CN109583393A - A kind of lane line endpoints recognition methods and device, equipment, medium - Google Patents
A kind of lane line endpoints recognition methods and device, equipment, medium Download PDFInfo
- Publication number
- CN109583393A CN109583393A CN201811478746.8A CN201811478746A CN109583393A CN 109583393 A CN109583393 A CN 109583393A CN 201811478746 A CN201811478746 A CN 201811478746A CN 109583393 A CN109583393 A CN 109583393A
- Authority
- CN
- China
- Prior art keywords
- lane line
- bounding box
- endpoints
- target detection
- sample 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000001514 detection method Methods 0.000 claims abstract description 77
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 26
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 19
- 238000003709 image segmentation Methods 0.000 claims description 23
- 230000011218 segmentation Effects 0.000 claims description 10
- 238000004891 communication Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 14
- 230000008569 process Effects 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000012549 training Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 238000010801 machine learning Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
- G06T2207/30256—Lane; Road marking
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
This application discloses a kind of lane line endpoints recognition methods and device, equipment, media.This method includes at least: according to the lane line for including in lane line sample image, define the bounding box for confining the lane line endpoints for including in lane line sample image, according to the definition of the bounding box to lane line endpoints, utilize the algorithm of target detection based on convolutional neural networks, target detection is carried out in images to be recognized, to identify the bounding box of lane line endpoints, according to target detection as a result, determining the position of the lane line endpoints.The application is returned based on bounding box by being that lane line endpoints define suitable bounding box according to lane line and carries out target detection, helped accurately to identify lane line endpoints in images to be recognized, determine its position.
Description
Technical field
This application involves machine learning techniques field more particularly to a kind of lane line endpoints recognition methods and device, equipment,
Medium.
Background technique
With the rapid development of machine learning techniques, the depth model based on deep learning is also used in more and more
Side, including critical point detection field.Bigger difference may be had by closing application effect of the measuring point detection in different scenes.
For example, under the scene of face critical point detection, due to face key point be often positioned in it is among image or close
Intermediate region, helps to obtain accurate testing result；Under the scape of high-precision map fabricating yard, an important task
It is the endpoint for extracting dotted line lane line, in the prior art, also by the scheme of face critical point detection, for detecting dotted line vehicle
The endpoint of diatom.
But since the endpoint of dotted line lane line is usually located at the region of image border, be often not readily available compared with
For accurate testing result.
Summary of the invention
The embodiment of the present application provides a kind of lane line endpoints recognition methods and device, equipment, medium, to solve existing skill
Following technical problem in art: existing critical point detection scheme is not readily available accurate dotted line lane line end-point detection
As a result.
The embodiment of the present application adopts the following technical solutions:
A kind of lane line endpoints recognition methods, comprising:
According to the lane line for including in lane line sample image, definition includes for confining in the lane line sample image
Lane line endpoints bounding box；
According to the definition of the bounding box to the lane line endpoints, calculated using the target detection based on convolutional neural networks
Method carries out target detection in images to be recognized, to identify the bounding box of lane line endpoints；
According to the target detection as a result, determining the position of the lane line endpoints.
Optionally, it is described according to the target detection as a result, determine the position of the lane line endpoints, specifically include:
Using image segmentation algorithm, image segmentation is carried out in the bounding box identified, to carry out prospect and background
Segmentation；
According to the target detection as a result, and described image segmentation as a result, determining the position of the lane line endpoints.
Optionally, described according to the lane line for including in lane line sample image, it defines for confining the lane line sample
The bounding box for the lane line endpoints for including in this image, specifically includes:
Define the bounding box for confining the lane line for including in lane line sample image；
According to the width and/or height of the bounding box of the lane line, definition is wrapped for confining in the lane line sample image
The bounding box of the lane line endpoints contained.
Optionally, the width and/or height of the bounding box according to the lane line is defined for confining the lane line sample
The bounding box for the lane line endpoints for including in this image, further includes:
According to preset size threshold, the full-size of the bounding box of the lane line endpoints is limited.
Optionally, the shape of the bounding box of the lane line endpoints is square, the side length of the square be not more than with
Under minimum value in several persons: the size threshold, the width of the bounding box of the lane line, height.
Optionally, definition of the basis to the bounding box of the lane line endpoints, using based on convolutional neural networks
Algorithm of target detection carries out target detection in images to be recognized, specifically includes:
Obtain multiple lane line sample images at least one lane scene；
The lane line and lane line endpoints that include in the multiple lane line sample image are labeled respectively；
According to the multiple lane line sample image and its mark, and the definition of the bounding box to lane line endpoints, benefit
With the algorithm of target detection based on convolutional neural networks, training bounding box regression model；
Using the bounding box regression model trained, target detection is carried out in images to be recognized.
Optionally, it is described according to the target detection as a result, determine the position of the lane line endpoints, specifically include:
According to the central point of the bounding box of the lane line endpoints identified, the position of the lane line endpoints is determined.
Optionally, the lane line is dotted line lane line.
