CN105137412B - A kind of 2D laser radars range image middle conductor feature Accurate Curve-fitting method - Google Patents
A kind of 2D laser radars range image middle conductor feature Accurate Curve-fitting method Download PDFInfo
- Publication number
- CN105137412B CN105137412B CN201510511455.4A CN201510511455A CN105137412B CN 105137412 B CN105137412 B CN 105137412B CN 201510511455 A CN201510511455 A CN 201510511455A CN 105137412 B CN105137412 B CN 105137412B
- Authority
- CN
- China
- Prior art keywords
- data
- segment
- segmentation
- line segment
- range image
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of 2D laser radars range image middle conductor feature Accurate Curve-fitting method, laser radar range image data are obtained first and carry out data prediction；It is converted into Polar Coordinate Model data；Then adaptive threshold region segmentation forming region data are used to Polar Coordinate Model data；And then line segment segmentation is carried out again；Line segment merging finally is carried out to segment data and output merges segment data, the extraction segment data of laser radar range image is used as.The present invention is first split on the basis of traditional Split and Merge algorithms using two-stage, then is fitted the line segment information obtained in laser radar range image by merging；And the selection of segmentation threshold is using adaptive dynamic method, and is optimized in the criterion of merging, and overcoming conventional method needs the shortcoming of the adjusting parameter under different scenes；The line segment feature in laser radar data image is rapidly and accurately extracted, it is adaptable to different indoor scenes, while ensure that the accuracy of fitting.
Description
Technical field
The present invention relates to sensing detection field, more particularly to a kind of 2D laser radars range image middle conductor feature is accurately intended
Conjunction method, it is adaptable to the Accurate Curve-fitting of 2D laser radar range image middle conductor features under doors structure scene.
Background technology
2D laser radars with the characteristics of the great cost performances such as its precision is high, investigative range is wide, anti-interference strong, moderate cost into
For the primary selection of mobile robot independent navigation under circumstances not known.Laser radar is using laser as signal source, by laser
The pulse laser that device is launched at an angle runs into testee back reflection and returns receiver, so that according between the time of transmitting-receiving
The distance of testee is measured every (TOF), wherein 2D laser radars are by certain limit detection range letter from a plane
Breath.
Under the scene of structuring indoors, the spy of line segment is often presented in the range image of laser radar sensing external environment
Levy.For example under corridor environment, two parallel segments are presented in image, under the environment of corner, image show two it is orthogonal
Line segment.The range information in image is only quickly and accurately extracted, environment could be modeled.
How the range information obtained according to laser radar accurately fits line segment information, is mobile robot to unknown
The key detected under environment, at present, existing matching line segment method often use the line segments extraction method based on 2D laser radars,
Mainly have:Split-and-Merge algorithms, Incremental algorithms, Hough transformation algorithm etc.；Wherein, Split-and-
Merge algorithm speeds are fast, but the dependence that merging criterion was chosen and split to the effect extracted to threshold value is larger, different
Need to select suitable parameter under scene.Incremental algorithm speeds are fast, and complexity is low, but the algorithm is primarily adapted for use in
Scene simple in construction, the fitting effect for intersecting straight lines under complex scene is bad.Hough transformation has good noise immunity.
But amount of calculation is very big, it should not be used in the independent navigation higher to requirement of real-time, and the selection of threshold value is also compared
It is difficult.
Because above-mentioned algorithm is present, adaptability is bad, computation complexity is higher, precision is not high, be difficult in adapt to complex scene
Problem.Therefore, urgent need is a kind of both has good adaptability, and rapidity, the laser radar line segments extraction of accuracy are can guarantee that again
Method.
The content of the invention
In view of this, the technical problems to be solved by the invention are to provide a kind of 2D laser radars range image middle conductor spy
Levy Accurate Curve-fitting method, it is adaptable to the accurate plan of 2D laser radar range image middle conductor features under doors structure scene
Close.
