CN107480592A - A kind of multilane detection method and tracking - Google Patents
A kind of multilane detection method and tracking Download PDFInfo
- Publication number
- CN107480592A CN107480592A CN201710568932.XA CN201710568932A CN107480592A CN 107480592 A CN107480592 A CN 107480592A CN 201710568932 A CN201710568932 A CN 201710568932A CN 107480592 A CN107480592 A CN 107480592A
- Authority
- CN
- China
- Prior art keywords
- line
- straight line
- lane
- lane line
- article
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 claims abstract description 25
- 230000009466 transformation Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 description 7
- 238000005259 measurement Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 229940074869 marquis Drugs 0.000 description 1
- VBUNOIXRZNJNAD-UHFFFAOYSA-N ponazuril Chemical compound CC1=CC(N2C(N(C)C(=O)NC2=O)=O)=CC=C1OC1=CC=C(S(=O)(=O)C(F)(F)F)C=C1 VBUNOIXRZNJNAD-UHFFFAOYSA-N 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000002922 simulated annealing 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)
- Traffic Control Systems (AREA)
Abstract
The present invention, which provides a kind of multilane detection method and tracking, the detection method, to be included：And the candidate region in image is determined, first lane line of left and right side of vehicle in candidate region is then detected using bicycle road line detecting method；A Chosen Point is selected immediately below end point, the Chosen Point is crossed and makees a horizontal line, the horizontal line intersects at one first intersection point and one second intersection point respectively with first lane line of left and right side；Judge that the left and right side of vehicle whether there is adjacent lane, if there is adjacent lane in left and right side, then determine each candidate straight line of left and right side Article 2 lane line, the distance between and calculate each candidate straight line of first and second intersection point and left and right side Article 2 lane line, and by the straight line cluster of distance each candidate rectilinear(-al) left and right side Article 2 lane line within a predetermined range, then each bar fitting a straight line left and right side Article 2 lane line in the straight line cluster.The present invention can quickly and easily position lane line, and computation complexity is low, and validity is high.
Description
Technical field
The present invention relates to image detecting method, more particularly to a kind of multilane detection method and tracking.
Background technology
The basic skills of lane detection is the result according to Image Edge-Detection, judges which marginal point belongs to same
Straight line.For the lane line on highway, near-end track can be characterized with straight line.Meanwhile as intelligent transportation environment sense
The part known, the detection of multilane can be to automatic Pilot auxiliary, intelligent driving and unmanned offer subenvironment driving
The support data of route planning.
At present, detecting the method for current rectilinear stretch has Hough transformation, beeline approaching, the curve based on simulated annealing
The methods of detection, correlation detection, inverse perspective are hinted obliquely at.In these methods, Hough transformation is due to can effectively make up straight line
The missing of upper characteristic point, it is used widely so as to improve the accuracy rate of lane detection.
Because the related algorithms of current ADAS (advanced drive assist system) focus mainly on current lane, and to multilane
Lane detection research it is very few, however as reaching its maturity for automatic Pilot technology, progressively perfecting for relevant laws and regulations is more
The detection in track is all the more important to the path planning of automatic driving vehicle.
The content of the invention
For above-mentioned the deficiencies in the prior art, it is an object of the invention to provide a kind of multilane based on forward sight camera
Detection method and tracking, accurately to detect and track a plurality of lane line, supported for unmanned offer data.