A kind of lane line endpoints identification device, comprising:
Definition module is defined according to the lane line for including in lane line sample image for confining the lane line sample
The bounding box for the lane line endpoints for including in image；
Identification module utilizes the mesh based on convolutional neural networks according to the definition of the bounding box to the lane line endpoints
Detection algorithm is marked, carries out target detection, in images to be recognized to identify the bounding box of lane line endpoints；
Determination module, according to the target detection as a result, determining the position of the lane line endpoints.
Optionally, the determination module is according to the target detection as a result, determining the position of the lane line endpoints, specifically
Include:
The determination module utilize image segmentation algorithm, carry out image segmentation in the bounding box identified, with into
The segmentation of row prospect and background；
According to the target detection as a result, and described image segmentation as a result, determining the position of the lane line endpoints.
Optionally, for the definition module according to the lane line for including in lane line sample image, definition is described for confining
The bounding box for the lane line endpoints for including in lane line sample image, specifically includes:
The definition module defines the bounding box for confining the lane line for including in lane line sample image；
According to the width and/or height of the bounding box of the lane line, definition is wrapped for confining in the lane line sample image
The bounding box of the lane line endpoints contained.
Optionally, the definition module defines described for confining according to the width and/or height of the bounding box of the lane line
The bounding box for the lane line endpoints for including in lane line sample image, further includes:
The definition module limits the full-size of the bounding box of the lane line endpoints according to preset size threshold.
Optionally, the shape of the bounding box of the lane line endpoints is square, the side length of the square be not more than with
Under minimum value in several persons: the size threshold, the width of the bounding box of the lane line, height.
Optionally, the identification module is according to the definition of the bounding box to the lane line endpoints, using based on convolution mind
Algorithm of target detection through network carries out target detection in images to be recognized, specifically includes:
The identification module obtains multiple lane line sample images at least one lane scene；
The lane line and lane line endpoints that include in the multiple lane line sample image are labeled respectively；
According to the multiple lane line sample image and its mark, and the definition of the bounding box to lane line endpoints, benefit
With the algorithm of target detection based on convolutional neural networks, training bounding box regression model；
Using the bounding box regression model trained, target detection is carried out in images to be recognized.
Optionally, the determination module is according to the target detection as a result, determining the position of the lane line endpoints, specifically
Include:
The central point of the bounding box for the lane line endpoints that the determination module is identified according to determines the lane line end
The position of point.
Optionally, the lane line is dotted line lane line.
A kind of lane line endpoints identification equipment, comprising:
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
A processor executes so that at least one described processor can:
According to the lane line for including in lane line sample image, definition includes for confining in the lane line sample image
Lane line endpoints bounding box；
According to the definition of the bounding box to the lane line endpoints, calculated using the target detection based on convolutional neural networks
Method carries out target detection in images to be recognized, to identify the bounding box of lane line endpoints；
According to the target detection as a result, determining the position of the lane line endpoints.
A kind of lane line endpoints identification nonvolatile computer storage media, is stored with computer executable instructions, described
Computer executable instructions setting are as follows:
According to the lane line for including in lane line sample image, definition includes for confining in the lane line sample image
Lane line endpoints bounding box；
According to the definition of the bounding box to the lane line endpoints, calculated using the target detection based on convolutional neural networks
Method carries out target detection in images to be recognized, to identify the bounding box of lane line endpoints；
According to the target detection as a result, determining the position of the lane line endpoints.
At least one above-mentioned technical solution that the embodiment of the present application uses can reach following the utility model has the advantages that by according to vehicle
Diatom be lane line endpoints define suitable bounding box, based on bounding box return carry out target detection, facilitate accurately to
It identifies and identifies lane line endpoints in image, determine its position.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen
Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings:
Fig. 1 is a kind of flow diagram for lane line endpoints recognition methods that some embodiments of the present application provide；
Fig. 2 is a kind of dotted line lane line endpoints and its bounding box schematic diagram that some embodiments of the present application provide；
Fig. 3 is a kind of detailed process signal for the above-mentioned lane line endpoints recognition methods that some embodiments of the present application provide
Figure；
Fig. 4 is a kind of structure for lane line endpoints identification device corresponding to Fig. 1 that some embodiments of the present application provide
Schematic diagram；
Fig. 5 is the structure that a kind of lane line endpoints corresponding to Fig. 1 that some embodiments of the present application provide identify equipment
Schematic diagram.
Specific embodiment
To keep the purposes, technical schemes and advantages of the application clearer, below in conjunction with the application specific embodiment and
Technical scheme is clearly and completely described in corresponding attached drawing.Obviously, described embodiment is only the application one
Section Example, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall in the protection scope of this application.
In some embodiments of the present application, the definition mode of the bounding box for lane line endpoints is proposed, is defined
Bounding box be different from the bounding box of the existing object (automobile, aircraft etc.) for occupying boxed area, existing boundaries frame
It is generally necessary to object edge is relatively accurately approached with rectangle small as far as possible, and application-defined bounding box can not have this
Limitation, on the one hand because object is endpoint, on the other hand because bounding box is referred to corresponding lane line and is defined.
