CN105426864A - Multiple lane line detecting method based on isometric peripheral point matching - Google Patents
Multiple lane line detecting method based on isometric peripheral point matching Download PDFInfo
- Publication number
- CN105426864A CN105426864A CN201510884340.XA CN201510884340A CN105426864A CN 105426864 A CN105426864 A CN 105426864A CN 201510884340 A CN201510884340 A CN 201510884340A CN 105426864 A CN105426864 A CN 105426864A
- Authority
- CN
- China
- Prior art keywords
- point
- lane line
- marginal
- candidate
- lane
- 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
- G06V20/588—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 invention discloses a multiple lane line detecting method based on isometric peripheral point matching. The method comprises the following steps: collecting an image of a road in front of an automobile, conducting inverse perspective projection change on the collected image of the road in front of the automobile by utilizing the interrelation of world coordinate, camera coordinate and image coordinate to obtain an aerial view of the image of the road, conducting edge detection on the aerial view, conducting pretreatment on the detected edge view, conducting local area matching on all edge points of the pretreated edge view, reserving successfully matched edge point pairs, obtaining a feature point diagram through the edge point pairs, conducting partitioning on diagram according to distribution of the feature points, extracting all feature points of one lane line I, isolating feature point areas which belong to other lane lines II, and taking the feature point areas as the candidate area of the lane line I. According to the method, the technical problems of multiple lane line detection, lane line positioning and bending, and solid lane line and dotted lane line discrimination in the conventional method can be effectively solved.
Description
Technical field
The invention belongs to technical field of computer vision, more specifically, relate to a kind of Multi-lane Lines Detection method of mating based on equidistant marginal point.
Background technology
Along with popularizing of family car, traffic safety problem receives the concern of people day by day.According to statistics, worldwide, the annual murderous number of road traffic accident is 500,000 people, and become life security one threatens greatly.
In order to improve road safety, avoid traffic accident generation, and many research institutions and company assist in driving in intelligence and carried out large quantifier elimination, and wishing can provide some supplementarys before accident occurs, and ensures traffic safety.Intelligence DAS (Driver Assistant System) adopts GPS, laser, radar usually, and the various ways such as computer vision carry out Collaboration.Wherein, adopt the intelligent householder method of computer vision, because its cost is low, the feature that extendability is strong, paying attention to by people all the more.The method, by setting up camera on automobile, is analyzed the surrounding's image information collected, be can be used to carry out route deviation early warning, controls automobile lane change and turns to.As chief component wherein, lane detection, the especially detection of Multi-lane Lines, can assist driving to provide many key messages for intelligence, seem and be even more important.
Nowadays, the lane detection of computer vision is adopted to generally involve two aspects: the 1) extraction of lane line unique point: unique point is the pixel that can represent lane line position in the picture, and the key chosen is the position that can accurately represent specific lane line.The method of extract minutiae has following several: carry out Threshold segmentation according to color intensity, and the pixel selecting brightness high is as unique point; Extract edge as unique point etc.2) selection of lane line model: lane line model has straight line model, segment line model, parabola model, clothoid model, B-spline curves model etc.The key of Model Selection is can the shape of matching lane line accurately, comprise often occur turning or shunting etc. multiple differently curved.With this curve model, matching is carried out to unique point after selected model, represent lane line.
But how many existing method for detecting lane lines deposits the problem of following several respects:
1, the detection of Multi-lane Lines cannot be carried out.The detection of Multi-lane Lines can be vehicle lane-changing, the operation such as to turn to provide auxiliary.But the lane line on a current method for detecting lane lines great majority restriction both sides, track, detection automobile place, although reduce the difficulty of lane detection like this, lost many important informations.
2, bending lane line cannot quick and precisely be located.In existing method, some employing Hough transform carry out lane detection, can only detect straight lane line like this, can only ensure the short-range testing result in front when lane line is bending; Some employing parabola models carry out matching to bending track, although parabola model can matching simply turning situation and computing velocity than very fast, being similar to track shunting, under the road sight of the complexity of rotary island etc., easily there is deviation; And adopt more complicated model, being then faced with calculated amount increases, the problem that cannot detect in real time.
3, cannot distinguish real vehicle diatom and empty lane line.In the actual driving procedure of automobile, real vehicle diatom and empty lane line represent different implications, the result contributing to improving lane detection is distinguished to it, but due to the discontinuous feature of empty lane line, sometimes be difficult to detect, even if detected, also rarely method has been distinguished it.
