CN107122756A - A kind of complete non-structural road edge detection method - Google Patents

A kind of complete non-structural road edge detection method Download PDF

Info

Publication number
CN107122756A
CN107122756A CN201710327868.6A CN201710327868A CN107122756A CN 107122756 A CN107122756 A CN 107122756A CN 201710327868 A CN201710327868 A CN 201710327868A CN 107122756 A CN107122756 A CN 107122756A
Authority
CN
China
Prior art keywords
sequences
image
road
edge detection
complete
Prior art date
Application number
CN201710327868.6A
Other languages
Chinese (zh)
Inventor
不公告发明人
Original Assignee
南宁市正祥科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 南宁市正祥科技有限公司 filed Critical 南宁市正祥科技有限公司
Priority to CN201710327868.6A priority Critical patent/CN107122756A/en
Publication of CN107122756A publication Critical patent/CN107122756A/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00624Recognising scenes, i.e. recognition of a whole field of perception; recognising scene-specific objects
    • G06K9/00791Recognising scenes perceived from the perspective of a land vehicle, e.g. recognising lanes, obstacles or traffic signs on road scenes
    • G06K9/00798Recognition of lanes or road borders, e.g. of lane markings, or recognition of driver's driving pattern in relation to lanes perceived from the vehicle; Analysis of car trajectory relative to detected road

Abstract

The invention discloses a kind of complete non-structural road edge detection method, grey relational grade is introduced into complete unstructured road color images edge detection, the comparative sequences of three-dimensional component information generation by color space and the reference sequences formed by Sobel operators are calculated into its grey-relational degree, color component gray relation grades image is generated, road color images edge detection is then carried out.Approach for road detection of the present invention is independent of road shape and it is assumed that directly carries out rim detection to coloured image using Sobel operators, not only real-time is good, and edge detection results are clear, moreover it is possible to effectively reduce the noise of image.

Description

A kind of complete non-structural road edge detection method

Technical field

Present invention relates particularly to a kind of complete non-structural road edge detection method.

Background technology

One of key technology of intelligent vehicle autonomous driving system is the Road Detection of view-based access control model.In recent years, independently drive Systematic research is sailed, as computer and robot technology are continued to develop, development at full speed has also been obtained, wherein, how each Plant and accurately identify the problem that road ahead is autonomous driving system under Complex Natural Environment.It is higher for structuring degree Road, can typically be reduced to lane detection by the method for Road Detection.And unstructured road, because by season in natural environment The influence of section, weather or illumination etc., can all cause the change of road-surface features, come certain to unstructured road detection band Challenge.

The information that road how is obtained from coloured image is the key of the Road Detection of view-based access control model.Because road is in In real natural environment, the image that video camera is obtained often is influenceed and degenerated by various environmental factors.Further, since completely Some features of unstructured road image, are not natively it is obvious that this makes image procossing postorder mistake such as edge, profile The information extraction of journey is directly affected.

The content of the invention

The technical problem to be solved in the present invention is to provide a kind of complete non-structural road edge detection method.

A kind of complete non-structural road edge detection method, comprises the following steps:

S1:Image preprocessing:The video image collected is subjected to image preprocessing;

S2:Determine reference sequences and comparative sequences:

1)If reference sequences, comparative sequences , wherein N represents the calculation formula of the number, then incidence coefficient of component in reference sequences and sequence to be compared:

,

Wherein,Referred to as resolution ratio, value 0.205;Represent in k-th of moment comparative sequencesWithBetween Relative value, is defined as sequenceWithIn the incidence coefficient at k moment;

2)Two reference sequences formed by Sobel operators are:

,

3)Assuming that the size of each component image is, to some pixel in each component mapPoint, utilizes picture ElementEight neighborhood component value can form pixelComparative sequences are:

, wherein,;WithRepresent q-th of comparative sequences, wherein

4)Two grey relational grades are calculated, i.e., WithIf,, Suitable threshold value T can be selected, when, pixels illustrated pointAt least there is side in one direction Edge characteristic, then can beIt is determined as marginal point, otherwiseIt is not just marginal point;

S3:The colored gray relation grades image of generation and image merge:

Method as described in step S2, is respectively acting on two components H, S in HSV space, obtains two edge detection graphs PictureWith, then by or computing merge mode obtain color road image border.

The beneficial effects of the invention are as follows:

Approach for road detection of the present invention is independent of road shape and it is assumed that directly carries out side to coloured image using Sobel operators Edge detects that not only real-time is good, and edge detection results are clear, moreover it is possible to effectively reduce the noise of image.

Embodiment

The present invention is further elaborated for specific examples below, but not as a limitation of the invention.