The bounding box of lane line endpoints based on definition, can be by bounding box regression training model, in lane line
The bounding box of lane line endpoints is identified in images to be recognized other than sample image and sample, and determines lane line endpoints
Position.In order to improve identification accuracy, the processing such as image segmentation, image enhancement can also be further carried out, then synthetically sentence
Determine the position of lane line endpoints.The scheme of the application is described in detail below.
Fig. 1 is a kind of flow diagram for lane line endpoints recognition methods that some embodiments of the present application provide.At this
In process, for equipment angle, executing subject can be one or more and calculate equipment, for example, individual machine study clothes
Business device, machine learning server cluster, image segmentation server etc., for program angle, executing subject correspondingly be can be
It is equipped on these and calculates the program in equipment, for example, neural net model establishing platform, image processing platform etc..
Process in Fig. 1 may comprise steps of:
S102: it according to the lane line for including in lane line sample image, defines for confining the lane line sample image
In include lane line endpoints bounding box.
In some embodiments of the present application, generally, the lane line for including in each lane line image have one or
Two lane line endpoints.Lane line can be miscellaneous, depends on practical identification and needs, for example can be dotted line lane line
Perhaps solid line lane line can be bicycle diatom perhaps two-way traffic line can be white lane line or yellow lane line etc..?
In practical application, dotted line lane line is more difficult accurately to identify its lane line end since discrete a plurality of line segment is constituted
Point, and the scheme of the application can also reach preferable recognition effect for dotted line lane line endpoints, below some embodiment masters
It to be illustrated so that the lane line in Fig. 1 is dotted line lane line as an example.
In some embodiments of the present application, lane line sample image have it is multiple, for training corresponding machine learning mould
Type, the machine learning model are at least used to detect the bounding box of lane line endpoints based on the definition to bounding box.It can be according to more
Kind of factor, defines the bounding box of lane line endpoints, which is such as lane line itself, other objects compares in image
Example, preset size threshold, lane line endpoints degree etc. at a distance from image border.
S104: it according to the definition of the bounding box to the lane line endpoints, is examined using the target based on convolutional neural networks
Method of determining and calculating (refers mainly to the image other than sample image, for example, freshly harvested pavement of road figure to be identified in images to be recognized
As etc.) in carry out target detection, to identify the bounding box of lane line endpoints.
In some embodiments of the present application, images to be recognized is carried out at subregional part based on convolutional neural networks
Reason can be more quasi- further according to multiple Local treatments as a result, obtaining disposed of in its entirety as a result, for the lesser lane line endpoints of target
Really extract bounding box.
S106: according to the target detection as a result, determining the position of the lane line endpoints.
In some embodiments of the present application, by target detection, after the bounding box for identifying lane line endpoints, Ke Yizhi
It connects according to the bounding box, the position of lane line endpoints is determined, for example, by the central point position or the side of the bounding box
Arbitrary point position in the intermediate region of boundary's frame, is determined as the position of lane line endpoints；Alternatively, can also be using other calculations
Method further identifies in the bounding box, to determine the position of lane line endpoints.
It is returned by being that lane line endpoints define suitable bounding box according to lane line based on bounding box by the method for Fig. 1
Return carry out target detection, helps accurately to identify lane line endpoints in images to be recognized, determine its position.
Method based on Fig. 1, some embodiments of the present application additionally provide some specific embodiments of this method, and
Expansion scheme is illustrated below.
In some embodiments of the present application, for step S106, it is described according to the target detection as a result, determine should
The position of lane line endpoints, for example may include: to carry out image in the bounding box identified using image segmentation algorithm
Segmentation, to carry out the segmentation of prospect and background；According to described image segmentation as a result, the target detection result and institute
The combination for stating the result of image segmentation determines the position of the lane line endpoints.By taking latter approach as an example, for example, can will know
Not Chu at least one foreground pixel coordinate for being obtained with image segmentation of center point coordinate of bounding box of lane line be averaged,
Obtained coordinate is determined as to the position of the lane line endpoints.
Foreground pixel such as can be lane line pixel, can be more specifically lane line edge pixel, background pixel can
Think the road surface pixel other than lane line.Image segmentation can be using the model realization trained accordingly, if the model training
The mark of used sample (bounding box image of lane line etc.) is accurate enough (for example, being accurate to lane line endpoints pixel),
Then may directly it be split using lane line endpoints as prospect.For example, image, semantic partitioning algorithm can be used, image is carried out
Segmentation helps to obtain more accurate segmentation result.
Algorithm used by above-mentioned target detection and image segmentation is not specifically limited here, can use existing algorithm
Or it is adapted to the algorithm etc. that practical scene improves, it is able to achieve described effect.For example, using MASK RCNN
Algorithm etc..