The object of the present invention is achieved like this：
A kind of 2D laser radars range image middle conductor feature Accurate Curve-fitting method that the present invention is provided, including following step
Suddenly：
S1：Obtain laser radar range image data and carry out data prediction；
S2：Range image data after data prediction are converted into Polar Coordinate Model data；
S3：Adaptive threshold region segmentation forming region data are used to Polar Coordinate Model data；
S4：Line segment is carried out to each area data to split to form segmentation segment data；
S5：Line segment merging is carried out to the segment data in segmentation segment data；
S6：Finish until all segmentation segment datas merge and export merging segment data, the merging segment data is made
For the extraction segment data of laser radar range image.
Further, data prediction is to be filtered by choosing middle position value filtering method in range image data in the S1
Random disturbances data.
Further, adaptive threshold region segmentation is comprised the following steps that in the S3：
S31：Current data is chosen from Polar Coordinate Model data；
S32：Judge whether the range data in current data is more than zero, if it is, regarding current data sequence number as sequence
Number label flag；
S33：If it is not, then it is current data that return to step S31, which chooses next Polar Coordinate Model data,；
S34：Calculate the front and rear changing value of current data and a upper data；Calculate current data and upper one pole being not zero
The zero data number contained between coordinate model data；
S35：Whether changing value is more than first threshold theta1 before and after judging, or whether zero data number is more than Second Threshold
Theta2, if it is, current data is divided into the m+1 cut zone data, and return to step S31 chooses next pole and sat
Mark model data is current data；
S36：If it is not, then current data is divided into m-th of cut zone data；And return to step S31 chooses next
Polar Coordinate Model data are current data；
S37：To the last Polar Coordinate Model data；
S38：The number of contained data in each cut zone is calculated, number is considered as interference region simultaneously less than predetermined threshold value
Give up.
Further, the line segment segmentation in the S4 comprises the following steps：
S41：The polar data of area data in each cut zone is converted into rectangular co-ordinate data；
S42：Select two area datas in cut zone and fit straight line according to its rectangular co-ordinate data；
S43：Other interior data of the area data are calculated to the point of the straight line ultimate range, and calculate ultimate range；
S44：Judge whether ultimate range is more than adaptive threshold, if it is, the data are arranged at into first straight line number
According to region；If it is not, then the data are arranged at into second straight line data area；
Wherein, the adaptive threshold is calculated according to ultimate range according to below equation：
Wherein, (xk-1,yk-1) table
Show -1 rectangular co-ordinate data of kth, (xk, yk) represent k-th of rectangular co-ordinate data, (xk+1,yk+1) represent that+1 right angle of kth is sat
Mark data；
S45：Circulating repetition is split to first straight line data area and second straight line data area, until all data
Segmentation is finished；
S46：The number of contained segmentation straight line in each cut zone is calculated, number is considered as interference range less than predetermined threshold value
Simultaneously give up in domain.
Further, the S5 middle conductors, which merge, specifically includes following steps：
S51：Take Least Square method to carry out fitting a straight line to the coordinate data in segmentation segment data, form straight line
Equation y=k*x+b；
S52：First coordinate data and last coordinate data in line segment coordinate data are selected, and respectively to straight line
Equation y=k*x+b makees vertical line, and the starting that the linear equation is obtained by calculating the intersection point of the vertical line and the linear equation is sat
Mark and end coordinate；
S53：Linear equation is converted into polar equation；
S54：Calculate adjacent in the differential seat angle absolute value of the adjacent segments in segmentation segment data, and segmentation segment data
The range difference absolute value of line segment；
S55：Judge whether differential seat angle absolute value is less than angle fixed threshold, and whether range difference absolute value is less than distance admittedly
Determine threshold value；If it is, two adjacent segment line segments are merged to form renewal line segment, and the line before merging is replaced to update line segment
Section；If it is not, then into next step；
S56：Return to step S51 circulating repetitions, until all segmentation segment datas are all calculated and finished.
The beneficial effects of the present invention are：The present invention is first used on the basis of traditional Split-and-Merge algorithms
Two-stage is split, then is fitted the line segment information obtained in laser radar range image by merging.First order segmentation utilizes polar coordinates
Middle distance and angle information extract area information, and second level segmentation is then to extract line segment information for area information.And
The selection of segmentation threshold is using adaptive dynamic method, and overcoming conventional method needs the adjusting parameter under different scenes
Shortcoming.It is optimized in the criterion that line segment merges；The present invention can be extracted fast and accurately in laser radar data image
Line segment feature, the advantage of this method is need not to set threshold value manually, is applicable to different indoor scenes, ensure that simultaneously
The accuracy of fitting.
