle problem of camera causes the lane line of extraction to have comparatively big error at far-end; 3) for the region not having monitoring camera, lane line cannot be realized and extract.Another kind of Lane detection method is based on GPS locator data, its principle carries GPS positioning equipment to cruise along road-center, then extract lane line based on gps data, the weak point of the method is that the positioning error of gps data is comparatively large, and therefore Precision in Roadway Recognition is low.
Summary of the invention
For the deficiency of existing lane line extracting method, the present invention proposes the lane line extracting method based on track of vehicle, adopt track of vehicle reverse push guide-car diatom.The present invention is the lane line extracting method based on track of vehicle, first to take photo by plane video extraction track of vehicle based on unmanned plane low latitude, then track of vehicle image is built based on track of vehicle, the track adhesion of changing between different tracks that track of vehicle causes is removed in applied morphology closed operation, caustic solution, then the minimum enclosed rectangle of track of vehicle on every bar track is calculated, the center line of this minimum enclosed rectangle is the center line in track, namely achieves lane line and extracts.
Lane line extraction method of the present invention, to take photo by plane video extraction track of vehicle based on unmanned plane, then based on vehicle operating track reverse push guide-car diatom, do not rely on road conditions (with or without pavement marker or whether fuzzy), there is general applicability.
The present invention enters a little with brand-new research, for lane line propose a kind of can the blanket lane line extraction method based on track of vehicle, realized by following step:
Step 1: track of vehicle extracts;
Based on the low latitude Aerial Images of unmanned plane, carry out track of vehicle extraction, extract result for obtaining best lane line, the track of vehicle number that every bar track is extracted is no less than 20, and namely this track needs at least 20 cars to pass through.Track extraction method can be manual extraction, also can based on existing automatic orbit extracting method such as particle filter, KLT feature point tracking or TLD, for building track of vehicle image in next step.
Step 2: track of vehicle image creation;
Create trace image with the track of vehicle extracted based on unmanned plane Aerial Images in step, track of vehicle raw data is discrete point, needs track of vehicle point to connect into curve according to the sequencing extracted here in the process creating track.The image creation of trace image divides following three steps: the blank image 1) creating white background, and the Aerial Images of image size and extraction track etc. are large; 2) by track of vehicle Drawing of Curve to blank image.Because track of vehicle is vector, there is not width problem, but track drafting considers this live width factor to the process need of image, the live width of track of vehicle gives tacit consent to 1/6 of a transverse width of picking up the car here, and lateral direction of car width needs hand dipping to obtain;
Step 3: eliminate track adhesion between different track from corrosion based on closing operation of mathematical morphology;
The track of many cars, after coincidence, can form coincidence district, but can there is gap between track, adopts closing operation of mathematical morphology method to fill track space, to form complete connected region here.Vehicle lane-changing can cause track adhesion between different track in trace image, affect the accuracy that follow-up lane line extracts, therefore need to eliminate this adhesion, here morphological erosion method is adopted to process trace image, to eliminate track adhesion, wherein corrosion structure element adopts round, and radius of a circle is 1/5 of vehicle width.After Image erosion, have track fragment and exist, here by length characteristic, track fragment length being less than image level width 1/2 screens out.
Step 4: Minimum Enclosing Rectangle method extracts lane line
To the trace image having interrupted track adhesion in upper step, first calculate the minimum enclosed rectangle of track of vehicle on every bar track, then calculate the center line of minimum enclosed rectangle, namely obtain the lane line of road.
The invention has the advantages that:
(1) the present invention utilizes vehicle generally along the characteristic that track center line travels, and reverse push guide-car diatom, has method innovation;
(2) unmanned plane has moveable advantage, therefore can be used for carrying out lane line extraction to the section without traffic monitoring apparatus;
(3) the present invention extracts lane line and does not rely on roadmarking, therefore to without graticule or the fuzzy road area of graticule applicable equally;
(4) the present invention can extract the lane line of multilane simultaneously, and efficiency is high.
Accompanying drawing explanation
Fig. 1 is the track of vehicle stacking diagram of video extraction of taking photo by plane based on unmanned plane;
Fig. 2 is the trace image (bianry image) built based on track of vehicle;
Fig. 3 eliminates track adhesion figure between different track based on closing operation of mathematical morphology, corrosion;
Fig. 4 is for screen out track fragment based on track connected region length;
Fig. 5 is the minimum enclosed rectangle figure calculating track connected region;
Fig. 6 is the center line chart calculating minimum enclosed rectangle;
Fig. 7 is the lane line (Four-Lane Road) extracted;
Fig. 8 is method flow diagram of the present invention.
