CN109903338A - A kind of method for positioning mobile robot based on improvement ORB algorithm - Google Patents
A kind of method for positioning mobile robot based on improvement ORB algorithm Download PDFInfo
- Publication number
- CN109903338A CN109903338A CN201910193033.5A CN201910193033A CN109903338A CN 109903338 A CN109903338 A CN 109903338A CN 201910193033 A CN201910193033 A CN 201910193033A CN 109903338 A CN109903338 A CN 109903338A
- Authority
- CN
- China
- Prior art keywords
- mobile robot
- image
- algorithm
- point
- orb
- 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.)
- Pending
Links
Landscapes
- Image Analysis (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a kind of based on the method for positioning mobile robot for improving ORB algorithm, belong to localization for Mobile Robot technical field, a kind of method for positioning mobile robot based on improvement ORB algorithm, including mobile robot and computer, input picture pair: mobile robot carries out image procossing by video camera real-time image acquisition, and by Image Real-time Transmission to computer；It is image gridding: to divide the image into the not overlapped elements of G=20*20 using grid using GMS algorithm；ORB description: the characteristic point of ORB algorithm Detection and Extraction is utilized, the present invention may be implemented to extract the characteristic point with scale invariability using SIFT algorithm, BRIEF is carried out again to describe to obtain description, finally quality of match is significantly improved using a kind of GMS feature registration algorithm based on scoring frame, with preferable real-time and characteristic matching accuracy rate, odometer accumulated error in independent navigation is effectively reduced, mobile robot autonomous positioning precision is improved.
Description
Technical field
The present invention relates to localization for Mobile Robot technical field, more specifically to a kind of based on improving ORB algorithm
Method for positioning mobile robot.
Background technique
In recent years, with the development of artificial intelligence technology, mobile robot is widely applied, by personnel from heavy
It is freed in manual labor, while being also effectively saved human cost, mobile robot needs in its long range traveling process
To obtain its location information in real time, common Mobile Robotics Navigation method includes magnetic navigation, inertial navigation, vision guided navigation etc.,
Magnetic navigation reliability is higher, but higher cost and maintenance difficult；Inertial navigation is that mobile robot is commonly navigated mode, because of it
It will lead to accumulated error infinitely to increase and be not suitable for long-time precision navigation, currently used Mobile Robotics Navigation mode is base
There is biggish accumulated error although this air navigation aid stability is high in the movement statistics integrated navigation mode of grid.
ORB algorithm is characteristic point detection and the description algorithm of a kind of view-based access control model information, uses FAST (features
From accelerated segment test) algorithm detects the pixel value in a certain characteristic point to be measured and peripheral region, the area
The generally circular in cross section region in domain, meanwhile, ORB algorithm uses BRIEF (binary robust independent elementary
Features) algorithm carries out feature point description, and several pairs of random points, the picture of more every a pair of random point are chosen near characteristic point
Plain size simultaneously carries out binary system assignment, is then combined into a binary coding, as Feature Descriptor, according to match point with it is non-
The Hamming distance of match point realizes matching from given threshold, and ORB algorithm joined directional information to FAST feature, while in spy
BRIEF description is used when sign description, and it is improved, and makes its description that there is preferable Shandong to noise and rotation
Stick although ORB algorithm has faster operation speed, and all has preferable performance to noise and image rotation variation,
But since FAST feature does not have scale invariability, so the rescaling for image does not have robustness.
The positioning of view-based access control model odometer is a kind of current more novel localization for Mobile Robot technology, and robot is by taking the photograph
Camera acquires characteristics of image, carries out pose estimation and independent navigation according to characteristics of image and kinematic constraint fuse information, adopts at random
Sample consistency (random sample consensus, RANSAC) algorithm is one of common visual signature registration Algorithm, but
The error hiding rate and complexity of RANSAC algorithm are higher.