Summary of the invention
For above defect or the Improvement requirement of prior art, the invention provides the Multi-lane Lines Detection method of mating based on equidistant marginal point, its object is to, improve the robustness of lane detection, and solve the above-mentioned technical matters mentioned, the present invention adopts can the B-spline curves model of matching complex curve shape, the selection aspect of lane line unique point proposes a kind of edge fully utilizing lane line, direction, position relationship between lane line, the equidistant marginal point of the much informations such as lane line width is to extracting method, not only can extract the unique point on lane line that is straight or that turn fast, decrease the calculated amount of curve, and conveniently differentiation is made to solid line dotted line, Multi-lane Lines Detection in the existing method of effective solution, the bending lane line in location, distinguish real vehicle diatom and also have the technical matters run in empty lane line.
For achieving the above object, according to one aspect of the present invention, provide a kind of Multi-lane Lines Detection method of mating based on equidistant marginal point, comprise the following steps:
(1) vehicle front road image is gathered;
(2) utilize world coordinate system, mutual relationship between camera coordinates system and image coordinate system, inverse perspective projection change is done, to obtain the general view of road image to the vehicle front road image collected;
(3) rim detection is carried out to general view, and pre-service is carried out to the outline map detected,
(4) regional area coupling is carried out to marginal points all in pretreated outline map, retain the marginal point pair that the match is successful, and by these marginal points to obtaining feature point diagram;
(5) in feature point diagram, the distribution according to unique point carries out piecemeal to image, and extracts and comprise an all unique point of lane line, and isolation belongs to the region of other lane line unique points, as the candidate region of a lane line.
(6) be that solid line or dotted line are distinguished according to the right number of marginal point to what wherein comprise in each candidate region, if number is greater than a threshold value, then what show to comprise in this region is solid line, otherwise is dotted line;
(7) in the candidate region obtained in step (5), in the vertical direction, Stochastic choice unique point is as a reference mark at regular intervals, utilizes the some reference mark matching B-spline curves obtained, represents a lane line;
(8) lane line that step (7) obtains is carried out perspective change, and by the vehicle front road image that the result of perspective change projects back original, to obtain final Multi-lane Lines Detection result.
Preferably, step (3) comprises following sub-step:
(3.1) rim detection is carried out to general view, to obtain initial outline map;
(3.2) filter the outline map obtained, namely deletion direction and vertical direction angle are greater than the edge of angle threshold.
Preferably, step (4) comprises following sub-step:
(4.1) to each marginal point in pretreated outline map, fan annular region using this marginal point as the center of circle is set as region of search;
(4.2) random selecting marginal point in outline map, check whether there are other marginal points in the region of search of this marginal point, if do not had, then illustrate that this marginal point does not belong to lane line edge, by this point deletion, if there are one or more other marginal points, then choose wherein nearest one and form marginal point pair with the marginal point chosen, and preserve;
(4.3) marginal point obtained is deleted from outline map, all remaining marginal points are repeated to the operation of above-mentioned steps (4.2), until there is no vacant marginal point, and using mid point right for each marginal point as a unique point, multiple unique point constitutive characteristic point diagram.
Preferably, step (4.1) is specially, the width L of lane line in general view, can in the process calculating inverse perspective projection transformation, when pixel coordinate is changed to image coordinate system by world coordinate system, the width calculation substituting into lane line in world coordinate system in the horizontal direction obtains, and the external radius of fan ring is L+5 pixel, inside radius is L-5 pixel, and the angle of fan ring is arrive in the scope of positive 30 degree and 150 degree to 210 degree from the horizontal by negative 30 degree.
Preferably, step (5) specifically comprises following sub-step:
(5.1) in feature point diagram, carry out number to unique point in the vertical direction to add up, if cumulative number is greater than given threshold value, illustrate that may there be a lane line this position, using the horizontal ordinate of this position as a candidate point coordinate, if candidate point number is zero, then show there is no lane line in this figure.If there is multiple candidate point, then need its further checksum filter, can determine that the candidate point of an existence lane line is called location point by after filtration;
(5.2) choose the maximum candidate point of cumulative unique point number, preserve as a location point, calculate the distance between location point and the coordinate of other every other candidate points, and delete the candidate point that its middle distance is less than width threshold value;
(5.3) location point obtained in the previous step is deleted from the set of candidate point, and repeat step (5.2), until all candidate points are all disposed;
(5.4) each band of position represented by this location point extracts by the location point utilizing step (5.2) and (5.3) to determine in feature point diagram.
According to another aspect of the present invention, provide a kind of Multi-lane Lines Detection system of mating based on equidistant marginal point, comprising:
First module, for gathering vehicle front road image;
Second module, for utilizing world coordinate system, mutual relationship between camera coordinates system and image coordinate system, does inverse perspective projection change, to obtain the general view of road image to the vehicle front road image collected;
3rd module, for carrying out rim detection to general view, and carries out pre-service to the outline map detected;
Four module, for carrying out regional area coupling to marginal points all in pretreated outline map, retains the marginal point pair that the match is successful, and by these marginal points to obtaining feature point diagram;
5th module, in feature point diagram, the distribution according to unique point carries out piecemeal to image, and extracts and comprise an all unique point of lane line, and isolation belongs to the region of other lane line unique points, as the candidate region of a lane line.