To achieve these goals, one aspect of the present invention provides a kind of multilane detection method, and this method includes following step
Suddenly：
Step S1, shoots the image of vehicle front, and determines the candidate region in image, is then detected using bicycle diatom
Method detects first lane line of left and right side of vehicle in the candidate region；
Step S2, a Chosen Point is selected immediately below end point, crosses the Chosen Point and make a horizontal line, the horizontal line with it is left,
First, right side lane line intersects at one first intersection point and one second intersection point respectively；
Step S3, judge that the left and right side of vehicle whether there is adjacent lane, if left side has adjacent lane, it is determined that left
Each candidate straight line of side Article 2 lane line, and calculate first intersection point and left side Article 2 lane line each candidate straight line it
Between distance, and by the straight line cluster of Article 2 lane line on the left of distance each candidate rectilinear(-al) within a predetermined range, then root
According to Article 2 lane line on the left of each bar fitting a straight line in the straight line cluster；Right side Article 2 is determined if right side has adjacent lane
Each candidate straight line of lane line, and calculate between second intersection point and each candidate straight line of right side Article 2 lane line away from
From, and the straight line cluster of each candidate rectilinear(-al) right side Article 2 lane line by distance within a predetermined range, it is then straight according to this
Article 2 lane line on the right side of each bar fitting a straight line in line cluster.
Further, the bicycle road line detecting method in the step S1 comprises the following steps：
Step S11, the image of vehicle front is read in, and described image is pre-processed；
Step S12, the straight line in described image is detected, and be that positive straight line is straight as left side using the slope of vehicle left side
Line, the slope of vehicle right side are used as right side straight line for negative straight line；
Step S13, take the datum mark immediately below end point, calculate the datum mark and each left side straight line and right side straight line it
Between spacing, and by the straight line cluster of first lane line on the left of each left side rectilinear(-al) of spacing within a predetermined range, by spacing
The straight line cluster of first lane line on the right side of each right side rectilinear(-al) within a predetermined range；
Step S14, based on the left and right side straight line nearest apart from the datum mark, with first lane line of left and right side
Straight line cluster in other straight lines for auxiliary, be fitted described first lane line of left and right side.
Further, the step S12 is using the straight line in Hough transformation detection described image.
Further, in the step S3, each candidate straight line of left side Article 2 lane line is determined as follows：
In first side of left side the left side detected slope of lane line for just and less than first side of left side lane line slope it is straight
Line, and select the angle between described first lane line in left side therefrom and be used as left side second more than the straight line of predetermined angular
The candidate lane line of bar lane line.
Further, in the step S3, each candidate straight line of right side Article 2 lane line is determined as follows：
In first side of right side, the right side detected slope of lane line is the straight of slope negative and more than first side of right side lane line
Line, and select the angle between described first lane line in right side therefrom and be used as right side second more than the straight line of predetermined angular
The candidate lane line of bar lane line.
Further, in the step S3, left side Article 2 lane line is fitted as follows：From the left side
All straight lines that end point is intersected at described first lane line of left and right side, and root are chosen in the straight line cluster of two lane lines
According to left side Article 2 lane line described in the fitting a straight line of selection.
Further, in the step S4, right side Article 2 lane line is fitted as follows：From the right side
All straight lines that end point is intersected at described first lane line of left and right side, and root are chosen in the straight line cluster of two lane lines
According to right side Article 2 lane line described in the fitting a straight line of selection.
Another aspect of the present invention provides a kind of multilane tracking, and this method is to any one of preceding claims 1-7
Each bar lane line that described multilane detection method is detected is tracked, to predict each bar lane line in the position of subsequent time
Put.
Further, multilane tracking, it is characterised in that this method uses Kalman filter or particle filter
Each bar lane line is tracked.
In summary, the multilane detection in the present invention and tracking can quickly and easily position lane line, calculate
Complexity is low, and validity is high, can fast positioning and prediction lane line scope, resources occupation rate is low, suitable for hardware resource and
The limited vehicle-mounted data processing software of disposal ability.
Brief description of the drawings
Fig. 1 is that one-lane lane line is fitted schematic diagram；
Fig. 2 is the principle schematic that left and right Article 2 track is determined in the present invention；
Fig. 3 A-3C are schematic diagram when vehicle is located at different tracks；
Fig. 4 A-4C are the update status of each lane line buffer in the case of three kinds.
Embodiment
To make the present invention more obvious understandable, hereby with preferred embodiment, and accompanying drawing is coordinated to be described in detail below.