Summary of the invention
1. technical problems to be solved
Aiming at the problems existing in the prior art, the purpose of the present invention is to provide a kind of based on the moving machine for improving ORB algorithm
Device people's localization method, it may be implemented to extract the characteristic point with scale invariability using SIFT algorithm, then carry out BRIEF description
Description is obtained, finally quality of match is significantly improved using a kind of GMS feature registration algorithm based on scoring frame, has preferable
Real-time and characteristic matching accuracy rate, effectively reduce odometer accumulated error in independent navigation, improve mobile robot
Autonomous positioning precision.
2. technical solution
To solve the above problems, the present invention adopts the following technical scheme that.
A kind of method for positioning mobile robot based on improvement ORB algorithm, including mobile robot and computer, including with
Lower step:
(1) input picture pair: mobile robot by video camera real-time image acquisition, and by Image Real-time Transmission to computer into
Row image procossing；
(2) image gridding: to divide the image into the not overlapped elements of G=20*20 using grid using GMS algorithm；
(3) ORB description: pass through BRIEF algorithm progress feature point description again using the characteristic point of ORB algorithm Detection and Extraction and obtain
Description；
(4) Bayes's visual modeling；
(5) network weights count；
(6) network statistics value matches: using the GMS feature registration algorithm based on scoring frame, since motion smoothing makes to match
Feature vertex neighborhood has more correct matching pair, will correct and erroneous matching characteristic nonlinear function as score function, according to
Given threshold determines characteristic matching standard；
(7) matching result is exported.
This programme may be implemented to extract the characteristic point with scale invariability using SIFT algorithm, then carry out BRIEF description
Description is obtained, finally quality of match is significantly improved using a kind of GMS feature registration algorithm based on scoring frame, has preferable
Real-time and characteristic matching accuracy rate, effectively reduce odometer accumulated error in independent navigation, improve mobile robot
Autonomous positioning precision.
Further, specific step is as follows for step (3):
The detection of (3-1) characteristic point: use the method for SIFT by original image and gaussian kernel function convolution tectonic scale space, structure first
Pyramid is made, extreme point detection is carried out to the image of different scale, the extreme point being achieved in that has scale invariability, then leads to
Over-fitting function clicks through row interpolation to discrete space and obtains continuous space extreme point, determines position and the scale of key point；
(3-2) description generates: describing method using BRIEF and characteristic point is described, BRIEF Feature Descriptor is to image
The description of block bi-level digital string, by comparing the image grayscale of bi-level digital string, by obtained result composition bi-level digital string
Form, the specific judgement of two-value are as follows:
Gray value of p (x) the representative image block p at point x in formula, selects n (x, y) test points pair, and the n of generation ties up binary number
Word string are as follows:
Specific practice is drawn and is justified as radius using certain length using key point as the center of circle, several with a certain model selection in circle
Point pair, by above formula method compare each pair of point as a result, by comparison result composition binary character string be this feature point
Description.
Further, it is the rotational invariance for increasing description in the step (3-1), histogram is passed through using SIFT method
Figure counts pixel gradient and direction in crucial vertex neighborhood, is the principal direction of key point by histogram peak direction.
Further, the step (3-2) is in order to guarantee that description has rotation consistency, using SIFT method by coordinate
Axis rotates the direction as key point, to ensure rotational invariance.
Further, specific step is as follows for step (6):
Neighborhood is sorted out by calculating score of the matching to field, gives unit to { i, j }, then score are as follows:
In formula | Xik,jk| it is unit to { ik,jkBetween matching to number, k is the disjoint areal for matching i, when grid
G=20*20, feature point number 10000, the characteristic point n=25 of unit each in this way pass through score SijBy unit to being divided into
Correct and Error Set is combined into { T, F }:
A in formulaiIt is threshold value, α=6, niIt is the number of all characteristic points, as score SijIt is then correct when greater than certain threshold value
Matching neighborhood, then only retaining all matchings pair in the region.
Further, specific step is as follows for step (2):
The not overlapped elements of G=20*20 are divided the image into using grid, and use half-space width mobile net on x, the direction y
Lattice mode and in triplicate, solves the problems, such as that many points are located at Grid Edge edge in practical application.