6th module, for being that solid line or dotted line are distinguished according to the right number of marginal point to what wherein comprise in each candidate region, if number is greater than a threshold value, then what show to comprise in this region is solid line, otherwise is dotted line;
7th module, in the candidate region that obtains in the 5th module, in the vertical direction, Stochastic choice unique point is as a reference mark at regular intervals, utilizes the some reference mark matching B-spline curves obtained, represents a lane line;
8th module, the lane line for the 7th module being obtained carries out perspective change, and by the vehicle front road image that the result of perspective change projects back original, to obtain final Multi-lane Lines Detection result.
In general, the above technical scheme conceived by the present invention compared with prior art, can obtain following beneficial effect:
1, the present invention can solve the problem accurately cannot carrying out Multi-lane Lines Detection fast existed in existing method: first in step (2), carry out inverse perspective projection transformation to image, different lane lines is reverted to state parallel to each other, again according to the unique point obtaining lane line in step (4), last according to the method mentioned in step (5), lane line extracted region is carried out to image, thus the region of every bar lane line is accurately extracted, in each lane line region, carry out lane detection respectively, realize the detection of Multi-lane Lines.
(2) the present invention can solve the problem cannot distinguishing real vehicle diatom and empty lane line existed in existing method: owing to obtaining the unique point of every bar lane line in step (4), in step (6), according to empty lane line and real vehicle diatom, the marked difference on the number of unique point is distinguished, and therefore can solve the problem of real vehicle diatom and empty Lane detection.
(3) the present invention can solve the problem accurately cannot locating bending lane line existed in existing method: owing to first obtaining the unique point that accurately can represent lane line position in step (4), choose in step (7) wherein some unique points as reference mark matching B-spline curves, B-spline curves have the feature that can bend arbitrarily, be applicable to the expression of complicated lane line situation, we reduce the number of unique point simultaneously, decrease the complexity that B-spline curves calculate, therefore accurately and fast can locate bending lane line.
Accompanying drawing explanation
Fig. 1 is the overall flow schematic diagram of the Multi-lane Lines Detection method that the present invention is based on equidistant marginal point coupling.
Fig. 2 is marginal point coupling schematic flow sheet.
Fig. 3 is that process flow diagram is extracted in the lane line band of position.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.In addition, if below in described each embodiment of the present invention involved technical characteristic do not form conflict each other and just can mutually combine.
The present invention is divided into six parts, is respectively road image against perspective projection transformation, outline map pre-service, fixed width marginal point coupling, the extraction of the lane line band of position, the identification of solid line dotted line, curve model matching.
As shown in Figure 1, the Multi-lane Lines Detection method that the present invention is based on equidistant marginal point coupling comprises the following steps:
(1) vehicle front road image is gathered.
(2) utilize world coordinate system, mutual relationship between camera coordinates system and image coordinate system, inverse perspective projection change is done, to obtain the general view of road image to the vehicle front road image collected.
(3) carry out rim detection to general view, and carry out pre-service to the outline map detected, this step specifically comprises following sub-step:
(3.1) rim detection is carried out to general view, to obtain initial outline map; Specifically, in the present embodiment, rim detection adopts canny operator.
(3.2) filter the outline map obtained, namely deletion direction and vertical direction angle are greater than the edge of angle threshold; Preferably, angle threshold elects 60 degree as;
(4) regional area coupling is carried out to marginal points all in pretreated outline map, retain the marginal point pair that the match is successful, and by these marginal points to obtaining feature point diagram; As shown in Figure 2, this step specifically comprises following sub-step:
(4.1) to each marginal point in pretreated outline map, set with this marginal point as the center of circle, the fan annular region of one fixed width and angle is as region of search.
Specifically, in general view, owing to eliminating the impact of perspective transform, lane line shows as fixing width.If a marginal point belongs to lane line, so should find other corresponding with it marginal points in lane line width range, and angle between the two should remain on rational scope.The width L of lane line in general view, can in the process calculating inverse perspective projection transformation, and when pixel coordinate is changed to image coordinate system by world coordinate system, the width calculation substituting into lane line in world coordinate system in the horizontal direction obtains.In world coordinate system, lane line width is generally 150mm.The external radius of fan ring is L+5 pixel, and inside radius is L-5 pixel, and the angle of fan ring is elected as and arrived positive 30 degree and 150 degree to 210 degree within the scope of this from the horizontal by bearing 30 degree.
(4.2) random selecting marginal point in outline map, check whether there are other marginal points in the region of search of this marginal point, if do not had, then illustrate that this marginal point does not belong to lane line edge, by this point deletion, if there are one or more other marginal points, then choose wherein nearest one and form marginal point pair with the marginal point chosen, and preserve.