A kind of multilane detection method of the present invention comprises the following steps：
Step S1, shoots the image of vehicle front, and determines the candidate region in image, is then detected using bicycle diatom
First lane line of left and right side of vehicle in method detection candidate region.Wherein, bicycle road line detecting method is as follows
Realize：Step S11, the image of vehicle front is read in, and image is pre-processed；Step S12, detected and schemed using Hough transformation
Straight line as in, and be positive straight line straight line as on the left of using the slope of vehicle left side, the slope of vehicle right side is negative straight line
As right side straight line；Step S13, determines end point, takes the datum mark immediately below end point, calculates the datum mark and each left side
Spacing between straight line and right side straight line, and first lane line in each left side rectilinear(-al) left side by spacing within a predetermined range
Straight line cluster, by the straight line cluster of first lane line on the right side of each right side rectilinear(-al) of spacing within a predetermined range；Step S14,
Based on the left and right side straight line nearest apart from datum mark, with other straight lines in the straight line cluster of first lane line of left and right side
For auxiliary, first lane line (as shown in Figure 1) of left and right side is fitted.
Step S2, as shown in Fig. 2 selecting a Chosen Point immediately below end point, cross the Chosen Point and make a horizontal line, the level
Line intersects at one first intersection point PL and one second intersection point PR respectively with first lane line of left and right side.
Because vehicle on road when driving, it is impacted it is most be vehicle and road from current lane and adjacent lane
Condition, therefore, Multi-lane Lines Detection of the invention at most only consider the detection of four lane lines.When current lane is middle lane,
Each two lane lines of left and right vehicle wheel are detected, as shown in Figure 3A；When leftmost side track of the vehicle in road, current lane is only detected
With right adjacent track, when rightmost side track of the vehicle in road, only detect current lane and its left adjacent track, respectively such as Fig. 3 B and
Shown in Fig. 3 C.
Therefore, in step s3, it is necessary first to judge that the left and right side of vehicle whether there is adjacent lane, if left side is present
Adjacent lane, it is determined that each candidate straight line of left side Article 2 lane line, and calculate the first intersection point PL and left side Article 2 track
The distance between each candidate straight line of line, and each candidate rectilinear(-al) left side Article 2 lane line by distance within a predetermined range
Straight line cluster, then Article 2 lane line on the left of each bar fitting a straight line in the straight line cluster；If there is adjacent lane in right side
Each candidate straight line of right side Article 2 lane line is then determined, and calculates the second intersection point PR and right side Article 2 lane line each candidate
The distance between straight line, and the straight line cluster of each candidate rectilinear(-al) right side Article 2 lane line by distance within a predetermined range,
Then Article 2 lane line on the right side of each bar fitting a straight line in the straight line cluster.
Wherein, in step s3, each candidate straight line of left side Article 2 lane line is determined as follows：In left side
The left side detected slope of side lane line for just and less than the first side of left side lane line slope straight line, and select therefrom with
Angle between the lane line of first, left side is more than marquis of the straight line of predetermined angular (such as 10 degree) as left side Article 2 lane line
Select lane line.Meanwhile each candidate straight line of right side Article 2 lane line is determined as follows：In the first side of right side lane line
Right side detected slope be the straight line of slope negative and more than the first side of right side lane line, and select therefrom and first, right side
Angle between lane line is more than candidate lane line of the straight line of predetermined angular (such as 10 degree) as right side Article 2 lane line.
In addition, in step s3, left side Article 2 lane line is fitted as follows：From left side Article 2 lane line
All straight lines that end point is intersected at first lane line of left and right side are chosen in straight line cluster, and according to the fitting a straight line of selection
Left side Article 2 lane line.Meanwhile right side Article 2 lane line is fitted as follows：From the straight of right side Article 2 lane line
All straight lines that end point is intersected at first lane line of left and right side are chosen in line cluster, and it is right according to the fitting a straight line of selection
Side Article 2 lane line.