3. compared with the prior art, the present invention has the advantages that
(1) this programme may be implemented to extract the characteristic point with scale invariability using SIFT algorithm, then carries out BRIEF and retouch
State to obtain description, finally significantly improve quality of match using a kind of GMS feature registration algorithm based on scoring frame, have compared with
Good real-time and characteristic matching accuracy rate, effectively reduces odometer accumulated error in independent navigation, improves mobile machine
People's autonomous positioning precision.
(2) it on the basis of ORB algorithm, is improved for ORB algorithm without the shortcomings that scale invariability, it will
There is SIFT preferable scale invariability to be integrated on ORB algorithm, to make description have rotational invariance, uses SIFT and calculates
A characteristic point principal direction establishes the mode of reference axis in method, obtains BRIEF description with rotational invariance, finally uses base
Accurate matching is realized in the GMS feature registration algorithm of scoring frame.
Detailed description of the invention
Fig. 1 is the solution of the present invention flow chart；
Fig. 2 is the runing time comparison diagram of ORB algorithm of the present invention and improved 5 width image of ORB algorithm；
Fig. 3 is the accuracy rate comparison diagram of ORB algorithm of the present invention and improved 5 width image of ORB algorithm.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description；Obviously, described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments, is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
In the description of the present invention, it should be noted that the instruction such as term " on ", "lower", "inner", "outside", " top/bottom end "
Orientation or positional relationship be based on the orientation or positional relationship shown in the drawings, be merely for convenience of description the present invention and simplification retouch
It states, rather than the device or element of indication or suggestion meaning must have a particular orientation, be constructed and operated in a specific orientation,
Therefore it is not considered as limiting the invention.In addition, term " first ", " second " are used for description purposes only, and cannot understand
For indication or suggestion relative importance.
In the description of the present invention, it should be noted that unless otherwise clearly defined and limited, term " installation " " is set
Be equipped with ", " be arranged/connect ", " connection " etc., shall be understood in a broad sense, such as " connection ", may be a fixed connection, be also possible to removable
Connection is unloaded, or is integrally connected, can be mechanical connection, is also possible to be electrically connected, can be directly connected, it can also be in
Between medium be indirectly connected, can be the connection inside two elements, for the ordinary skill in the art, can be specific
Situation understands the concrete meaning of above-mentioned term in the present invention.
Embodiment 1:
Fig. 1-3 is please referred to, it is a kind of based on the method for positioning mobile robot for improving ORB algorithm, including mobile robot and calculating
Machine, the experiment running environment of computer be Intel (R) Core (TM) I5-4590M 3.3GHz 4GB, VS2013 and
Opencv3.2, comprising the following steps:
(1) input picture pair: mobile robot by video camera real-time image acquisition, and by Image Real-time Transmission to computer into
Row image procossing；
(2) image gridding: to divide the image into the not overlapped elements of G=20*20 using grid using GMS algorithm；
(3) ORB description: pass through BRIEF algorithm progress feature point description again using the characteristic point of ORB algorithm Detection and Extraction and obtain
Description；
(4) Bayes's visual modeling；
(5) network weights count；
(6) network statistics value matches: using the GMS feature registration algorithm based on scoring frame, since motion smoothing makes to match
Feature vertex neighborhood has more correct matching pair, will correct and erroneous matching characteristic nonlinear function as score function, according to
Given threshold determines characteristic matching standard；
(7) matching result is exported.
Specific step is as follows for step (3):
The detection of (3-1) characteristic point: use the method for SIFT by original image and gaussian kernel function convolution tectonic scale space, structure first
Pyramid is made, extreme point detection is carried out to the image of different scale, the extreme point being achieved in that has scale invariability, then leads to
Over-fitting function clicks through row interpolation to discrete space and obtains continuous space extreme point, determines position and the scale of key point；
(3-2) description generates: describing method using BRIEF and characteristic point is described, BRIEF Feature Descriptor is to image
The description of block bi-level digital string, by comparing the image grayscale of bi-level digital string, by obtained result composition bi-level digital string
Form, the specific judgement of two-value are as follows:
Gray value of p (x) the representative image block p at point x in formula, selects n (x, y) test points pair, and the n of generation ties up binary number
Word string are as follows:
Specific practice is drawn and is justified as radius using certain length using key point as the center of circle, several with a certain model selection in circle
Point pair, by above formula method compare each pair of point as a result, by comparison result composition binary character string be this feature point
Description.