A kind of complete non-structural road edge detection method, comprises the following steps:

S1:Image preprocessing:The video image collected is subjected to image preprocessing;

S2:Determine reference sequences and comparative sequences:

1)If reference sequences, comparative sequences, Wherein N represents the calculation formula of the number, then incidence coefficient of component in reference sequences and sequence to be compared:

,

Wherein,Referred to as resolution ratio, value 0.205;Represent in k-th of moment comparative sequencesWithBetween Relative value, is defined as sequenceWithIn the incidence coefficient at k moment;

2)Two reference sequences formed by Sobel operators are:

,

3)Assuming that the size of each component image is, to some pixel in each component mapPoint, utilizes picture ElementEight neighborhood component value can form pixelComparative sequences are:

, wherein,;WithRepresent q-th of comparative sequences, wherein

4)Two grey relational grades are calculated, i.e., WithIf,, Suitable threshold value T can be selected, when, pixels illustrated pointAt least there is side in one direction Edge characteristic, then can beIt is determined as marginal point, otherwiseIt is not just marginal point;

S3:The colored gray relation grades image of generation and image merge:

Method as described in step S2, is respectively acting on two components H, S in HSV space, obtains two edge detection graphs PictureWith, then by or computing merge mode obtain color road image border.

Claims (1)

1. a kind of complete non-structural road edge detection method, it is characterised in that comprise the following steps:
S1:Image preprocessing:The video image collected is subjected to image preprocessing;
S2:Determine reference sequences and comparative sequences:
1)If reference sequences, comparative sequences, Wherein N represents the calculation formula of the number, then incidence coefficient of component in reference sequences and sequence to be compared:
,
Wherein,Referred to as resolution ratio, value 0.205;Represent in k-th of moment comparative sequencesWithBetween Relative value, is defined as sequenceWithIn the incidence coefficient at k moment;
2)Two reference sequences formed by Sobel operators are:
,
3)Assuming that the size of each component image is, to some pixel in each component mapPoint, utilizes pixelEight neighborhood component value can form pixelComparative sequences are:
, wherein,;WithRepresent q-th of comparative sequences, wherein
4)Two grey relational grades are calculated, i.e., WithIf,, Suitable threshold value T can be selected, when, pixels illustrated pointAt least there is side in one direction Edge characteristic, then can beIt is determined as marginal point, otherwiseIt is not just marginal point;
S3:The colored gray relation grades image of generation and image merge:
Method as described in step S2, is respectively acting on two components H, S in HSV space, obtains two edge detection graphs PictureWith, then by or computing merge mode obtain color road image border.
CN201710327868.6A 2017-05-11 2017-05-11 A kind of complete non-structural road edge detection method CN107122756A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710327868.6A CN107122756A (en) 2017-05-11 2017-05-11 A kind of complete non-structural road edge detection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710327868.6A CN107122756A (en) 2017-05-11 2017-05-11 A kind of complete non-structural road edge detection method

Publications (1)

Publication Number Publication Date
CN107122756A true CN107122756A (en) 2017-09-01

Family

ID=59727924

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710327868.6A CN107122756A (en) 2017-05-11 2017-05-11 A kind of complete non-structural road edge detection method

Country Status (1)

Country Link
CN (1) CN107122756A (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103914688A (en) * 2014-03-27 2014-07-09 北京科技大学 Urban road obstacle recognition system
CN104392212A (en) * 2014-11-14 2015-03-04 北京工业大学 Method for detecting road information and identifying forward vehicles based on vision

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103914688A (en) * 2014-03-27 2014-07-09 北京科技大学 Urban road obstacle recognition system
CN104392212A (en) * 2014-11-14 2015-03-04 北京工业大学 Method for detecting road information and identifying forward vehicles based on vision

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
兰丽 等: ""一种基于灰关联和Sobel算子的完全非结构化道路边缘检测方法"", 《湖南工程学院学报》 *

Similar Documents

Publication Publication Date Title
CN105260699B (en) A kind of processing method and processing device of lane line data
Alvarez et al. Combining priors, appearance, and context for road detection
Zhou et al. Efficient road detection and tracking for unmanned aerial vehicle
Hillel et al. Recent progress in road and lane detection: a survey
Neubert et al. Appearance change prediction for long-term navigation across seasons
Son et al. Real-time illumination invariant lane detection for lane departure warning system
US9435885B2 (en) Road-terrain detection method and system for driver assistance systems
CN103020920B (en) Method for enhancing low-illumination images
Alonso et al. Combination of feature extraction methods for SVM pedestrian detection
Gong et al. The recognition and tracking of traffic lights based on color segmentation and camshift for intelligent vehicles
Yenikaya et al. Keeping the vehicle on the road: A survey on on-road lane detection systems
Kurihata et al. Rainy weather recognition from in-vehicle camera images for driver assistance
Wu et al. Lane-mark extraction for automobiles under complex conditions
Broggi Robust real-time lane and road detection in critical shadow conditions
Milford et al. SeqSLAM: Visual route-based navigation for sunny summer days and stormy winter nights
CN103426182B (en) The electronic image stabilization method of view-based access control model attention mechanism
Sotelo et al. A color vision-based lane tracking system for autonomous driving on unmarked roads
Kühnl et al. Spatial ray features for real-time ego-lane extraction
CN104992145B (en) A kind of square samples track tracking detection method
CN104318258B (en) Time domain fuzzy and kalman filter-based lane detection method
Tan et al. Color model-based real-time learning for road following
Shen et al. A robust video based traffic light detection algorithm for intelligent vehicles
CN101840507B (en) Target tracking method based on character feature invariant and graph theory clustering
CN106599832A (en) Method for detecting and recognizing various types of obstacles based on convolution neural network
CN102314602B (en) Shadow removal in image captured by vehicle-based camera using optimized oriented linear axis

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20170901

WW01 Invention patent application withdrawn after publication