Embodiment
Below in conjunction with drawings and Examples, the present invention is described in further detail.
The invention provides a kind of lane line extracting method based on track of vehicle, track of vehicle used is based on video extraction of taking photo by plane.The present invention is the lane line extracting method based on track of vehicle, first to take photo by plane video extraction track of vehicle based on the low latitude of unmanned plane, then builds track of vehicle image based on track of vehicle; The track adhesion of changing between different tracks that track of vehicle causes is removed in applied morphology closed operation, caustic solution; Based on track regions length characteristic, screen out non-vehicle track regions; Then calculate the minimum enclosed rectangle of every bar track track, the center line of this minimum enclosed rectangle is the center line in track, namely achieves lane line and extracts.The above-mentioned lane line extracting method based on track of vehicle, as shown in Figure 8, concrete treatment step is as follows for flow process:
Step 1: track of vehicle extracts;
Based on the low latitude Aerial Images of unmanned plane, extract track of vehicle.Track extraction method can be manual extraction, also can based on existing automatic orbit extracting method such as particle filter, KLT feature point tracking or TLD, for building track of vehicle image in next step.Extract result for obtaining best lane line, the track of vehicle number that every bar track is extracted should be no less than 20, and namely every bar track needs at least 20 cars to pass through.The track of vehicle overlap of extraction is plotted on Aerial Images, Figure 1 shows that the Overlay figure of track of vehicle.
Step 2: track of vehicle picture construction;
Create trace image with the track of vehicle extracted based on unmanned plane Aerial Images in step, track of vehicle raw data is discrete point, needs track of vehicle point to connect into curve according to the sequencing extracted here in the process creating track.The image creation of trace image divides following three steps: the blank image 1) creating white background, and the Aerial Images of image size and extraction track etc. are large; 2) by track of vehicle Drawing of Curve to blank image.Because track of vehicle is vector, there is not width problem, but track drafting considers this live width factor to the process need of image, the live width of track of vehicle gives tacit consent to 1/6 of a transverse width of picking up the car here, and lateral direction of car width needs hand dipping to obtain.Figure 2 shows that the trace image of establishment, this figure is bianry image;
Step 3: eliminate track adhesion between different track from corrosion based on closing operation of mathematical morphology;
The track of many cars, after coincidence, can form coincidence district, but can there is gap between track, adopts closing operation of mathematical morphology method to fill track space, to form complete connected region here.Vehicle lane-changing can cause the track of vehicle adhesion in trace image between different track, as shown in Figure 2, article four, between track, there is the adhesion phenomenon of track, this can affect the accuracy that follow-up lane line extracts, therefore need to eliminate this adhesion, adopt morphological erosion method to process trace image, to eliminate track adhesion here, wherein corrosion structure element adopts circle, and radius of a circle is 1/5 of vehicle width.In trajectory diagram shown in Fig. 3, the track of Tu2Zhong Huan road vehicle is screened out, and namely eliminates track adhesion part.As shown in Fig. 3 lower right corner, the non-track regions of part is not eliminated, and needs to screen out, here based on the length characteristic of track of vehicle, track connected region length being less than image length 1/2 screens out, and in Fig. 4, display has screened out the shorter non-vehicle path portion in the lower right corner in Fig. 3.
Step 4: Minimum Enclosing Rectangle method extracts lane line
To the trace image having interrupted track adhesion in step 3, calculate the minimum enclosed rectangle of track of vehicle on every bar track, track of vehicle as shown in Figure 5 on each track is by a long rectangle outsourcing, then the center line of minimum enclosed rectangle is calculated, as shown in Figure 6, this center line is the lane line of road, namely achieves lane line and extracts, as shown in Figure 7, the lane line in four tracks is extracted in display.
Claims (4)
1., based on the lane line extraction method of track of vehicle, comprise following step:
Step 1: track of vehicle extracts;
Based on the low latitude Aerial Images of unmanned plane, carry out track of vehicle extraction;
Step 2: track of vehicle image creation;
Create and the equal-sized blank image of Aerial Images, track of vehicle raw data is discrete point, and the discrete point of track of vehicle is plotted to blank image according to the sequencing extracted, and discrete point connects into curve, form track of vehicle image, the live width of track of vehicle is set;
Step 3: eliminate track adhesion between different track based on morphological dilations corrosion;
After the track coincidence of many cars, form coincidence district, between track, there is gap, adopt closing operation of mathematical morphology to fill track space, form complete connected region; Vehicle lane-changing causes track adhesion between different track in trace image, adopts morphological erosion method to process trace image, eliminates track adhesion;
Step 4: Minimum Enclosing Rectangle method extracts lane line;
To the trace image obtained in step 3, obtain the minimum enclosed rectangle of track of vehicle on every bar track, calculate the center line of minimum enclosed rectangle, center line is the lane line of road.