Brief description of the drawings
In order that the object, technical solutions and advantages of the present invention are clearer, below in conjunction with accompanying drawing the present invention is made into
The detailed description of one step, wherein：
Fig. 1 is line segment approximating method flow chart provided in an embodiment of the present invention；
Fig. 2 is region segmentation flow chart provided in an embodiment of the present invention；
Fig. 3 is that line segment provided in an embodiment of the present invention splits flow chart；
Fig. 4 is that line segment provided in an embodiment of the present invention merges flow chart.
Embodiment
Hereinafter with reference to accompanying drawing, the preferred embodiments of the present invention are described in detail.It should be appreciated that preferred embodiment
Only for the explanation present invention, the protection domain being not intended to be limiting of the invention.
Embodiment 1
As shown in figure 1, the 2D laser radar range image middle conductor feature Accurate Curve-fitting methods that the present invention is provided, including with
Lower step：
S1：Obtain laser radar range image data and carry out data prediction；
S2：Range image data after data prediction are converted into Polar Coordinate Model data；The Polar Coordinate Model data
It is made up of angle value and distance value；The angle value is θiCorrespondence intersection point and the angle of X-axis formation；Distance value is piCorrespondence origin
To the distance of straight line；
S3：Adaptive threshold region segmentation forming region data are used to Polar Coordinate Model data；
Adaptive threshold region segmentation is comprised the following steps that in the S3：
S31：Current data is chosen from Polar Coordinate Model data；
S32：Judge whether the range data in current data is more than zero, if it is, regarding current data sequence number as sequence
Number label flag；
The flag of the present embodiment is used for the sequence number for recording last non-zero, when being recycled to non-zero next time,
The sequence number difference of non-zero next time and last non-zero can be calculated；
S33：If it is not, then it is current data that return to step S31, which chooses next Polar Coordinate Model data,；
S34：Calculate the front and rear changing value of current data and a upper data；Calculate current data and upper one pole being not zero
The zero data number contained between coordinate model data；
S35：Whether changing value is more than first threshold theta1 before and after judging, and whether zero data number is more than Second Threshold
Theta2, if it is, current data is divided into the m+1 cut zone data, and return to step S31 chooses next pole and sat
Mark model data is current data；
S36：If it is not, then current data is divided into m-th of cut zone data；And return to step S31 chooses next
Polar Coordinate Model data are current data；
S37：To the last Polar Coordinate Model data；
S38：The number of contained data in each cut zone is calculated, number is considered as interference region simultaneously less than predetermined threshold value
Give up.
S4：Line segment is carried out to each area data to split to form segmentation segment data；
Line segment segmentation in the S4 comprises the following steps：
S41：The polar data of area data in each cut zone is converted into rectangular co-ordinate data；
S42：Select two area datas in cut zone and fit straight line according to its rectangular co-ordinate data；
S43：Other interior data of the area data are calculated to the point of the straight line ultimate range, and calculate ultimate range；
S44：Judge whether ultimate range is more than adaptive threshold, if it is, the data are arranged at into first straight line number
According to region；If it is not, then the data are arranged at into second straight line data area；
Wherein, the adaptive threshold is calculated according to ultimate range according to below equation：
Wherein, theta represents adaptive
Answer threshold value；(xk-1,yk-1) represent -1 rectangular co-ordinate data of kth, (xk, yk) represent k-th of rectangular co-ordinate data, (xk+1,yk+1)
Represent+1 rectangular co-ordinate data of kth；
S45：Circulating repetition is split to first straight line data area and second straight line data area, until all data
Segmentation is finished；
S46：The number of contained segmentation straight line in each cut zone is calculated, number is considered as interference range less than predetermined threshold value
Simultaneously give up in domain.
S5：Line segment merging is carried out to the segment data in segmentation segment data；
The S5 middle conductors, which merge, specifically includes following steps：
S51：Take Least Square method to carry out fitting a straight line to the coordinate data in segmentation segment data, form straight line
Equation y=k*x+b；Y represents ordinate, k straight slopes, and x represents abscissa, and b represents intercept on ordinate.
