CN106447651A - Traffic sign detection method based on orthogonal Gauss-Hermite moment - Google Patents
Traffic sign detection method based on orthogonal Gauss-Hermite moment Download PDFInfo
- Publication number
- CN106447651A CN106447651A CN201610805781.0A CN201610805781A CN106447651A CN 106447651 A CN106447651 A CN 106447651A CN 201610805781 A CN201610805781 A CN 201610805781A CN 106447651 A CN106447651 A CN 106447651A
- Authority
- CN
- China
- Prior art keywords
- image
- moment
- detection method
- pixel
- order
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 20
- 238000001914 filtration Methods 0.000 claims abstract description 11
- 238000002474 experimental method Methods 0.000 claims abstract description 6
- 241000566145 Otus Species 0.000 claims abstract description 4
- 230000000877 morphologic effect Effects 0.000 claims abstract description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 238000004040 coloring Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 2
- 230000000694 effects Effects 0.000 abstract description 2
- 230000011218 segmentation Effects 0.000 description 4
- 238000009499 grossing Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000003709 image segmentation Methods 0.000 description 2
- 230000006698 induction Effects 0.000 description 2
- 241001463014 Chazara briseis Species 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- UREBDLICKHMUKA-CXSFZGCWSA-N dexamethasone Chemical compound C1CC2=CC(=O)C=C[C@]2(C)[C@]2(F)[C@@H]1[C@@H]1C[C@@H](C)[C@@](C(=O)CO)(O)[C@@]1(C)C[C@@H]2O UREBDLICKHMUKA-CXSFZGCWSA-N 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010230 functional analysis Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T2207/10024—Color 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
- G06T2207/20024—Filtering details
-
- 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
- G06T2207/20036—Morphological image processing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a traffic sign detection method based on an orthogonal Gauss-Hermite moment. The method comprises the following steps of: obtaining a first image of an area of interest; converting an obtained colorful image into a gray level, wherein the method directly calls a library function; independently carrying out calculation along an X direction and a Y direction, and subsequently, calculating the moment in a vertical direction, wherein in specific implementation, a first-order moment is replaced with the first-order difference of a filtered image, a third-order moment is replaced with the linear group of the first-order difference and the third-order difference of the filtered image, and an experiment only uses the third-order moment; applying an Otus method to segment the image; and citing morphological filtering to remove noise. By use of the method, an image in a traffic scene is selected to do an experiment, and an experiment result indicates that the traffic sign detection method is effective. Meanwhile, the traffic sign detection method is compared with other methods, the difference of the effects of first-order moment detection is small, the third-order moment can be further used, and the advantages of the traffic sign detection method are embodied.
Description
Technical field
The present invention relates to a kind of traffic mark detection method based on orthogonal Gauss-Hermite square.
Background technology
Progress with urbanization and the popularization of automobile, vehicles number rolls up, congested in traffic aggravation, traffic thing
Therefore take place frequently, the safety of highway communication and conevying efficiency problem become to become increasingly conspicuous.And the driving auxiliary based on computer vision
System is to solve the problems, such as one of important measures of traffic safety and conevying efficiency so that it gradually obtains in intelligent transportation system
Application.Its research is substantially carried out at road Identification, collision recognition, Traffic Sign Recognition etc. three aspect.In road Identification, collision
Identification aspect research is relatively early, also obtains many preferably results, but research is less in terms of pavement marking identification.Due to handing over
Many important transport information, the such as change of driving road ahead situation, pavement, rate limitation, driving are comprised in logical mark
The information such as is pointed in direction, provides these information to be conducive to driver's reaction in good time to driver it is ensured that driving safety in good time, it is to avoid
Traffic accident occurs, and has great importance.
Content of the invention
The invention aims to overcoming deficiency of the prior art, thus providing one kind to be based on orthogonal Gauss-
The traffic mark detection method of Hermit square.
A kind of traffic mark detection method based on orthogonal Gauss-Hermit square, comprises the following steps:
Step one:Obtain image I (x, y) in interest region;
Step 2:The coloured image of acquisition is converted into gray scale (256 grades) image f (x, y).
If 3 colouring components that (x, y) puts are (R, G, B), then
F (x, y)=0.3R (x, y)+0.59G (x, y)+0.11B (x, y), this method directly invokes built-in function
I (x, y)=rgb2gray (I);
Step 3:Calculate OGHM with Y-direction respectively in X direction, i.e. the square in first calculated level directionWherein n=2k+1, k=0,1,2 ..., the subsequent vertical direction that calculates
Square, that is,a2i+1It is constant;In our algorithm, σ is taken as 5;Specifically real
In existing, first moment is replaced by we with the first-order difference of filtered image, third moment with filtered image first-order difference with
Linear group of third order difference to replace entirely;Third moment is also only used in our experiment;
Step 4:With Otus method, image is split;If the total pixel of image is N, gray level is L, and gray scale is i's
Pixel is NiIndividual;W (k) and u (k) is made to represent that 0 arrives probability and the average gray that the pixel of gray level k occurs respectively, Set as being divided into M pixel class C againj(there is M-1 thresholding 0≤t1<t2<...<tM-1
≤ L-1), divide the image into M pixel class Cj,Cj∈[tj-1+1,tj], j=1 ..., M, t0=0, tM=L-1. then CjOccur
Probability wj, average gray and variance are respectively:wj=w (tj)-w(tj-1),Then interior class variance is
Inter-class varianceTherefore dividing method is to make inter- object distance minimum and between class distance
Maximum threshold values group { t1,t2,t3,...,tM-1As M threshold values optimal threshold values group;As M=2, that is, it is divided into 2 classes, now
For maximizing σ2=ω0(μ-μ0)2+ω1(μ-μ1)2Obtain image segmentation thresholding T;And with T, binaryzation is implemented with image, that is,
Step 5:Quote morphologic filtering and remove noise, first carry out a computingThen carry out opening operation X
B, here B be taken as a little 8 connection.