In some embodiments of the present application, it is assumed that the bounding box that lane line is defined according to lane line itself, then for step
Rapid S102, it is described according to the lane line for including in lane line sample image, it defines for confining in the lane line sample image
The bounding box for the lane line endpoints for including, for example may include: definition for confining the lane for including in lane line sample image
The bounding box of line；According to the width and/or height of the bounding box of the lane line, define for confining in the lane line sample image
The bounding box for the lane line endpoints for including.The bounding box of lane line can refer to the bounding box of each line segment in dotted line lane line,
It can also refer to the bounding box of lane line entirety, the method for determination of the bounding box of lane line is referred to: to automobile, aircraft etc.
The method of determination of the bounding box of the obvious object of profile.
Lane line endpoints lane line where with it be it is directly related, in the bounding box for defining lane line endpoints, referring to should
The size of place lane line is more reasonable, helps so that in every lane line sample image, defined bounding box may not
It is onesize, but compared to lane line where it, size is suitable, is conducive to the feature for more effectively extracting end region.
For example, the bounding box of lane line can be determined first, obtain the width and height of the bounding box, then wide and senior middle school take compared with
Small value defines the width and/or height of the bounding box of lane line endpoints according to the smaller value.
Further, it is contemplated that, then may according to its bounding box when accounting is relatively large in the picture for lane line itself
Lane line endpoints bounding box bigger than normal is defined, it, can be with predetermined size threshold value, for limiting the vehicle for this problem
The full-size of the bounding box of road line endpoints.Size threshold such as can be the identical value for the unified setting of each samples pictures
(for example, being set as being no more than 50 pixels etc. for the wide, high of bounding box), it is also possible to be adapted to each samples pictures
The self-adapting changeable value that size is set separately is (for example, width, the high width being set as no more than corresponding image for bounding box
With the 20 of senior middle school minimum value/first-class).
In some embodiments of the present application, the bounding box of lane line endpoints can be defined as to rectangle, but can also be with
The bounding box definition of lane line endpoints is square, square symmetry is better, can also reduce the size ginseng of bounding box
Several numbers (the two wide and high parameters merge into this parameter of side length), economize on resources.In addition, in certain images, lane
Line endpoints may be very close to image border, in this case, if bounding box definition is square, the side length of square can
Can be too small, it is unfavorable for extracting the feature in bounding box, in such a case, it is possible to bounding box is defined as rectangle, on side
The vertical direction of edge extracts feature as far as possible.
More intuitively, some embodiments of the present application provide a kind of dotted line lane line endpoints and its bounding box schematic diagram,
As shown in Figure 2.Fig. 2 shows dotted line lane lines, and illustratively denote the dotted line lane line with dashed square box
The bounding box of two endpoints respectively, and the central point (being indicated with cross) of bounding box can be considered as the endpoint.
In some embodiments of the present application, for step S102, bounding box of the basis to the lane line endpoints
Definition carry out target detection in images to be recognized using the algorithm of target detection based on convolutional neural networks, such as can be with
It include: the multiple lane line sample images obtained at least one lane scene；To being wrapped in the multiple lane line sample image
The lane line and lane line endpoints contained is labeled respectively；According to the multiple lane line sample image and its mark, and
Definition to the bounding box of lane line endpoints, using the algorithm of target detection based on convolutional neural networks, training bounding box is returned
Model；Using the bounding box regression model trained, target detection is carried out in images to be recognized.
Certain features that lane line image under different lane scenes may have scene to limit distinguish lane lane
Scene facilitates subsequent more accurately identification lane line endpoints.Lane scene can identify requirement definition according to practical, for example, single
Lane scene, crossroad scene, the crossing scene that turns around, actual situation two-wire scene, two-way multilane scene, bend scene etc., this
In be not specifically limited, only citing help understand.
According to explanation above, some embodiments of the present application additionally provide one kind of above-mentioned lane line endpoints recognition methods
Detailed process, as shown in Figure 3.
Step in Fig. 3 may comprise steps of:
S302: the great amount of samples image under various lane scenes including dotted line lane line is collected.
S304: according to lane scene, to dotted line lane line itself and it includes lane line endpoints be labeled.
S306: are defined by one and is directed to lane according to corresponding dotted line lane line size itself for each lane line endpoints
The bounding box of line endpoints, being specifically defined is: the bounding box is square, and the side length of the bounding box is preset size threshold, right
The width of the bounding box for the dotted line lane line answered and the minimum value of senior middle school, for example it is expressed as min (min (w, h), 50), min () table
Show and be minimized function, w and h respectively indicate the width and height, and 50 indicate the size threshold.
S308: utilizing algorithm of target detection, identifies the bounding box of lane line endpoints, the central point of the bounding box identified can
To be considered as lane line endpoints.
S310: in the bounding box identified, image, semantic partitioning algorithm is further utilized, prospect and back are partitioned into
Scape, wherein prospect is considered as lane line endpoints pixel, and background is considered as other pixels.
S312: according to image, semantic segmentation result, lane line endpoints are extracted.