Another aspect of the present invention provides a kind of multilane tracking, the party on the basis of foregoing multilane detection method
Each bar lane line that method is detected using Kalman filter or particle filter to foregoing multilane detection method is tracked,
To predict each bar lane line in the position of subsequent time, consequently facilitating the area-of-interest of subsequent time image is determined, to choose
Straight line in the area-of-interest of subsequent time image provides foundation as candidate lane for lane detection.
Underneath with exemplified by Kalman filter come explain lane line tracking principle：
Xp=A*X (1)
In formula (1), A is state-transition matrix, and institute's containing parameter is the speed of Current vehicle, turns to angle information；X is current
State matrix, its component include the slope and intercept of lane line.Xp is predicted value.
Pp=A*P*At+Q (2)
In formula (2), P is process noise, and Pp value is the predicted value of process error and the sum of noise, is referred to as process and misses
Difference.
K=Pp/ (Pp+R) (3)
Formula (3) is used to calculate kalman gain, and Pp represents process error, and R represents measurement error, and K meaning is, weighs
The ratio that process error accounts for global error (process error+measurement error) is much.The result of tracking is represented with formula (4)：
Xk+1=Xp+K* (Z-Xp) (4)
In formula (4), Z is measurement result, it is meant that tracking result be predicted value and correction value and.Wherein：
Correction value=kalman gain * (measured value-predicted value) (5)
After tracking result obtains amendment, process noise is also corrected：
Pk+1=(I-K) * Pp (6).
In addition, the present invention devises a buffer, renewals of the Buffer in different conditions for each lane line
Situation is different, and Fig. 4 A-4C illustrate the update status of buffer in the case of three kinds.
Fig. 4 A are the update status of buffer under normally travel state.In Tk‐1At the moment, export Lk‐1, as TkMoment track
The predicted value of line parameter；In TkAt the moment, by lane detection and then secondary utilization track with new algorithm, export LkTied as tracking
Fruit, as Tk+1The predicted value of the lane line parameter at moment.
Fig. 4 B are to detect the update status of buffer under effective track state.In Tk‐1At the moment, export Lk‐1, as Tk
The predicted value of moment lane line parameter；In TkMoment, and to detect lane line, therefore Lane tracking is also come to nothing, herein
In the case of, still by Lk‐1As TkThe tracking result at moment, as Tk+1The predicted value of the lane line parameter at moment.
Fig. 4 C are the situations that buffer is emptied because of lane change.If detecting that vehicle is in during lane change, empty
Buffer, predicted value is implanted into buffer again after lane line is normally detected.
It the above is only some embodiments of the present invention, it is noted that come for those skilled in the art
Say, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should be regarded as
Protection scope of the present invention.
Claims (9)
1. a kind of multilane detection method, it is characterised in that this method comprises the following steps：
Step S1, shoots the image of vehicle front, and determines the candidate region in image, then utilizes bicycle road line detecting method
Detect first lane line of left and right side of vehicle in the candidate region；
Step S2, a Chosen Point is selected immediately below end point, cross the Chosen Point and make a horizontal line, the horizontal line and left and right side
First lane line intersects at one first intersection point and one second intersection point respectively；
Step S3, judge that the left and right side of vehicle whether there is adjacent lane, if left side has adjacent lane, it is determined that left side
Each candidate straight line of two lane lines, and between calculating first intersection point and each candidate straight line of left side Article 2 lane line
Distance, and the straight line cluster of each candidate rectilinear(-al) left side Article 2 lane line by distance within a predetermined range, then basis should
Article 2 lane line on the left of each bar fitting a straight line in straight line cluster；Right side Article 2 track is determined if right side has adjacent lane
The distance between each candidate straight line of line, and calculate each candidate straight line of second intersection point and right side Article 2 lane line, and
By the straight line cluster of Article 2 lane line on the right side of distance each candidate rectilinear(-al) within a predetermined range, then according in the straight line cluster
Each bar fitting a straight line on the right side of Article 2 lane line.