S52：First coordinate data and last coordinate data in line segment coordinate data are selected, and respectively to straight line
Equation y=k*x+b makees vertical line, and the starting that the linear equation is obtained by calculating the intersection point of the vertical line and the linear equation is sat
Mark and end coordinate；
S53：Linear equation is converted into polar equation；
S54：Calculate adjacent in the differential seat angle absolute value of the adjacent segments in segmentation segment data, and segmentation segment data
The range difference absolute value of line segment；
S55：Judge whether differential seat angle absolute value is less than angle fixed threshold, and whether range difference absolute value is less than distance admittedly
Determine threshold value；If it is, two adjacent segment line segments are merged to form renewal line segment, and the line before merging is replaced to update line segment
Section；If it is not, then into next step；
S56：Return to step S51 circulating repetitions, until all segmentation segment datas are all calculated and finished.
S6：Finish until all segmentation segment datas merge and export merging segment data, the merging segment data is made
For the extraction segment data of laser radar range image.
Embodiment 2
As shown in Fig. 2 Fig. 2 is region segmentation flow chart provided in an embodiment of the present invention；The region segmentation that the present invention is provided
Method, is comprised the following steps that：
The middle position value filtering method of regular length will be used to acquired laser radar data (p1,p2…pi) filtered
After ripple processing, region segmentation is carried out to the laser radar data, that is, image of adjusting the distance carries out region using adaptive threshold value and drawn
Point, mainly including following 3 parts：
1) according to the scanning range and angular resolution of laser radar by range data (p1,p2…pi) it is converted into polar coordinates
(θ1,p1),(θ2,p2)…(θi,pi)；Within a scan period, it is one group apart from number to obtain environmental information from 2D laser radars
According to (p1,p2…pi), the range data obtained more than laser radar range is 0；According to the scanning range of the radar and angle-resolved
Range data, can be converted into polar coordinates (θ by ratei,pi) form, original is obtained to environment sensing material is thus formed laser radar
The range image of beginning.
2) according to polar coordinates (θ1,p1),(θ2,p2)…(θi,pi), using adaptive threshold, data are divided into region
(D1,D2…Dm), wherein, DmIt is (θi,pi) set, refer to and be partitioned into region.Partition principle has two, and first principle is basis
Change before and after range data is compared with first threshold theta1, and second principle is the number containing 0 data between data
Compared with Second Threshold theta2；Theta1 selection is dynamic, if current i=k, first threshold theta1 values are
pkWith pk-1Difference add pk-1With pk-2Difference；
That is i=flag, theta1=pi-pi-1+pi-1-pi-2, theta2=α, α are the fixed error of laser radar；Take solid
Determine threshold value 10；(θi,pi) refer to original angle-range coordinate；
First region segmentation of the present embodiment is not iteration, after meeting segmentation condition, repetitive cycling iterations,
All data until covering；
3) to region (D1,D2…Dm) in interference region handled, give up interference region.
Embodiment 3
As shown in figure 3, Fig. 3, which is line segment provided in an embodiment of the present invention, splits flow chart；The line segment segmentation that the present invention is provided
Method, is comprised the following steps that：
Line segment segmentation is carried out to each area data D, region D is divided into (L by the linear feature of line segment1,L2…Ln),
Mainly include following 3 parts：
1) Coordinate Conversion：To regional ensemble DiMiddle polar coordinates (θ1,p1),(θ2,p2)…(θi,pi) it is converted into rectangular co-ordinate
(x1,y1),(x2,y2)…(xi,yi), wherein, Xi=pi*Cosθi, Yi=pi*Sinθi。
2) selection of threshold value is adaptive, if current i=k, theta values are distance between current point and front and rear point
Sum.