S314: combining target testing result and image, semantic segmentation result (for example, the two coordinate is averaged), most
The position of lane line endpoints is determined eventually.
Based on same thinking, some embodiments of the present application additionally provide the corresponding device of the above method, equipment and non-
Volatile computer storage medium.
Fig. 4 is a kind of structure for lane line endpoints identification device corresponding to Fig. 1 that some embodiments of the present application provide
Schematic diagram, the device include:
Definition module 401 is defined according to the lane line for including in lane line sample image for confining the lane line sample
The bounding box for the lane line endpoints for including in this image；
Identification module 402, according to the definition of the bounding box to the lane line endpoints, using based on convolutional neural networks
Algorithm of target detection carries out target detection in images to be recognized, to identify the bounding box of lane line endpoints；
Determination module 403, according to the target detection as a result, determining the position of the lane line endpoints.
Optionally, the determination module 403 according to the target detection as a result, determine the position of the lane line endpoints,
It specifically includes:
The determination module 403 utilizes image segmentation algorithm, carries out image segmentation in the bounding box identified, with
The segmentation of carry out prospect and background；
According to the target detection as a result, and described image segmentation as a result, determining the position of the lane line endpoints.
Optionally, the definition module 401 is defined according to the lane line for including in lane line sample image for confining
The bounding box for stating the lane line endpoints for including in lane line sample image, specifically includes:
The definition module 401 defines the bounding box for confining the lane line for including in lane line sample image；
According to the width and/or height of the bounding box of the lane line, definition is wrapped for confining in the lane line sample image
The bounding box of the lane line endpoints contained.
Optionally, the definition module 401 is defined according to the width and/or height of the bounding box of the lane line for confining
The bounding box for the lane line endpoints for including in the lane line sample image, further includes:
The definition module 401 limits the maximum ruler of the bounding box of the lane line endpoints according to preset size threshold
It is very little.
Optionally, the shape of the bounding box of the lane line endpoints is square, the side length of the square be not more than with
Under minimum value in several persons: the size threshold, the width of the bounding box of the lane line, height.
Optionally, the identification module 402 is according to the definition of the bounding box to the lane line endpoints, using being based on convolution
The algorithm of target detection of neural network carries out target detection in images to be recognized, specifically includes:
The identification module 402 obtains multiple lane line sample images at least one lane scene；
The lane line and lane line endpoints that include in the multiple lane line sample image are labeled respectively；
According to the multiple lane line sample image and its mark, and the definition of the bounding box to lane line endpoints, benefit
With the algorithm of target detection based on convolutional neural networks, training bounding box regression model；
Using the bounding box regression model trained, target detection is carried out in images to be recognized.
Optionally, the determination module 403 according to the target detection as a result, determine the position of the lane line endpoints,
It specifically includes:
The determination module 403 determines the lane according to the central point of the bounding box of the lane line endpoints identified
The position of line endpoints.
Optionally, the lane line is dotted line lane line.
Fig. 5 is the structure that a kind of lane line endpoints corresponding to Fig. 1 that some embodiments of the present application provide identify equipment
Schematic diagram, the equipment include:
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
A processor executes so that at least one described processor can:
According to the lane line for including in lane line sample image, definition includes for confining in the lane line sample image
Lane line endpoints bounding box；
According to the definition of the bounding box to the lane line endpoints, calculated using the target detection based on convolutional neural networks
Method carries out target detection in images to be recognized, to identify the bounding box of lane line endpoints；
According to the target detection as a result, determining the position of the lane line endpoints.
A kind of lane line endpoints identification non-volatile computer corresponding to Fig. 1 that some embodiments of the present application provide is deposited
Storage media is stored with computer executable instructions, computer executable instructions setting are as follows:
According to the lane line for including in lane line sample image, definition includes for confining in the lane line sample image
Lane line endpoints bounding box；
According to the definition of the bounding box to the lane line endpoints, calculated using the target detection based on convolutional neural networks
Method carries out target detection in images to be recognized, to identify the bounding box of lane line endpoints；
According to the target detection as a result, determining the position of the lane line endpoints.
Various embodiments are described in a progressive manner in the application, same and similar part between each embodiment
It may refer to each other, each embodiment focuses on the differences from other embodiments.Especially for device, set
For standby and media embodiment, since it is substantially similar to the method embodiment, so be described relatively simple, related place referring to
The part of embodiment of the method illustrates.
Device, equipment and medium provided by the embodiments of the present application and method be it is one-to-one, therefore, device, equipment and
The advantageous effects that medium also has corresponding method similar, due to above to the advantageous effects of method into
Go detailed description, therefore, the advantageous effects of which is not described herein again device, equipment and medium.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including described want
There is also other identical elements in the process, method of element, commodity or equipment.
The above description is only an example of the present application, is not intended to limit this application.For those skilled in the art
For, various changes and changes are possible in this application.All any modifications made within the spirit and principles of the present application are equal
Replacement, improvement etc., should be included within the scope of the claims of this application.