(4.3) marginal point obtained is deleted from outline map, all remaining marginal points are repeated to the operation of above-mentioned steps (4.2), until there is no vacant marginal point, and using mid point right for each marginal point as a unique point, multiple unique point constitutive characteristic point diagram.
(5) in feature point diagram, the distribution according to unique point carries out piecemeal to image, and extracts and comprise an all unique point of lane line, and isolation belongs to the region of other lane line unique points, as the candidate region of a lane line.As shown in Figure 3, this step specifically comprises following sub-step:
(5.1) in feature point diagram, number is carried out to unique point in the vertical direction and adds up, if cumulative number is greater than given threshold value, illustrate that may there be a lane line this position, using the horizontal ordinate of this position as a candidate point coordinate.If candidate point number is zero, then show there is no lane line in this figure.If there is multiple candidate point, then need its further checksum filter, can determine that the candidate point of an existence lane line is called location point by after filtration.
Specifically, in order to better inspection vehicle diatom, we need the unique point belonging to different lane line to keep apart.After inverse perspective change, lane line keeps vertical direction substantially, the horizontal ordinate of the unique point that same lane line extracts relatively is concentrated, and the horizontal ordinate of the unique point on different lane line has larger distance, and the distribution of the unique point that interference produces then compares dispersion.At the number of vertical direction statistical nature point, a peak value will be formed near lane line.Can according to this to lane line zone location.Unique point quantity threshold elects 15 to 30 as, preferably, elects 20 as.
(5.2) choose the maximum candidate point of cumulative unique point number, preserve as a location point, calculate the distance between location point and the coordinate of other every other candidate points, and delete the candidate point that its middle distance is less than width threshold value W.
Specifically, due to same lane line may produce multiple candidate point, we only need to retain maximum one of wherein accumulated amount.Checksum filter can be carried out to it according to the distance between the number of unique point cumulative on candidate point and candidate point.Width threshold value W elects 1/1 to three/8th of feature point diagram width as, and preferably, width threshold value W elects 1/5th of feature point diagram width as.
(5.3) location point obtained in the previous step is deleted from the set of candidate point, and repeat step (5.2), until all candidate points are all disposed.
(5.4) each band of position represented by this location point extracts by the location point utilizing step (5.2) and (5.3) to determine in feature point diagram; Specifically, the band of position is a rectangular area, and be highly the height of feature point diagram, width is fixed value, and the horizontal ordinate of central point is the coordinate of location point.The width putting region is 1/1 to nine/6th of feature point diagram width, preferably, is 1/8th of feature point diagram width.
(6) be that solid line or dotted line are distinguished according to the right number of marginal point to what wherein comprise in each candidate region, if number is greater than certain threshold value, then what show to comprise in this region is solid line, otherwise is dotted line.
Specifically, in actual applications, real vehicle diatom and empty lane line have different implication, distinguish to it practical application contributing to lane detection system.The span of threshold value is 20 to 40, preferably elects 30 as.
(7), in the candidate region obtained in step (5), in the vertical direction, Stochastic choice location point is as a reference mark at regular intervals, utilizes the some reference mark matching B-spline curves obtained, represents a lane line.Preferably, the span at interval is 1/1 to five/3rd of candidate region height.Preferably elect 1/4th of candidate region height as.
(8) lane line previous step obtained carries out perspective change, and by the vehicle front road image that the result of perspective change projects back original, to obtain final Multi-lane Lines Detection result.
Multi-lane Lines Detection method of the present invention, first utilizes inverse perspective mapping to obtain general view, eliminates the impact of having an X-rayed in road image.Guarantee that the information such as the direction of the position relationship between lane line, lane line, lane line width can not change because of the distance of distance camera.These information contribute to us and identify lane line, filter and verify.
For general general view, normally obtain lane line location point with the filtering of dimensional Gaussian masterplate, such method shortcoming is cannot by walkway, other traffic mark such as rotation arrow and lane line are separated, and have suppression to a certain degree to departing from the larger bending lane line of vertical direction.The present invention is directed to this point, propose a kind of marginal point matching process based on fixed width, not only can filtering interfering, and the direction of lane line can not be limited, bending lane line can be represented more accurately.
Those skilled in the art will readily understand; the foregoing is only preferred embodiment of the present invention; not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.
Claims (6)
1., based on the Multi-lane Lines Detection method that equidistant marginal point mates, it is characterized in that, comprise the following steps:
(1) vehicle front road image is gathered;
(2) utilize world coordinate system, mutual relationship between camera coordinates system and image coordinate system, inverse perspective projection change is done, to obtain the general view of road image to the vehicle front road image collected;
(3) rim detection is carried out to general view, and pre-service is carried out to the outline map detected;
(4) regional area coupling is carried out to marginal points all in pretreated outline map, retain the marginal point pair that the match is successful, and by these marginal points to obtaining feature point diagram;
(5) in feature point diagram, the distribution according to unique point carries out piecemeal to image, extracts and comprises an all unique point of lane line, and isolation belongs to the region of other lane line unique points, as the candidate region of a lane line.