It is the rotational invariance for increasing description in step (3-1), statistics with histogram key point is passed through using SIFT method
Histogram peak direction is the principal direction of key point by pixel gradient and direction in neighborhood.
Step (3-2) has rotation consistency to guarantee that description is sub, is rotated reference axis as pass using SIFT method
The direction of key point, to ensure rotational invariance.
Specific step is as follows for step (6):
Neighborhood is sorted out by calculating score of the matching to field, gives unit to { i, j }, then score are as follows:
In formula | Xik,jk| it is unit to { ik,jkBetween matching to number, k is the disjoint areal for matching i, when grid
G=20*20, feature point number 10000, the characteristic point n=25 of unit each in this way pass through score SijBy unit to being divided into
Correct and Error Set is combined into { T, F }:
A in formulaiIt is threshold value, α=6, niIt is the number of all characteristic points, as score SijIt is then correct when greater than certain threshold value
Matching neighborhood, then only retaining all matchings pair in the region.
Specific step is as follows for step (2):
The not overlapped elements of G=20*20 are divided the image into using grid, and use half-space width mobile net on x, the direction y
Lattice mode and in triplicate, solves the problems, such as that many points are located at Grid Edge edge in practical application.
Mobile robot reaches designated position, mobile robot Real Time Obstacle Avoiding in navigation procedure according to preset path from starting point
Path planning updates current time pose according to feature registration, realizes autonomous positioning and map rejuvenation, and mobile robot uses base
Present image feature is matched with road sign three-dimensional in map office in the GMS algorithm of scoring frame, according to odometer and characteristic matching
Fuse information realizes pose and map rejuvenation, and the present invention may be implemented to extract the feature with scale invariability using SIFT algorithm
Point, then carry out BRIEF and describe to obtain description, finally significantly mentioned using a kind of GMS feature registration algorithm based on scoring frame
High quality of match has preferable real-time and characteristic matching accuracy rate, effectively reduces odometer accumulation in independent navigation and misses
Difference improves mobile robot autonomous positioning precision.
Referring to Fig. 1, being changed for ORB algorithm without the shortcomings that scale invariability on the basis of ORB algorithm
Into, there is preferable scale invariability to be integrated on ORB algorithm SIFT, it is sub with rotational invariance to make to describe, it uses
A characteristic point principal direction establishes the mode of reference axis in SIFT algorithm, obtains BRIEF description with rotational invariance, please join
Fig. 3 is read, accurate matching is finally realized using the GMS feature registration algorithm based on scoring frame, matching accuracy rate obtains significantly
It is promoted, please refers to the high speed of service that Fig. 2 innovatory algorithm also remains with traditional ORB algorithm.
More than, it is merely preferred embodiments of the present invention；But scope of protection of the present invention is not limited thereto.It is any
Those familiar with the art in the technical scope disclosed by the present invention, according to the technique and scheme of the present invention and its improves
Design is subject to equivalent substitution or change, should be covered by the scope of protection of the present invention.
Claims (6)
1. a kind of existed based on the method for positioning mobile robot for improving ORB algorithm, including mobile robot and computer, feature
In: the following steps are included:
(1) input picture pair: mobile robot by video camera real-time image acquisition, and by Image Real-time Transmission to computer into
Row image procossing；
(2) image gridding: to divide the image into the not overlapped elements of G=20*20 using grid using GMS algorithm；
(3) ORB description: pass through BRIEF algorithm progress feature point description again using the characteristic point of ORB algorithm Detection and Extraction and obtain
Description；
(4) Bayes's visual modeling；
(5) network weights count；
(6) network statistics value matches: using the GMS feature registration algorithm based on scoring frame, since motion smoothing makes to match
Feature vertex neighborhood has more correct matching pair, will correct and erroneous matching characteristic nonlinear function as score function, according to
Given threshold determines characteristic matching standard；
(7) matching result is exported.