2. the lane line extraction method based on track of vehicle according to claim 1, in described step 1, the track of vehicle number that every bar track is extracted is no less than 20.
3. the lane line extraction method based on track of vehicle according to claim 1, in described step 2, the live width of track of vehicle is 1/6 of lateral direction of car width.
4. the lane line extraction method based on track of vehicle according to claim 1, in described step 3, in morphological erosion method, corrosion structure element adopts round, and radius of a circle is 1/5 of vehicle width; After Image erosion, track connected region length being less than image length 1/2 screens out.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510890424.4A CN105488485B (en) | 2015-12-07 | 2015-12-07 | Lane line extraction method based on track of vehicle |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510890424.4A CN105488485B (en) | 2015-12-07 | 2015-12-07 | Lane line extraction method based on track of vehicle |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105488485A true CN105488485A (en) | 2016-04-13 |
CN105488485B CN105488485B (en) | 2019-01-22 |
Family
ID=55675457
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510890424.4A Expired - Fee Related CN105488485B (en) | 2015-12-07 | 2015-12-07 | Lane line extraction method based on track of vehicle |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105488485B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109284674A (en) * | 2018-08-09 | 2019-01-29 | 浙江大华技术股份有限公司 | A kind of method and device of determining lane line |
CN109785667A (en) * | 2019-03-11 | 2019-05-21 | 百度在线网络技术(北京)有限公司 | Deviation recognition methods, device, equipment and storage medium |
CN109871752A (en) * | 2019-01-04 | 2019-06-11 | 北京航空航天大学 | A method of lane line is extracted based on monitor video detection wagon flow |
CN110906940A (en) * | 2019-10-26 | 2020-03-24 | 武汉中海庭数据技术有限公司 | Lane sideline aggregation method based on track direction |
CN110942038A (en) * | 2019-11-29 | 2020-03-31 | 腾讯科技(深圳)有限公司 | Traffic scene recognition method, device, medium and electronic equipment based on vision |
CN111551958A (en) * | 2020-04-28 | 2020-08-18 | 北京踏歌智行科技有限公司 | Mining area unmanned high-precision map manufacturing method |
CN111738207A (en) * | 2020-07-13 | 2020-10-02 | 腾讯科技(深圳)有限公司 | Lane line detection method and device, electronic device and readable storage medium |
CN112092815A (en) * | 2020-09-02 | 2020-12-18 | 北京航空航天大学 | Vehicle track changing tracking control method based on model prediction |
CN114906152A (en) * | 2022-04-22 | 2022-08-16 | 合众新能源汽车有限公司 | Lane line construction method, lane line construction device, electronic device, and computer-readable storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103295420A (en) * | 2013-01-30 | 2013-09-11 | 吉林大学 | Method for recognizing lane line |
CN103500322A (en) * | 2013-09-10 | 2014-01-08 | 北京航空航天大学 | Automatic lane line identification method based on low-altitude aerial images |
CN103676949A (en) * | 2012-09-03 | 2014-03-26 | 现代摩比斯株式会社 | Lane keeping control system and method |
CN105005771A (en) * | 2015-07-13 | 2015-10-28 | 西安理工大学 | Method for detecting full line of lane based on optical flow point locus statistics |
EP2955664A1 (en) * | 2014-06-13 | 2015-12-16 | Fujitsu Limited | Traffic lane boundary line extraction apparatus, traffic lane boundary line extraction method, and program |
-
2015
- 2015-12-07 CN CN201510890424.4A patent/CN105488485B/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103676949A (en) * | 2012-09-03 | 2014-03-26 | 现代摩比斯株式会社 | Lane keeping control system and method |
CN103295420A (en) * | 2013-01-30 | 2013-09-11 | 吉林大学 | Method for recognizing lane line |
CN103500322A (en) * | 2013-09-10 | 2014-01-08 | 北京航空航天大学 | Automatic lane line identification method based on low-altitude aerial images |
EP2955664A1 (en) * | 2014-06-13 | 2015-12-16 | Fujitsu Limited | Traffic lane boundary line extraction apparatus, traffic lane boundary line extraction method, and program |
CN105005771A (en) * | 2015-07-13 | 2015-10-28 | 西安理工大学 | Method for detecting full line of lane based on optical flow point locus statistics |