(6) be that solid line or dotted line are distinguished according to the right number of marginal point to what wherein comprise in each candidate region, if number is greater than threshold value, then what show to comprise in this region is solid line, otherwise is dotted line;
(7) in the candidate region obtained in step (5), in the vertical direction, Stochastic choice unique point is as a reference mark at regular intervals, utilizes the some reference mark matching B-spline curves obtained, represents a lane line;
(8) lane line that step (7) obtains is carried out perspective change, and by the vehicle front road image that the result of perspective change projects back original, to obtain final Multi-lane Lines Detection result.
2. multilane detection method according to claim 1, is characterized in that, step (3) comprises following sub-step:
(3.1) rim detection is carried out to general view, to obtain initial outline map;
(3.2) filter the outline map obtained, namely deletion direction and vertical direction angle are greater than the edge of angle threshold.
3. multilane detection method according to claim 2, is characterized in that, step (4) comprises following sub-step:
(4.1) to each marginal point in pretreated outline map, fan annular region using this marginal point as the center of circle is set as region of search;
(4.2) random selecting marginal point in outline map, check whether there are other marginal points in the region of search of this marginal point, if do not had, then illustrate that this marginal point does not belong to lane line edge, by this point deletion, if there are one or more other marginal points, then choose wherein nearest one and form marginal point pair with the marginal point chosen, and preserve;
(4.3) marginal point obtained is deleted from outline map, all remaining marginal points are repeated to the operation of above-mentioned steps (4.2), until there is no vacant marginal point, and using mid point right for each marginal point as a unique point, multiple unique point constitutive characteristic point diagram.
4. multilane detection method according to claim 3, it is characterized in that, specifically, step (4.1) is specially, the width L of lane line in general view, can in the process calculating inverse perspective projection transformation, when pixel coordinate is changed to image coordinate system by world coordinate system, the width calculation substituting into lane line in world coordinate system in the horizontal direction obtains, the external radius of fan ring is L+5 pixel, inside radius is L-5 pixel, and the angle of fan ring is arrive in the scope of positive 30 degree and 150 degree to 210 degree from the horizontal by negative 30 degree.
5. multilane detection method according to claim 3, is characterized in that, step (5) specifically comprises following sub-step:
(5.1) in feature point diagram, carry out number to unique point in the vertical direction to add up, if cumulative number is greater than given threshold value, illustrate that may there be a lane line this position, using the horizontal ordinate of this position as a candidate point coordinate, if candidate point number is zero, then show there is no lane line in this figure.If there is multiple candidate point, then need its further checksum filter, can determine that the candidate point of an existence lane line is called location point by after filtration;
(5.2) choose the maximum candidate point of cumulative unique point number, preserve as a location point, calculate the distance between location point and the coordinate of other every other candidate points, and delete the candidate point that its middle distance is less than width threshold value;
(5.3) location point obtained in the previous step is deleted from the set of candidate point, and repeat step (5.2), until all candidate points are all disposed;
(5.4) each band of position represented by this location point extracts by the location point utilizing step (5.2) and (5.3) to determine in feature point diagram.
6., based on the Multi-lane Lines Detection system that equidistant marginal point mates, it is characterized in that, comprising:
First module, for gathering vehicle front road image;
Second module, for utilizing world coordinate system, mutual relationship between camera coordinates system and image coordinate system, does inverse perspective projection change, to obtain the general view of road image to the vehicle front road image collected;
3rd module, for carrying out rim detection to general view, and carries out pre-service to the outline map detected;
Four module, for carrying out regional area coupling to marginal points all in pretreated outline map, retains the marginal point pair that the match is successful, and by these marginal points to obtaining feature point diagram;
5th module, in feature point diagram, the distribution according to unique point carries out piecemeal to image, and extracts and comprise an all unique point of lane line, and isolation belongs to the region of other lane line unique points, as the candidate region of a lane line.