Claims (10)
1. a kind of lane line endpoints recognition methods characterized by comprising
According to the lane line for including in lane line sample image, define for confining the vehicle for including in the lane line sample image
The bounding box of road line endpoints；
According to the definition of the bounding box to the lane line endpoints, using the algorithm of target detection based on convolutional neural networks,
Target detection is carried out in images to be recognized, to identify the bounding box of lane line endpoints；
According to the target detection as a result, determining the position of the lane line endpoints.
2. the method as described in claim 1, which is characterized in that it is described according to the target detection as a result, determining the lane
The position of line endpoints, specifically includes:
Using image segmentation algorithm, image segmentation is carried out in the bounding box identified, to carry out point of prospect and background
It cuts；
According to the target detection as a result, and described image segmentation as a result, determining the position of the lane line endpoints.
3. the method as described in claim 1, which is characterized in that it is described according to the lane line for including in lane line sample image,
The bounding box for confining the lane line endpoints for including in the lane line sample image is defined, is specifically included:
Define the bounding box for confining the lane line for including in lane line sample image；
According to the width and/or height of the bounding box of the lane line, definition includes for confining in the lane line sample image
The bounding box of lane line endpoints.
4. method as claimed in claim 3, which is characterized in that the width and/or height of the bounding box according to the lane line,
Define the bounding box for confining the lane line endpoints for including in the lane line sample image, further includes:
According to preset size threshold, the full-size of the bounding box of the lane line endpoints is limited.
5. method as claimed in claim 4, which is characterized in that the shape of the bounding box of the lane line endpoints is square,
The side length of the square is no more than the minimum value in following several persons: the size threshold, the lane line bounding box width,
It is high.
6. a kind of lane line endpoints identification device characterized by comprising
Definition module is defined according to the lane line for including in lane line sample image for confining the lane line sample image
In include lane line endpoints bounding box；
Identification module is examined according to the definition of the bounding box to the lane line endpoints using the target based on convolutional neural networks
Method of determining and calculating carries out target detection in images to be recognized, to identify the bounding box of lane line endpoints；
Determination module, according to the target detection as a result, determining the position of the lane line endpoints.
7. device as claimed in claim 6, which is characterized in that the determination module is according to the target detection as a result, sentencing
The position of the fixed lane line endpoints, specifically includes:
The determination module utilizes image segmentation algorithm, carries out image segmentation in the bounding box identified, before carrying out
The segmentation of scape and background；
According to the target detection as a result, and described image segmentation as a result, determining the position of the lane line endpoints.
8. device as claimed in claim 6, which is characterized in that the definition module includes according in lane line sample image
Lane line defines the bounding box for confining the lane line endpoints for including in the lane line sample image, specifically includes:
The definition module defines the bounding box for confining the lane line for including in lane line sample image；
According to the width and/or height of the bounding box of the lane line, definition includes for confining in the lane line sample image
The bounding box of lane line endpoints.
9. device as claimed in claim 8, which is characterized in that the definition module is according to the width of the bounding box of the lane line
And/or it is high, define the bounding box for confining the lane line endpoints for including in the lane line sample image, further includes:
The definition module limits the full-size of the bounding box of the lane line endpoints according to preset size threshold.
10. a kind of lane line endpoints identify equipment characterized by comprising
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
Manage device execute so that at least one described processor can:
According to the lane line for including in lane line sample image, define for confining the vehicle for including in the lane line sample image
The bounding box of road line endpoints；
According to the definition of the bounding box to the lane line endpoints, using the algorithm of target detection based on convolutional neural networks,