To regional ensemble (D1,D2…Di) in each region DiSplit again, if DiData in set are
(x1,y1),(x2,y2)…(xn,yn),
With first coordinate points (x1,y1) and last coordinate points (xn,yn) fit straight line：
(yn-y1)*x-(xn-x1)*y+y1*xn-yn*x1=0, find other coordinate points (x in region2,y2)…(xn-1,
yn-1) to the point of the straight line ultimate range, calculate ultimate range；The point of the straight line ultimate range, if meeting segmentation condition, i.e.,
Ultimate range be more than threshold value, then with the point by the region segmentation into two lines section.By the region segmentation into L1、L2, wherein LiRepresent
One line segment, is (xn,yn) set, then update whole region Di。
It is the process of an iteration to the segmentation in region, to two straight line L1、L2Continue to use same dividing method.Directly
To all linearity region (L1,L2…Ln) all meet line segment feature；
The present embodiment is in order to avoid common least square method (OLS) is in the case of independent variable and dependent variable are all present under error
There is larger error in fitting, so being fitted using total least square method (TLS) to straight line.
The selection of wherein threshold value is to be calculated to obtain according to adaptive approach：
If the fixed mistake of laser radar
Difference is a, if theta<A, then theta take fixed threshold a.After segmentation terminates, (L is formed1,L2…Ln)。
The present embodiment is by the universal model y=k of line segmenti*x+biIt is converted into (θi,pi) model form.Cross origin and make corresponding straight
The vertical line of line, wherein piOrigin is corresponded to the distance of straight line, θiCorrespondence intersection point and the angle of X-axis formation.Then (θ is utilizedi,pi)
Model goes to merge line segment using threshold value, prevents over-segmentation.
Overcome in this way when line segment and X-axis close to it is vertical when, kiValue is excessive, causes line segment pooled error larger
Shortcoming.Therefore, based on this, the line segment fitting algorithm that the present embodiment is provided both ensure that the rapidity of traditional algorithm, and one
Determine to improve adaptability and accuracy in degree.
3) to (L1,L2…Ln) in interference line segment handled, give up interference line segment；I.e. to line segment aggregate (L1,L2…
Ln) in LiIf, contained (x in Ln,yn) number of coordinates be less than 3, be considered as interference line segment, given up.
Wherein, DmRefer to region, be the set of line segment, LiRefer to the line segment inside region, be the set of coordinate (x, y)；
Dm.length refer to and comprise only a L under the line segment bar number contained in the region, primary condition1；(xk,yk) give directions and arrive (x1,y1),
(xn,yn) connection straight line maximum distance point, Max_Dis refers to the point to the maximum distance of straight line, and theta refers to segmentation adaptive thresholding
Value, if meeting segmentation condition, using k as cut-point, by LiIt is divided into Ltempt1、Ltempt2, wherein using Ltempt1Replace Li,
Dm.Insert(i,Ltempt2) refer to Ltempt2It is inserted into region DmMiddle LiBehind.
Embodiment 4
As shown in figure 4, Fig. 4, which is line segment provided in an embodiment of the present invention, merges flow chart, the line segment that the present invention is provided merges
Method, is comprised the following steps that：
To the line segment (L split1,L2…Ln) adjacent line segment merges according to certain feature, prevent excessively point
Cut, it is main to include with 4 steps：
1) respectively to LiIn the coordinate data (x that includes1,y1),(x2,y2)…(xk,yk) fitting a straight line is carried out, because variable
X, Y contain random error, in order to avoid common least square method (OLS) all exists under error condition in independent variable and dependent variable
There is larger error in fitting, so the method for fitting uses Least Square method (TLS).
2) according to L after fittingiLinear equation y=ki*x+bi, LiFirst data point (x in1,y1) and last
Individual data point (xk,yk) vertical line is made to linear equation respectively, by determining origin coordinates and end coordinate on linear equation, most
Line segment is determined eventually, and intersection point is the origin coordinates and end coordinate of the line segment.
3) by LiModel y=ki*x+biIt is converted into (θi,pi) model, cross the vertical line that origin makees line correspondence, wherein piIt is former
Point arrives the distance of straight line, θiCorrespondence intersection point and the angle of X-axis formation, -180 ° of ＜ θi180 ° of ＜.