6th module, for being that solid line or dotted line are distinguished according to the right number of marginal point to what wherein comprise in each candidate region, if number is greater than a threshold value, then what show to comprise in this region is solid line, otherwise is dotted line;
7th module, in the candidate region that obtains in the 5th module, in the vertical direction, Stochastic choice unique point is as a reference mark at regular intervals, utilizes the some reference mark matching B-spline curves obtained, represents a lane line;
8th module, the lane line for the 7th module being obtained carries out perspective change, and by the vehicle front road image that the result of perspective change projects back original, to obtain final Multi-lane Lines Detection result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510884340.XA CN105426864B (en) | 2015-12-04 | 2015-12-04 | One kind being based on the matched Multi-lane Lines Detection method of equidistant marginal point |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510884340.XA CN105426864B (en) | 2015-12-04 | 2015-12-04 | One kind being based on the matched Multi-lane Lines Detection method of equidistant marginal point |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105426864A true CN105426864A (en) | 2016-03-23 |
| CN105426864B CN105426864B (en) | 2018-09-04 |
Family
ID=55505063
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510884340.XA Active CN105426864B (en) | 2015-12-04 | 2015-12-04 | One kind being based on the matched Multi-lane Lines Detection method of equidistant marginal point |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105426864B (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105930800A (en) * | 2016-04-21 | 2016-09-07 | 北京智芯原动科技有限公司 | Lane line detection method and device |
| CN106803066A (en) * | 2016-12-29 | 2017-06-06 | 广州大学 | A kind of vehicle yaw angle based on Hough transform determines method |
| CN107045629A (en) * | 2017-04-19 | 2017-08-15 | 南京理工大学 | A kind of Multi-lane Lines Detection method |
| CN107643049A (en) * | 2017-09-26 | 2018-01-30 | 沈阳理工大学 | Vehicle position detection system and method on weighbridge based on monocular structure light |
| CN107862290A (en) * | 2017-11-10 | 2018-03-30 | 智车优行科技（北京）有限公司 | Method for detecting lane lines and system |
| CN108062512A (en) * | 2017-11-22 | 2018-05-22 | 北京中科慧眼科技有限公司 | A kind of method for detecting lane lines and device |
| CN108256445A (en) * | 2017-12-29 | 2018-07-06 | 北京华航无线电测量研究所 | Method for detecting lane lines and system |
| CN107025432B (en) * | 2017-02-28 | 2018-08-21 | 合肥工业大学 | A kind of efficient lane detection tracking and system |
| CN108509906A (en) * | 2018-03-30 | 2018-09-07 | 长安大学 | Unmanned plane road baseline spot captures dual threshold Radon recognition methods |
| CN108573242A (en) * | 2018-04-26 | 2018-09-25 | 南京行车宝智能科技有限公司 | A kind of method for detecting lane lines and device |
| CN109886175A (en) * | 2019-02-13 | 2019-06-14 | 合肥思艾汽车科技有限公司 | A kind of method for detecting lane lines that straight line is combined with circular arc |
| CN110068274A (en) * | 2019-05-29 | 2019-07-30 | 易思维（杭州）科技有限公司 | Gluing sensor adhesive tape detects teaching method |
| CN110135252A (en) * | 2019-04-11 | 2019-08-16 | 长安大学 | A kind of adaptive accurate lane detection and deviation method for early warning for unmanned vehicle |
| CN110163176A (en) * | 2019-05-28 | 2019-08-23 | 北京百度网讯科技有限公司 | The recognition methods of lane line change location, device, equipment and medium |
| CN110414487A (en) * | 2019-08-16 | 2019-11-05 | 东软睿驰汽车技术（沈阳）有限公司 | Method and device for identifying lane line |
| CN110443225A (en) * | 2019-08-15 | 2019-11-12 | 安徽半问科技有限公司 | Virtual and real lane line identification method and device based on feature pixel statistics |
| CN111353466A (en) * | 2020-03-12 | 2020-06-30 | 北京百度网讯科技有限公司 | Lane line recognition processing method, lane line recognition processing device, and storage medium |
| CN111402609A (en) * | 2020-03-17 | 2020-07-10 | 北京百度网讯科技有限公司 | Special lane driving reminding method, device, equipment and storage medium |
| CN111738035A (en) * | 2019-03-25 | 2020-10-02 | 比亚迪股份有限公司 | Method, device and equipment for calculating yaw angle of vehicle |
| CN111767853A (en) * | 2020-06-30 | 2020-10-13 | 北京百度网讯科技有限公司 | Lane line detection method and device |
| WO2021116081A1 (en) | 2019-12-13 | 2021-06-17 | Connaught Electronics Ltd. | A method and system for detecting traffic lane boundaries |