Target detection is carried out in images to be recognized, to identify the bounding box of lane line endpoints；
According to the target detection as a result, determining the position of the lane line endpoints.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811478746.8A CN109583393B (en) | 2018-12-05 | 2018-12-05 | Lane line end point identification method and device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811478746.8A CN109583393B (en) | 2018-12-05 | 2018-12-05 | Lane line end point identification method and device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109583393A true CN109583393A (en) | 2019-04-05 |
| CN109583393B CN109583393B (en) | 2023-08-11 |
Family
ID=65926316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811478746.8A Active CN109583393B (en) | 2018-12-05 | 2018-12-05 | Lane line end point identification method and device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109583393B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110688971A (en) * | 2019-09-30 | 2020-01-14 | 上海商汤临港智能科技有限公司 | Method, device and equipment for detecting dotted lane line |
| CN112053407A (en) * | 2020-08-03 | 2020-12-08 | 杭州电子科技大学 | Automatic lane line detection method based on AI technology in traffic law enforcement image |
| CN113449648A (en) * | 2021-06-30 | 2021-09-28 | 北京纵目安驰智能科技有限公司 | Method, system, equipment and computer readable storage medium for detecting indicator line |
| WO2022028383A1 (en) * | 2020-08-06 | 2022-02-10 | 长沙智能驾驶研究院有限公司 | Lane line labeling method, detection model determining method, lane line detection method, and related device |
| EP4047520A1 (en) * | 2021-02-23 | 2022-08-24 | Beijing Tusen Zhitu Technology Co., Ltd. | Method and apparatus for detecting corner points of lane lines, electronic device and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036246A (en) * | 2014-06-10 | 2014-09-10 | 电子科技大学 | Lane line positioning method based on multi-feature fusion and polymorphism mean value |
| CN104036253A (en) * | 2014-06-20 | 2014-09-10 | 智慧城市系统服务（中国）有限公司 | Lane line tracking method and lane line tracking system |
| CN105740782A (en) * | 2016-01-25 | 2016-07-06 | 北京航空航天大学 | Monocular vision based driver lane-changing process quantization method |
| US20160229399A1 (en) * | 2015-02-10 | 2016-08-11 | Honda Motor Co., Ltd. | Vehicle travel support system and vehicle travel support method |
| CN106663207A (en) * | 2014-10-29 | 2017-05-10 | 微软技术许可有限责任公司 | Whiteboard and document image detection method and system |
| CN106682646A (en) * | 2017-01-16 | 2017-05-17 | 北京新能源汽车股份有限公司 | Method and apparatus for recognizing lane line |
| CN108090456A (en) * | 2017-12-27 | 2018-05-29 | 北京初速度科技有限公司 | A kind of Lane detection method and device |
| CN108545019A (en) * | 2018-04-08 | 2018-09-18 | 多伦科技股份有限公司 | A kind of safety driving assist system and method based on image recognition technology |
| US20180293447A1 (en) * | 2017-04-05 | 2018-10-11 | Denso Corporation | Road parameter calculator |
-
2018
- 2018-12-05 CN CN201811478746.8A patent/CN109583393B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036246A (en) * | 2014-06-10 | 2014-09-10 | 电子科技大学 | Lane line positioning method based on multi-feature fusion and polymorphism mean value |
| CN104036253A (en) * | 2014-06-20 | 2014-09-10 | 智慧城市系统服务（中国）有限公司 | Lane line tracking method and lane line tracking system |
| CN106663207A (en) * | 2014-10-29 | 2017-05-10 | 微软技术许可有限责任公司 | Whiteboard and document image detection method and system |
| US20160229399A1 (en) * | 2015-02-10 | 2016-08-11 | Honda Motor Co., Ltd. | Vehicle travel support system and vehicle travel support method |
| CN105740782A (en) * | 2016-01-25 | 2016-07-06 | 北京航空航天大学 | Monocular vision based driver lane-changing process quantization method |
| CN106682646A (en) * | 2017-01-16 | 2017-05-17 | 北京新能源汽车股份有限公司 | Method and apparatus for recognizing lane line |
| US20180293447A1 (en) * | 2017-04-05 | 2018-10-11 | Denso Corporation | Road parameter calculator |
| CN108090456A (en) * | 2017-12-27 | 2018-05-29 | 北京初速度科技有限公司 | A kind of Lane detection method and device |
| CN108545019A (en) * | 2018-04-08 | 2018-09-18 | 多伦科技股份有限公司 | A kind of safety driving assist system and method based on image recognition technology |
Non-Patent Citations (12)
| Title |
|---|
| BRODY HUVAL等: "An Empirical Evaluation of Deep Learning on Highway Driving", 《ARXIV.ORG》 * |
| BRODY HUVAL等: "An Empirical Evaluation of Deep Learning on Highway Driving", 《ARXIV.ORG》, 17 April 2015 (2015-04-17), pages 1 - 7, XP055392563 * |
| CEM ÜNSALAN等: "Road Network Detection Using Probabilistic and Graph Theoretical Methods", 《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》 * |
| CEM ÜNSALAN等: "Road Network Detection Using Probabilistic and Graph Theoretical Methods", 《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》, vol. 50, no. 11, 17 April 2012 (2012-04-17), pages 4441 - 4453, XP011472361, DOI: 10.1109/TGRS.2012.2190078 * |
| 宋洪军等: "基于车道线检测与图像拐点的道路能见度估计", 《计算机应用》 * |
| 宋洪军等: "基于车道线检测与图像拐点的道路能见度估计", 《计算机应用》, no. 12, 26 January 2013 (2013-01-26), pages 3397 - 3403 * |
| 张晓林: "基于深度学习的证件照人脸识别方法", 《计算机系统应用》 * |
| 张晓林: "基于深度学习的证件照人脸识别方法", 《计算机系统应用》, vol. 27, no. 5, 15 May 2018 (2018-05-15), pages 203 - 208 * |