2. according to claim 1 a kind of based on the method for positioning mobile robot for improving ORB algorithm, it is characterised in that:
Specific step is as follows for step (3):
The detection of (3-1) characteristic point: use the method for SIFT by original image and gaussian kernel function convolution tectonic scale space, structure first
Pyramid is made, extreme point detection is carried out to the image of different scale, the extreme point being achieved in that has scale invariability, then leads to
Over-fitting function clicks through row interpolation to discrete space and obtains continuous space extreme point, determines position and the scale of key point；
(3-2) description generates: describing method using BRIEF and characteristic point is described, BRIEF Feature Descriptor is to image
The description of block bi-level digital string, by comparing the image grayscale of bi-level digital string, by obtained result composition bi-level digital string
Form, the specific judgement of two-value are as follows:
Gray value of p (x) the representative image block p at point x in formula, selects n (x, y) test points pair, and the n of generation ties up binary number
Word string are as follows:
Specific practice is drawn and is justified as radius using certain length using key point as the center of circle, several with a certain model selection in circle
Point pair, by above formula method compare each pair of point as a result, by comparison result composition binary character string be this feature point
Description.
3. according to claim 1 or 2 a kind of based on the method for positioning mobile robot for improving ORB algorithm, feature exists
In: it is the rotational invariance for increasing description in the step (3-1), it is adjacent by statistics with histogram key point using SIFT method
Histogram peak direction is the principal direction of key point by pixel gradient and direction in domain.
4. according to claim 1 or 2 a kind of based on the method for positioning mobile robot for improving ORB algorithm, feature exists
In: the step (3-2) is rotated reference axis as crucial in order to guarantee that description has rotation consistency, using SIFT method
The direction of point, to ensure rotational invariance.
5. according to claim 1 a kind of based on the method for positioning mobile robot for improving ORB algorithm, it is characterised in that:
Specific step is as follows for step (6):
Neighborhood is sorted out by calculating score of the matching to field, gives unit to { i, j }, then score are as follows:
In formula | Xik,jk| it is unit to { ik,jkBetween matching to number, k is the disjoint areal for matching i, G when grid
=20*20, feature point number 10000, the characteristic point n=25 of unit each in this way pass through score SijBy unit to being divided into
Correct and Error Set is combined into { T, F }:
A in formulaiIt is threshold value, α=6, niIt is the number of all characteristic points, as score SijIt is then correct when greater than certain threshold value
Neighborhood is matched, then only retaining all matchings pair in the region.
6. according to claim 1 a kind of based on the method for positioning mobile robot for improving ORB algorithm, it is characterised in that:
Specific step is as follows for step (2):
The not overlapped elements of G=20*20 are divided the image into using grid, and use half-space width mobile net on x, the direction y
Lattice mode and in triplicate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910193033.5A CN109903338A (en) | 2019-03-14 | 2019-03-14 | A kind of method for positioning mobile robot based on improvement ORB algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910193033.5A CN109903338A (en) | 2019-03-14 | 2019-03-14 | A kind of method for positioning mobile robot based on improvement ORB algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109903338A true CN109903338A (en) | 2019-06-18 |
Family
ID=66952250
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910193033.5A Pending CN109903338A (en) | 2019-03-14 | 2019-03-14 | A kind of method for positioning mobile robot based on improvement ORB algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109903338A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110598783A (en) * | 2019-09-10 | 2019-12-20 | 中国科学技术大学 | Visual consistency method based on distributed mobile robot system |
| CN110619338A (en) * | 2019-09-18 | 2019-12-27 | 成都信息工程大学 | Image feature extraction method capable of long-time dependence |