| CN113221861A (en) * | 2021-07-08 | 2021-08-06 | 中移(上海)信息通信科技有限公司 | Multi-lane line detection method, device and detection equipment |
| CN113728370A (en) * | 2019-04-23 | 2021-11-30 | 株式会社电装 | Vehicle device, vehicle program, and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8116523B2 (en) * | 2007-07-24 | 2012-02-14 | Renesas Electronics Corporation | On-vehicle image processing apparatus |
| CN102663744A (en) * | 2012-03-22 | 2012-09-12 | 杭州电子科技大学 | Complex road detection method under gradient point pair constraint |
| CN103617412A (en) * | 2013-10-31 | 2014-03-05 | 电子科技大学 | Real-time lane line detection method |
| CN104657727A (en) * | 2015-03-18 | 2015-05-27 | 厦门麦克玛视电子信息技术有限公司 | Lane line detection method |
-
2015
- 2015-12-04 CN CN201510884340.XA patent/CN105426864B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8116523B2 (en) * | 2007-07-24 | 2012-02-14 | Renesas Electronics Corporation | On-vehicle image processing apparatus |
| CN102663744A (en) * | 2012-03-22 | 2012-09-12 | 杭州电子科技大学 | Complex road detection method under gradient point pair constraint |
| CN103617412A (en) * | 2013-10-31 | 2014-03-05 | 电子科技大学 | Real-time lane line detection method |
| CN104657727A (en) * | 2015-03-18 | 2015-05-27 | 厦门麦克玛视电子信息技术有限公司 | Lane line detection method |
Non-Patent Citations (1)
| Title |
|---|
| 张利平等: "基于数学形态学的多车道线检测及MATLAB仿真", 《农业装备与车辆工程》 * |
Cited By (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105930800B (en) * | 2016-04-21 | 2019-02-01 | 北京智芯原动科技有限公司 | A kind of method for detecting lane lines and device |
| CN105930800A (en) * | 2016-04-21 | 2016-09-07 | 北京智芯原动科技有限公司 | Lane line detection method and device |
| CN106803066A (en) * | 2016-12-29 | 2017-06-06 | 广州大学 | A kind of vehicle yaw angle based on Hough transform determines method |
| CN107025432B (en) * | 2017-02-28 | 2018-08-21 | 合肥工业大学 | A kind of efficient lane detection tracking and system |
| CN107045629A (en) * | 2017-04-19 | 2017-08-15 | 南京理工大学 | A kind of Multi-lane Lines Detection method |
| CN107045629B (en) * | 2017-04-19 | 2020-06-26 | 南京理工大学 | Multi-lane line detection method |
| CN107643049A (en) * | 2017-09-26 | 2018-01-30 | 沈阳理工大学 | Vehicle position detection system and method on weighbridge based on monocular structure light |
| CN107643049B (en) * | 2017-09-26 | 2023-06-20 | 沈阳理工大学 | System and method for detecting vehicle position on wagon balance based on monocular structured light |
| CN107862290A (en) * | 2017-11-10 | 2018-03-30 | 智车优行科技（北京）有限公司 | Method for detecting lane lines and system |
| CN107862290B (en) * | 2017-11-10 | 2021-09-24 | 智车优行科技（北京）有限公司 | Lane line detection method and system |
| CN108062512A (en) * | 2017-11-22 | 2018-05-22 | 北京中科慧眼科技有限公司 | A kind of method for detecting lane lines and device |
| CN108256445A (en) * | 2017-12-29 | 2018-07-06 | 北京华航无线电测量研究所 | Method for detecting lane lines and system |
| CN108256445B (en) * | 2017-12-29 | 2020-11-06 | 北京华航无线电测量研究所 | Lane line detection method and system |
| CN108509906B (en) * | 2018-03-30 | 2022-02-08 | 长安大学 | Double-threshold Radon identification method for line light point capture of unmanned aerial vehicle aerial photography roadbed |
| CN108509906A (en) * | 2018-03-30 | 2018-09-07 | 长安大学 | Unmanned plane road baseline spot captures dual threshold Radon recognition methods |
| CN108573242A (en) * | 2018-04-26 | 2018-09-25 | 南京行车宝智能科技有限公司 | A kind of method for detecting lane lines and device |
| CN109886175A (en) * | 2019-02-13 | 2019-06-14 | 合肥思艾汽车科技有限公司 | A kind of method for detecting lane lines that straight line is combined with circular arc |
| CN109886175B (en) * | 2019-02-13 | 2021-06-01 | 合肥思艾汽车科技有限公司 | Method for detecting lane line by combining straight line and circular arc |
| CN111738035A (en) * | 2019-03-25 | 2020-10-02 | 比亚迪股份有限公司 | Method, device and equipment for calculating yaw angle of vehicle |
| CN110135252A (en) * | 2019-04-11 | 2019-08-16 | 长安大学 | A kind of adaptive accurate lane detection and deviation method for early warning for unmanned vehicle |
| CN113728370B (en) * | 2019-04-23 | 2023-09-22 | 株式会社电装 | Vehicle device, vehicle program, and storage medium |
| CN113728370A (en) * | 2019-04-23 | 2021-11-30 | 株式会社电装 | Vehicle device, vehicle program, and storage medium |
| CN110163176B (en) * | 2019-05-28 | 2021-06-18 | 北京百度网讯科技有限公司 | Lane line change position identification method, device, equipment and medium |
| CN110163176A (en) * | 2019-05-28 | 2019-08-23 | 北京百度网讯科技有限公司 | The recognition methods of lane line change location, device, equipment and medium |