| 郭剑鹰等: "一种基于机器学习的ADAS车道类型判别方法", 《汽车电器》 * |
| 郭剑鹰等: "一种基于机器学习的ADAS车道类型判别方法", 《汽车电器》, 20 December 2017 (2017-12-20), pages 22 - 24 * |
| 金月等: "基于暗通道先验的实时低能见度辅助驾驶系统", 《华东理工大学学报(自然科学版)HTTPS://DOI.ORG/10.14135/J.CNKI.1006-3080.20180118002》 * |
| 金月等: "基于暗通道先验的实时低能见度辅助驾驶系统", 《华东理工大学学报(自然科学版)HTTPS://DOI.ORG/10.14135/J.CNKI.1006-3080.20180118002》, 19 June 2018 (2018-06-19), pages 1 - 8 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110688971A (en) * | 2019-09-30 | 2020-01-14 | 上海商汤临港智能科技有限公司 | Method, device and equipment for detecting dotted lane line |
| WO2021063228A1 (en) * | 2019-09-30 | 2021-04-08 | 上海商汤临港智能科技有限公司 | Dashed lane line detection method and device, and electronic apparatus |
| CN110688971B (en) * | 2019-09-30 | 2022-06-24 | 上海商汤临港智能科技有限公司 | Method, device and equipment for detecting dotted lane line |
| CN112053407A (en) * | 2020-08-03 | 2020-12-08 | 杭州电子科技大学 | Automatic lane line detection method based on AI technology in traffic law enforcement image |
| CN112053407B (en) * | 2020-08-03 | 2024-04-09 | 杭州电子科技大学 | Automatic lane line detection method based on AI technology in traffic law enforcement image |
| WO2022028383A1 (en) * | 2020-08-06 | 2022-02-10 | 长沙智能驾驶研究院有限公司 | Lane line labeling method, detection model determining method, lane line detection method, and related device |
| EP4047520A1 (en) * | 2021-02-23 | 2022-08-24 | Beijing Tusen Zhitu Technology Co., Ltd. | Method and apparatus for detecting corner points of lane lines, electronic device and storage medium |
| CN113449648A (en) * | 2021-06-30 | 2021-09-28 | 北京纵目安驰智能科技有限公司 | Method, system, equipment and computer readable storage medium for detecting indicator line |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109583393B (en) | 2023-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109583393A (en) | A kind of lane line endpoints recognition methods and device, equipment, medium | |
| CN110287276A (en) | High-precision map updating method, device and storage medium | |
| WO2018068653A1 (en) | Point cloud data processing method and apparatus, and storage medium | |
| Mattyus et al. | Enhancing road maps by parsing aerial images around the world | |
| JP6259928B2 (en) | Lane data processing method, apparatus, storage medium and equipment | |
| KR102106359B1 (en) | Method and apparatus for identifying intersections in an electronic map | |
| Huang et al. | Spatial-temproal based lane detection using deep learning | |
| CN108875595A (en) | A kind of Driving Scene object detection method merged based on deep learning and multilayer feature | |
| US20130243343A1 (en) | Method and device for people group detection | |
| CN110119148A (en) | A kind of six-degree-of-freedom posture estimation method, device and computer readable storage medium | |
| CN109584294A (en) | A kind of road surface data reduction method and apparatus based on laser point cloud | |
| CN103206957B (en) | The lane detection and tracking method of vehicular autonomous navigation | |
| CN106327576B (en) | A kind of City scenarios method for reconstructing and system | |
| CN109858349A (en) | A kind of traffic sign recognition method and its device based on improvement YOLO model | |
| CN114581744A (en) | Image target detection method, system, equipment and storage medium | |
| CN115203352A (en) | Lane level positioning method and device, computer equipment and storage medium | |
| CN110299063B (en) | Visual display method and device for trajectory data | |
| CN109190687A (en) | A kind of nerve network system and its method for identifying vehicle attribute | |
| CN110188607A (en) | A kind of the traffic video object detection method and device of multithreads computing | |
| CN109711341A (en) | A kind of virtual lane line recognition methods and device, equipment, medium | |
| CN109800684A (en) | The determination method and device of object in a kind of video | |
| CN113269806B (en) | Method, device and processor for measuring blood flow inside blood vessel | |
| CN116129386A (en) | Method, system and computer readable medium for detecting a travelable region | |
| CN117011692A (en) | Road identification method and related device | |
| Kiran et al. | Automatic hump detection and 3D view generation from a single road image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address |
Address after: Room 108-27, Building 1, No. 611 Yunxiu South Road, Wuyang Street, Deqing County, Huzhou City, Zhejiang Province, 313200 (Moganshan National High tech Zone) Patentee after: Kuandong (Huzhou) Technology Co.,Ltd. Address before: 811, 8 / F, 101, 3-8 / F, building 17, rongchuang Road, Chaoyang District, Beijing 100012 Patentee before: KUANDENG (BEIJING) TECHNOLOGY Co.,Ltd. |
|
| CP03 | Change of name, title or address | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20190405 Assignee: Zhejiang Kuandong Yuntu Technology Co.,Ltd. Assignor: Kuandong (Huzhou) Technology Co.,Ltd. Contract record no.: X2024980001061 Denomination of invention: A lane line endpoint recognition method, device, equipment, and medium Granted publication date: 20230811 License type: Common License Record date: 20240119 |
|
| EE01 | Entry into force of recordation of patent licensing contract |