| CN110675437A (en) * | 2019-09-24 | 2020-01-10 | 重庆邮电大学 | Image matching method based on improved GMS-ORB characteristics and storage medium |
| CN110689578A (en) * | 2019-10-11 | 2020-01-14 | 南京邮电大学 | Unmanned aerial vehicle obstacle identification method based on monocular vision |
| CN111739066A (en) * | 2020-07-27 | 2020-10-02 | 深圳大学 | Visual positioning method, system and storage medium based on Gaussian process |
| CN114111803A (en) * | 2022-01-26 | 2022-03-01 | 中国人民解放军战略支援部队航天工程大学 | Visual navigation method of indoor satellite platform |
-
2019
- 2019-03-14 CN CN201910193033.5A patent/CN109903338A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110598783A (en) * | 2019-09-10 | 2019-12-20 | 中国科学技术大学 | Visual consistency method based on distributed mobile robot system |
| CN110619338A (en) * | 2019-09-18 | 2019-12-27 | 成都信息工程大学 | Image feature extraction method capable of long-time dependence |
| CN110619338B (en) * | 2019-09-18 | 2022-02-08 | 成都信息工程大学 | Image feature extraction method capable of long-time dependence |
| CN110675437A (en) * | 2019-09-24 | 2020-01-10 | 重庆邮电大学 | Image matching method based on improved GMS-ORB characteristics and storage medium |
| CN110675437B (en) * | 2019-09-24 | 2023-03-28 | 重庆邮电大学 | Image matching method based on improved GMS-ORB characteristics and storage medium |
| CN110689578A (en) * | 2019-10-11 | 2020-01-14 | 南京邮电大学 | Unmanned aerial vehicle obstacle identification method based on monocular vision |
| CN111739066A (en) * | 2020-07-27 | 2020-10-02 | 深圳大学 | Visual positioning method, system and storage medium based on Gaussian process |
| CN114111803A (en) * | 2022-01-26 | 2022-03-01 | 中国人民解放军战略支援部队航天工程大学 | Visual navigation method of indoor satellite platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109903338A (en) | A kind of method for positioning mobile robot based on improvement ORB algorithm | |
| Sun et al. | Leveraging crowdsourced GPS data for road extraction from aerial imagery | |
| CN103310233B (en) | With similarity method for digging between class behavior multi views and Activity recognition method | |
| CN104200523B (en) | A kind of large scene three-dimensional rebuilding method for merging additional information | |
| CN109520500B (en) | Accurate positioning and street view library acquisition method based on terminal shooting image matching | |
| CN107481315A (en) | A kind of monocular vision three-dimensional environment method for reconstructing based on Harris SIFT BRIEF algorithms | |
| Lian et al. | DeepWindow: Sliding window based on deep learning for road extraction from remote sensing images | |
| CN109684921A (en) | A kind of road edge identification and tracking based on three-dimensional laser radar | |
| CN103473774B (en) | A kind of vehicle positioning method based on pavement image characteristic matching | |
| CN109636854A (en) | A kind of augmented reality three-dimensional Tracing Registration method based on LINE-MOD template matching | |
| CN107369183A (en) | Towards the MAR Tracing Registration method and system based on figure optimization SLAM | |
| CN105096386A (en) | Method for automatically generating geographic maps for large-range complex urban environment | |
| JP2022531679A (en) | Device positioning methods, devices, and devices | |
| CN105069809A (en) | Camera positioning method and system based on planar mixed marker | |
| CN104036524A (en) | Fast target tracking method with improved SIFT algorithm | |
| CN110675453B (en) | Self-positioning method for moving target in known scene | |
| CN105279769A (en) | Hierarchical particle filtering tracking method combined with multiple features | |
| CN107025449A (en) | A kind of inclination image linear feature matching process of unchanged view angle regional area constraint | |
| CN110136174B (en) | Target object tracking method and device | |
| CN108537844A (en) | A kind of vision SLAM winding detection methods of fusion geological information | |
| CN105427333A (en) | Real-time registration method of video sequence image, system and shooting terminal | |
| CN111126205A (en) | Optical remote sensing image airplane target detection method based on rotary positioning network | |
| Yang et al. | CubeSLAM: Monocular 3D object detection and SLAM without prior models | |
| Liu et al. | Autonomous road detection and modeling for UGVs using vision-laser data fusion | |
| CN114758086B (en) | Method and device for constructing urban road information model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190618 |
|
| WD01 | Invention patent application deemed withdrawn after publication |