4) (the L in the D of region1,L2…Ln) straight line model (θ1,p1),(θ2,p2)…(θn,pn), to regional ensemble
In adjacent line segment merge, when the absolute value of the difference of adjacent segments angle is less than fixed threshold and apart from the absolute of its difference
Value is less than fixed threshold, then merges two sections of line segments, and update area D.The same method using iteration, until all phases
Adjacent line segment is satisfied by.
Wherein, D in Fig. 4mRefer to a region, be the set of line segment, LiRefer to the line segment inside region, Dm.length Zhi Gai areas
The line segment bar number contained in domain.(θi,pi) it is straight line LiStraight line model, ang_theta is angle threshold, dis_theta be away from
From threshold value, threshold value when it is merging.If merging condition is met, by LiWith Li-1It is merged into Ltempt, replace Li-1, same to time shift
Except Li, Dm.remove(Li) refer to removal DmL in regioni。
Finally illustrate, the above embodiments are merely illustrative of the technical solutions of the present invention and it is unrestricted, although pass through ginseng
According to the preferred embodiments of the present invention, invention has been described, it should be appreciated by those of ordinary skill in the art that can
So that various changes are made to it in the form and details, the spirit and scope limited without departing from the present invention.
Claims (3)
1. a kind of 2D laser radars range image middle conductor feature Accurate Curve-fitting method, it is characterised in that：Comprise the following steps：
S1：Obtain laser radar range image data and carry out data prediction；
S2：Range image data after data prediction are converted into Polar Coordinate Model data；
S3：Adaptive threshold region segmentation forming region data are used to Polar Coordinate Model data；
S4：Line segment is carried out to each area data to split to form segmentation segment data；
S5：Line segment merging is carried out to the segment data in segmentation segment data；
S6：Finish until all segmentation segment datas merge and export merging segment data, the merging segment data is as sharp
The extraction segment data of optical radar range image；
Data prediction is to filter the random disturbances in range image data by choosing middle position value filtering method in the S1
Data；
Adaptive threshold region segmentation is comprised the following steps that in the S3：
S31：Current data is chosen from Polar Coordinate Model data；
S32：Judge whether the range data in current data is more than zero, if it is, regarding current data sequence number as sequence number mark
Sign flag；
S33：If it is not, then it is current data that return to step S31, which chooses next Polar Coordinate Model data,；
S34：Calculate the front and rear changing value of current data and a upper data；Calculate current data and upper one polar coordinates being not zero
The zero data number contained between model data；
S35：Whether changing value is more than first threshold theta1 before and after judging, or whether zero data number is more than Second Threshold
Theta2, if it is, current data is divided into the m+1 cut zone data, and return to step S31 chooses next pole and sat
Mark model data is current data；
S36：If it is not, then current data is divided into m-th of cut zone data；And return to step S31 chooses next pole and sat
Mark model data is current data；
S37：To the last Polar Coordinate Model data；
S38：The number of contained data is calculated in each cut zone, number is less than being considered as interference region and giving up for predetermined threshold value
Abandon.
2. 2D laser radars range image middle conductor feature Accurate Curve-fitting method according to claim 1, it is characterised in that：
Line segment segmentation in the S4 comprises the following steps：
S41：The polar data of area data in each cut zone is converted into rectangular co-ordinate data；
S42：Select two area datas in cut zone and fit straight line according to its rectangular co-ordinate data；
S43：Other interior data of the area data are calculated to the point of the straight line ultimate range, and calculate ultimate range；
S44：Judge whether ultimate range is more than adaptive threshold, if it is, the data are arranged at into first straight line data field
Domain；
If it is not, then the data are arranged at into second straight line data area；
Wherein, the adaptive threshold is calculated according to ultimate range according to below equation：
Wherein, (xk-1,yk-1) expression kth-
1 rectangular co-ordinate data, (xk, yk) represent k-th of rectangular co-ordinate data, (xk+1,yk+1) represent+1 rectangular co-ordinate data of kth；
S45：Circulating repetition is split to first straight line data area and second straight line data area, until the segmentation of all data
Finish；
S46：The number of contained segmentation straight line in each cut zone is calculated, number is considered as interference region simultaneously less than predetermined threshold value
Give up.