Non-Patent Citations (1)
Title |
---|
陈漪: "基于GPS数据的城市路网立交桥识别技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109284674A (en) * | 2018-08-09 | 2019-01-29 | 浙江大华技术股份有限公司 | A kind of method and device of determining lane line |
CN109284674B (en) * | 2018-08-09 | 2020-12-08 | 浙江大华技术股份有限公司 | Method and device for determining lane line |
CN109871752A (en) * | 2019-01-04 | 2019-06-11 | 北京航空航天大学 | A method of lane line is extracted based on monitor video detection wagon flow |
CN109785667A (en) * | 2019-03-11 | 2019-05-21 | 百度在线网络技术(北京)有限公司 | Deviation recognition methods, device, equipment and storage medium |
CN110906940A (en) * | 2019-10-26 | 2020-03-24 | 武汉中海庭数据技术有限公司 | Lane sideline aggregation method based on track direction |
CN110942038A (en) * | 2019-11-29 | 2020-03-31 | 腾讯科技(深圳)有限公司 | Traffic scene recognition method, device, medium and electronic equipment based on vision |
CN111551958A (en) * | 2020-04-28 | 2020-08-18 | 北京踏歌智行科技有限公司 | Mining area unmanned high-precision map manufacturing method |
CN111551958B (en) * | 2020-04-28 | 2022-04-01 | 北京踏歌智行科技有限公司 | Mining area unmanned high-precision map manufacturing method |
CN111738207A (en) * | 2020-07-13 | 2020-10-02 | 腾讯科技(深圳)有限公司 | Lane line detection method and device, electronic device and readable storage medium |
CN112092815A (en) * | 2020-09-02 | 2020-12-18 | 北京航空航天大学 | Vehicle track changing tracking control method based on model prediction |
CN114906152A (en) * | 2022-04-22 | 2022-08-16 | 合众新能源汽车有限公司 | Lane line construction method, lane line construction device, electronic device, and computer-readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN105488485B (en) | 2019-01-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105488485A (en) | Lane line automatic extraction method based on vehicle trajectory | |
CN102622886B (en) | Video-based method for detecting violation lane-changing incident of vehicle | |
CN106297330B (en) | Reduce the method and system that pedestrian's street crossing influences plane perceptual signal control efficiency | |
CN109190481B (en) | Method and system for extracting road material of remote sensing image | |
CN102509457B (en) | Vehicle tracking method and device | |
Yang et al. | Generating lane-based intersection maps from crowdsourcing big trace data | |
Maire et al. | Obstacle-free range determination for rail track maintenance vehicles | |
CN107301776A (en) | Track road conditions processing and dissemination method based on video detection technology | |
CN103500322A (en) | Automatic lane line identification method based on low-altitude aerial images | |
CN105740809A (en) | Expressway lane line detection method based on onboard camera | |
CN112699708A (en) | Method and device for generating lane-level topology network | |
CN105513342A (en) | Video-tracking-based vehicle queuing length calculating method | |
CN105005778A (en) | Expressway vehicle detection method based on visual background extraction | |
CN102867177A (en) | People number counting method based on image grey level matching | |
CN104065920A (en) | Vehicle monitoring and tracking method, system and server | |
CN101364347A (en) | Detection method for vehicle delay control on crossing based on video | |
US20220237919A1 (en) | Method, Apparatus, and Computing Device for Lane Recognition | |
CN104504904A (en) | Mobile acquisition method of transportation facility | |
CN109241855B (en) | Intelligent vehicle travelable area detection method based on stereoscopic vision | |
CN116824859A (en) | Intelligent traffic big data analysis system based on Internet of things | |
CN110443142B (en) | Deep learning vehicle counting method based on road surface extraction and segmentation | |
Wei et al. | Damage inspection for road markings based on images with hierarchical semantic segmentation strategy and dynamic homography estimation | |
CN110765224A (en) | Processing method of electronic map, vehicle vision repositioning method and vehicle-mounted equipment | |
Nagaraj et al. | Traffic jam detection using image processing | |
CN113286081A (en) | Target identification method, device, equipment and medium for airport panoramic video |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190122 |