2. multilane detection method according to claim 1, it is characterised in that the bicycle diatom detection in the step S1
Method comprises the following steps：
Step S11, the image of vehicle front is read in, and described image is pre-processed；
Step S12, the straight line in described image is detected, and be positive straight line straight line, car as on the left of using the slope of vehicle left side
Slope on the right side of is used as right side straight line for negative straight line；
Step S13, the datum mark immediately below end point is taken, calculated between the datum mark and each left side straight line and right side straight line
Spacing, and the straight line cluster of first lane line in each left side rectilinear(-al) left side by spacing within a predetermined range, by spacing pre-
Determine the straight line cluster of first lane line on the right side of each right side rectilinear(-al) in scope；
Step S14, based on the left and right side straight line nearest apart from the datum mark, with the straight of first lane line of left and right side
Other straight lines in line cluster are auxiliary, are fitted described first lane line of left and right side.
3. multilane detection method according to claim 2, it is characterised in that the step S12 is examined using Hough transformation
The straight line surveyed in described image.
4. multilane detection method according to claim 1, it is characterised in that in the step S3, by walking as follows
The rapid each candidate straight line for determining left side Article 2 lane line：In first side of left side the left side detected slope of lane line for just and
Less than the straight line of the slope of first side of left side lane line, and select therefrom between described first lane line in left side
Angle is more than candidate lane line of the straight line of predetermined angular as left side Article 2 lane line.
5. multilane detection method according to claim 1, it is characterised in that in the step S3, by walking as follows
The rapid each candidate straight line for determining right side Article 2 lane line：In first side of right side the right side detected slope of lane line to be negative and
More than the straight line of the slope of first side of right side lane line, and select therefrom between described first lane line in right side
Angle is more than candidate lane line of the straight line of predetermined angular as right side Article 2 lane line.
6. multilane detection method according to claim 1, it is characterised in that in the step S3, by walking as follows
Rapid fitting left side Article 2 lane line：All and left and right side is chosen from the straight line cluster of the left side Article 2 lane line
First lane line intersects at the straight line of end point, and the left side Article 2 lane line according to the fitting a straight line of selection.
7. multilane detection method according to claim 1, it is characterised in that in the step S4, by walking as follows
Rapid fitting right side Article 2 lane line：All and left and right side is chosen from the straight line cluster of the right side Article 2 lane line
First lane line intersects at the straight line of end point, and the right side Article 2 lane line according to the fitting a straight line of selection.
8. a kind of multilane tracking, it is characterised in that this method is to more cars any one of preceding claims 1-7
Each bar lane line that road detection method is detected is tracked, to predict each bar lane line in the position of subsequent time.
9. multilane tracking according to claim 8, it is characterised in that this method uses Kalman filter or grain
Subfilter is tracked to each bar lane line.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710568932.XA CN107480592B (en) | 2017-07-13 | 2017-07-13 | Multi-lane detection method and tracking method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710568932.XA CN107480592B (en) | 2017-07-13 | 2017-07-13 | Multi-lane detection method and tracking method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107480592A true CN107480592A (en) | 2017-12-15 |
| CN107480592B CN107480592B (en) | 2020-06-12 |
Family
ID=60595584
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710568932.XA Active CN107480592B (en) | 2017-07-13 | 2017-07-13 | Multi-lane detection method and tracking method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107480592B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108961146A (en) * | 2018-07-19 | 2018-12-07 | 深圳地平线机器人科技有限公司 | The method and apparatus of rendering perception map |
| 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 |
| CN110967025A (en) * | 2018-09-30 | 2020-04-07 | 长城汽车股份有限公司 | Lane line screening method and system |