| CN110068274B (en) * | 2019-05-29 | 2020-08-07 | 易思维（杭州）科技有限公司 | Adhesive tape detection demonstration method for gluing sensor |
| CN110068274A (en) * | 2019-05-29 | 2019-07-30 | 易思维（杭州）科技有限公司 | Gluing sensor adhesive tape detects teaching method |
| CN110443225B (en) * | 2019-08-15 | 2021-07-30 | 安徽卡思普智能科技有限公司 | Virtual and real lane line identification method and device based on feature pixel statistics |
| CN110443225A (en) * | 2019-08-15 | 2019-11-12 | 安徽半问科技有限公司 | Virtual and real lane line identification method and device based on feature pixel statistics |
| CN110414487B (en) * | 2019-08-16 | 2022-05-13 | 东软睿驰汽车技术（沈阳）有限公司 | Method and device for identifying lane line |
| CN110414487A (en) * | 2019-08-16 | 2019-11-05 | 东软睿驰汽车技术（沈阳）有限公司 | Method and device for identifying lane line |
| WO2021116081A1 (en) | 2019-12-13 | 2021-06-17 | Connaught Electronics Ltd. | A method and system for detecting traffic lane boundaries |
| DE102019134320A1 (en) * | 2019-12-13 | 2021-06-17 | Connaught Electronics Ltd. | Method and system for detecting lane boundaries |
| CN111353466B (en) * | 2020-03-12 | 2023-09-22 | 北京百度网讯科技有限公司 | Lane line recognition processing method, equipment and storage medium |
| CN111353466A (en) * | 2020-03-12 | 2020-06-30 | 北京百度网讯科技有限公司 | Lane line recognition processing method, lane line recognition processing device, and storage medium |
| CN111402609A (en) * | 2020-03-17 | 2020-07-10 | 北京百度网讯科技有限公司 | Special lane driving reminding method, device, equipment and storage medium |
| CN111767853A (en) * | 2020-06-30 | 2020-10-13 | 北京百度网讯科技有限公司 | Lane line detection method and device |
| CN111767853B (en) * | 2020-06-30 | 2024-04-23 | 北京百度网讯科技有限公司 | Lane line detection method and device |
| CN113221861B (en) * | 2021-07-08 | 2021-11-09 | 中移(上海)信息通信科技有限公司 | Multi-lane line detection method, device and detection equipment |
| CN113221861A (en) * | 2021-07-08 | 2021-08-06 | 中移(上海)信息通信科技有限公司 | Multi-lane line detection method, device and detection equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105426864B (en) | 2018-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105426864A (en) | Multiple lane line detecting method based on isometric peripheral point matching | |
| US11854272B2 (en) | Hazard detection from a camera in a scene with moving shadows | |
| Niu et al. | Robust lane detection using two-stage feature extraction with curve fitting | |
| CN103714538B (en) | road edge detection method, device and vehicle | |
| Yuan et al. | Robust lane detection for complicated road environment based on normal map | |
| CN102567705B (en) | Method for detecting and tracking night running vehicle | |
| Wu et al. | Applying a functional neurofuzzy network to real-time lane detection and front-vehicle distance measurement | |
| Mammeri et al. | Design of lane keeping assist system for autonomous vehicles | |
| CN101701818B (en) | Method for detecting long-distance barrier | |
| Huang et al. | Lane detection based on inverse perspective transformation and Kalman filter | |
| US8385601B2 (en) | In-vehicle white line recognition apparatus | |
| CN110379168B (en) | Traffic vehicle information acquisition method based on Mask R-CNN | |
| CN102982304B (en) | Utilize polarized light image to detect the method and system of vehicle location | |
| CN103077384A (en) | Method and system for positioning and recognizing vehicle logo | |
| CN103345618A (en) | Traffic violation detection method based on video technology | |
| CN113370977A (en) | Intelligent vehicle forward collision early warning method and system based on vision | |
| CN101916516A (en) | Lane deviation recognition method | |
| CN202134079U (en) | Unmanned vehicle lane marker line identification and alarm device | |
| CN107705577B (en) | Real-time detection method and system for calibrating illegal lane change of vehicle based on lane line | |
| US20200193174A1 (en) | Section line recognition device | |
| Li et al. | A lane marking detection and tracking algorithm based on sub-regions | |
| CN103902985A (en) | High-robustness real-time lane detection algorithm based on ROI | |
| CN103996031A (en) | Self adaptive threshold segmentation lane line detection system and method | |
| CN103794050A (en) | Real-time transport vehicle detecting and tracking method | |
| Xu et al. | Road lane modeling based on RANSAC algorithm and hyperbolic model |
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 |