3. 2D laser radars range image middle conductor feature Accurate Curve-fitting method according to claim 1, it is characterised in that：
The S5 middle conductors, which merge, specifically includes following steps：
S51：Take Least Square method to carry out fitting a straight line to the coordinate data in segmentation segment data, form linear equation
Y=k*x+b；
S52：First coordinate data and last coordinate data in line segment coordinate data are selected, and respectively to linear equation
Y=k*x+b makees vertical line, by calculate the intersection point of the vertical line and the linear equation obtain the linear equation origin coordinates and
End coordinate；
S53：Linear equation is converted into polar equation；
S54：Calculate the adjacent segments in the differential seat angle absolute value of the adjacent segments in segmentation segment data, and segmentation segment data
Range difference absolute value；
S55：Judge whether differential seat angle absolute value is less than angle fixed threshold, and whether range difference absolute value is less than the fixed threshold of distance
Value；
If it is, two adjacent segment line segments are merged to form renewal line segment, and the line segment before merging is replaced to update line segment；
If it is not, then into next step；
S56：Return to step S51 circulating repetitions, until all segmentation segment datas are all calculated and finished.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510511455.4A CN105137412B (en) | 2015-08-19 | 2015-08-19 | A kind of 2D laser radars range image middle conductor feature Accurate Curve-fitting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510511455.4A CN105137412B (en) | 2015-08-19 | 2015-08-19 | A kind of 2D laser radars range image middle conductor feature Accurate Curve-fitting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105137412A CN105137412A (en) | 2015-12-09 |
| CN105137412B true CN105137412B (en) | 2017-10-20 |
Family
ID=54722818
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510511455.4A Active CN105137412B (en) | 2015-08-19 | 2015-08-19 | A kind of 2D laser radars range image middle conductor feature Accurate Curve-fitting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105137412B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105987697B (en) * | 2016-04-26 | 2019-01-29 | 重庆大学 | The wheeled AGV navigation locating method of Mecanum and system under a kind of quarter bend |
| CN110399892B (en) * | 2018-04-24 | 2022-12-02 | 北京京东尚科信息技术有限公司 | Environmental feature extraction method and device |
| CN110412530B (en) * | 2018-04-27 | 2021-09-17 | 深圳市优必选科技有限公司 | Method and device for identifying charging pile and robot |
| CN108805896B (en) * | 2018-05-07 | 2022-02-18 | 南京理工大学 | Distance image segmentation method applied to urban environment |
| CN109443370A (en) * | 2018-09-13 | 2019-03-08 | 中通国脉物联科技南京有限公司 | A method of deviate detection track |
| CN110146910B (en) * | 2019-05-15 | 2023-02-28 | 重庆大学 | Positioning method and device based on data fusion of GPS and laser radar |
| CN110632607B (en) * | 2019-09-16 | 2022-02-01 | 广州小鹏自动驾驶科技有限公司 | Object boundary determining method and system based on ultrasonic data and vehicle |
| CN113093219A (en) * | 2021-02-20 | 2021-07-09 | 意欧斯物流科技(上海)有限公司 | 2D laser radar data line segment extraction method based on region growing algorithm |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1808711A2 (en) * | 2006-01-16 | 2007-07-18 | Omron Corporation | Object detector |
| CN102930280A (en) * | 2012-10-05 | 2013-02-13 | 中国电子科技集团公司第十研究所 | Method for identifying overhead high-voltage wire automatically from infrared image |
| CN103760569A (en) * | 2013-12-31 | 2014-04-30 | 西安交通大学 | Drivable region detection method based on laser radar |
| CN103777192A (en) * | 2012-10-24 | 2014-05-07 | 中国人民解放军第二炮兵工程学院 | Linear feature extraction method based on laser sensor |
| CN104008387A (en) * | 2014-05-19 | 2014-08-27 | 山东科技大学 | Lane line detection method based on feature point piecewise linear fitting |
| CN104200212A (en) * | 2014-06-25 | 2014-12-10 | 西安煤航信息产业有限公司 | Building outer boundary line extraction method based on onboard LiDAR (Light Detection and Ranging) data |