| CN113254563A (en) * | 2021-06-18 | 2021-08-13 | 智道网联科技(北京)有限公司 | Road number generation method and related device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101837780A (en) * | 2009-03-18 | 2010-09-22 | 现代自动车株式会社 | A lane departure warning system using a virtual lane and a system according to the same |
| CN102324017A (en) * | 2011-06-09 | 2012-01-18 | 中国人民解放军国防科学技术大学 | FPGA (Field Programmable Gate Array)-based lane line detection method |
| CN103440649A (en) * | 2013-08-23 | 2013-12-11 | 安科智慧城市技术（中国）有限公司 | Detection method and device for lane boundary line |
| CN106529443A (en) * | 2016-11-03 | 2017-03-22 | 温州大学 | Method for improving detection of lane based on Hough transform |
-
2017
- 2017-07-13 CN CN201710568932.XA patent/CN107480592B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101837780A (en) * | 2009-03-18 | 2010-09-22 | 现代自动车株式会社 | A lane departure warning system using a virtual lane and a system according to the same |
| CN102324017A (en) * | 2011-06-09 | 2012-01-18 | 中国人民解放军国防科学技术大学 | FPGA (Field Programmable Gate Array)-based lane line detection method |
| CN103440649A (en) * | 2013-08-23 | 2013-12-11 | 安科智慧城市技术（中国）有限公司 | Detection method and device for lane boundary line |
| CN106529443A (en) * | 2016-11-03 | 2017-03-22 | 温州大学 | Method for improving detection of lane based on Hough transform |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108961146A (en) * | 2018-07-19 | 2018-12-07 | 深圳地平线机器人科技有限公司 | The method and apparatus of rendering perception map |
| 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 |
| WO2020029667A1 (en) * | 2018-08-09 | 2020-02-13 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for lane line identification |
| 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 |
| US11335102B2 (en) | 2018-08-09 | 2022-05-17 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for lane line identification |
| CN110967025A (en) * | 2018-09-30 | 2020-04-07 | 长城汽车股份有限公司 | Lane line screening method and system |
| CN113254563A (en) * | 2021-06-18 | 2021-08-13 | 智道网联科技(北京)有限公司 | Road number generation method and related device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107480592B (en) | 2020-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107480592A (en) | A kind of multilane detection method and tracking | |
| JP6978491B2 (en) | Image processing methods for recognizing ground markings, and systems for detecting ground markings | |
| CN105612569B (en) | Parking vehicle detection apparatus, vehicle management system and control method | |
| US10081308B2 (en) | Image-based vehicle detection and distance measuring method and apparatus | |
| Huang et al. | Finding multiple lanes in urban road networks with vision and lidar | |
| CN110287905B (en) | Deep learning-based real-time traffic jam area detection method | |
| US8055445B2 (en) | Probabilistic lane assignment method | |
| RU2735567C1 (en) | Method for storing movement backgrounds, method for generating motion path model, method for estimating local position and storage device for storing movement backgrounds | |
| CN107798724A (en) | Automated vehicle 3D road models and lane markings define system | |
| US20120314070A1 (en) | Lane sensing enhancement through object vehicle information for lane centering/keeping | |
| CN106056100A (en) | Vehicle auxiliary positioning method based on lane detection and object tracking | |
| CN104183127A (en) | Traffic surveillance video detection method and device | |
| CN110530372A (en) | Localization method, determining method of path, device, robot and storage medium | |
| CN104318258A (en) | Time domain fuzzy and kalman filter-based lane detection method | |
| CN104751151A (en) | Method for identifying and tracing multiple lanes in real time | |
| CN112577526B (en) | Confidence calculating method and system for multi-sensor fusion positioning | |
| CN107901909A (en) | Control method and device for automatic lane replacement and controller | |
| CN104677361B (en) | A kind of method of comprehensive location | |
| CN103582907A (en) | Device for determining road profile, onboard image-recognition device, device for adjusting image-capturing axis, and lane-recognition method. | |
| CN105261034A (en) | Method and device for calculating traffic flow on highway | |
| CN110379168A (en) | A kind of vehicular traffic information acquisition method based on Mask R-CNN | |
| CN103730015A (en) | Method and device for detecting traffic flow at intersection | |
| US20200180646A1 (en) | Sensor fusion target prediction device and method for vehicles and vehicle including the device | |
| CN108458746A (en) | One kind being based on sensor method for self-adaption amalgamation | |
| CN109709944A (en) | A kind of generation method in enter the station method and its path of entering the station of automatic Pilot bus |
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 |