| CN104463871A (en) * | 2014-12-10 | 2015-03-25 | 武汉大学 | Streetscape facet extraction and optimization method based on vehicle-mounted LiDAR point cloud data |
-
2015
- 2015-08-19 CN CN201510511455.4A patent/CN105137412B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1808711A2 (en) * | 2006-01-16 | 2007-07-18 | Omron Corporation | Object detector |
| CN102930280A (en) * | 2012-10-05 | 2013-02-13 | 中国电子科技集团公司第十研究所 | Method for identifying overhead high-voltage wire automatically from infrared image |
| CN103777192A (en) * | 2012-10-24 | 2014-05-07 | 中国人民解放军第二炮兵工程学院 | Linear feature extraction method based on laser sensor |
| CN103760569A (en) * | 2013-12-31 | 2014-04-30 | 西安交通大学 | Drivable region detection method based on laser radar |
| CN104008387A (en) * | 2014-05-19 | 2014-08-27 | 山东科技大学 | Lane line detection method based on feature point piecewise linear fitting |
| CN104200212A (en) * | 2014-06-25 | 2014-12-10 | 西安煤航信息产业有限公司 | Building outer boundary line extraction method based on onboard LiDAR (Light Detection and Ranging) data |
| CN104463871A (en) * | 2014-12-10 | 2015-03-25 | 武汉大学 | Streetscape facet extraction and optimization method based on vehicle-mounted LiDAR point cloud data |
Non-Patent Citations (1)
| Title |
|---|
| 基于自适应阈值的距离图像线段特征提取;满增光 等;《深圳大学学报理工版》;20111130;第28卷(第6期);第483-488页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105137412A (en) | 2015-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105137412B (en) | A kind of 2D laser radars range image middle conductor feature Accurate Curve-fitting method | |
| Zheng et al. | Improved lane line detection algorithm based on Hough transform | |
| Pinggera et al. | Lost and found: detecting small road hazards for self-driving vehicles | |
| CN104318258B (en) | Time domain fuzzy and kalman filter-based lane detection method | |
| CN110738121A (en) | front vehicle detection method and detection system | |
| CN108647646A (en) | The optimizing detection method and device of low obstructions based on low harness radar | |
| CN107301664A (en) | Improvement sectional perspective matching process based on similarity measure function | |
| CN105261020A (en) | Method for detecting fast lane line | |
| CN110307791B (en) | Vehicle length and speed calculation method based on three-dimensional vehicle boundary frame | |
| CN109001757A (en) | A kind of parking space intelligent detection method based on 2D laser radar | |
| CN108960115A (en) | Multi-direction Method for text detection based on angle point | |
| CN102903111B (en) | Large area based on Iamge Segmentation low texture area Stereo Matching Algorithm | |
| CN105513094A (en) | Stereo vision tracking method and stereo vision tracking system based on 3D Delaunay triangulation | |
| CN110110678A (en) | Determination method and apparatus, storage medium and the electronic device of road boundary | |
| CN113177593A (en) | Fusion method of radar point cloud and image data in water traffic environment | |
| CN101383046B (en) | Three-dimensional reconstruction method on basis of image | |
| Wang et al. | Fast vanishing point detection method based on road border region estimation | |
| CN103279748B (en) | A kind of Approach for road detection based on SIFT – COF characteristic light stream | |
| CN105678791A (en) | Lane line detection and tracking method based on parameter non-uniqueness property | |
| CN101783014A (en) | Circle detection method | |
| CN113487631A (en) | Adjustable large-angle detection sensing and control method based on LEGO-LOAM | |
| Zhu et al. | A beam guardrail detection algorithm using lidar for intelligent vehicle | |
| CN107038710A (en) | A kind of Vision Tracking using paper as target | |
| CN111353481A (en) | Road obstacle identification method based on laser point cloud and video image | |
| CN103983270B (en) | A kind of image conversion processing method of sonar data |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201222 Address after: 402460 station No.14, no.6, 10th floor, innovation and development center, No.19 Lingfang Avenue, Changzhou street, Rongchang District, Chongqing Patentee after: Chongqing kezhiyuan Technology Co.,Ltd. Address before: 400044 No. 174 Sha Jie street, Shapingba District, Chongqing Patentee before: Chongqing University |
|
| TR01 | Transfer of patent right |