Beneficial effect using technique scheme is:
The image that the present invention chooses in a traffic scene is tested, test result indicate that, the present invention is effective.Meanwhile,
Also by it, carried out than state more relatively with other methods (as the method for CANNY), and the effect difference of first moment detection is less;But three ranks
Square can use further, therefore its advantage embodies.
Brief description
Fig. 1 is detection method segmentation figure road original image figure.
Fig. 2 is detection method segmentation figure road result figure.
Fig. 3 is detection method segmentation figure label original image figure.
Fig. 4 is detection method segmentation figure label result figure.
Specific embodiment
A kind of traffic mark detection method based on orthogonal Gauss-Hermit square, comprises the following steps:
Step one:Obtain image I (x, y) in interest region;
Step 2:The coloured image of acquisition is converted into gray scale (256 grades) image f (x, y).
If 3 colouring components that (x, y) puts are (R, G, B), then
F (x, y)=0.3R (x, y)+0.59G (x, y)+0.11B (x, y), this method directly invokes built-in function
I (x, y)=rgb2gray (I);
Step 3:Calculate OGHM with Y-direction respectively in X direction, i.e. the square in first calculated level directionWherein n=2k+1, k=0,1,2 ..., the subsequent vertical direction that calculates
Square, that is,a2i+1It is constant;In our algorithm, σ is taken as 5;Specifically real
In existing, first moment is replaced by we with the first-order difference of filtered image, third moment with filtered image first-order difference with
Linear group of third order difference to replace entirely;Third moment is also only used in our experiment;
Step 4:With Otus method, image is split;If the total pixel of image is N, gray level is L, and gray scale is i's
Pixel is NiIndividual;W (k) and u (k) is made to represent that 0 arrives probability and the average gray that the pixel of gray level k occurs respectively, Set as being divided into M pixel class C againj(there is M-1 thresholding 0≤t1<t2<...<tM-1
≤ L-1), divide the image into M pixel class Cj,Cj∈[tj-1+1,tj], j=1 ..., M, t0=0, tM=L-1. then CjOccur
Probability wj, average gray and variance are respectively:wj=w (tj)-w(tj-1),Then interior class variance is
Inter-class varianceTherefore dividing method is to make inter- object distance minimum and between class distance
Maximum threshold values group { t1,t2,t3,...,tM-1As M threshold values optimal threshold values group;As M=2, that is, it is divided into 2 classes, now
For maximizing σ2=ω0(μ-μ0)2+ω1(μ-μ1)2Obtain image segmentation thresholding T;And with T, binaryzation is implemented with image, that is,
Step 5:Quote morphologic filtering and remove noise, first carry out a computingThen carry out opening operation X B,
Here B is taken as 8 connections a little.
(A):Closed operation is a kind of elimination boundary point, makes the process that border is internally shunk;Can be used to eliminate little and nothing
The object of meaning;Algorithm is the structural element B with NxN, each pixel of scan image, two being covered with it with structural element
Value image does AND-operation, if being all 1, this pixel of result images is 1, otherwise for 0;
(B):Opening operation is that all background dots being contacted with object are merged in this object, makes border to outside expansion
Process.Can be used to fill up the cavity in object.Algorithm is the structural element B with NxN, each pixel knot of scan image
If it is all 0 that constitutive element and the bianry image that it covers do AND-operation, this pixel of result images is 0, otherwise for 1.
Orthogonal Gauss-Hermite square and fast algorithm
Square, such as geometric moment, orthogonal moment are widely used for pattern-recognition, image procossing, computer vision, many resolutions point
Analysis etc..
From the viewpoint of functional analysis, one can consider that geometric moment is the space that the base that signal forms in monomial generates
In projection, when we project a signal in a space when, we always wish that this space is orthogonal, its objective is
In order to reduce amount of calculation, and the reconstruct of signal can be conducive to.Orthogonal Gauss-Hermite square meets outside orthogonal requirement,
It smooths very much, and therefore it is to insensitive for noise.This is especially advantageous for effectively detecting moving target in environment made an uproar by band.Under
Face we will provide its definition, property and fast algorithm.
A) geometric moment
One-dimensional n rank geometric moment, MnX () is defined as follows:
Wherein [- w, w] is the interval of a 2w+1, and x is interval midpoint, and s (x) is exactly our signals to be analyzed.
Two-dimensional case is similar with one-dimensional, is defined as follows:
(x, y) is window (2w1+1)×(2w2+ 1) center, s (x, y) 2D is 2D signal, for example, piece image.
Geometric moment is the base 1, x that signal s (x) forms in monomial, x2,…,xN... the projection in the space of generation.
B) orthogonal Legendre moments
Legendre polynomial function can be used as one group of orthogonal basis .Legendre polynomial function and is defined as follows:
Here n is the exponent number of Legendre. according to this multinomial, we can define a yardstick Legendre multinomial
Function is:
Wherein n is the exponent number of yardstick Legendre polynomial function, and w>0.
C) Hermite square
The another one man cluster of orthogonal polynomial is Hermite multinomial. an orthogonal Hermite polynomial is:
Pn=Hn(t/σ) (5)
Wherein
Orthogonal Hermite n rank square MnX () is defined as:
Wherein s (x) is given signal.
Two-dimentional (p, q) rank Hermite square is defined as:
Wherein I (x, y) is image, Hp,q(u/ σ, v/ σ)=Hp(u/σ)Hq(v/σ).
It is true that two-dimentional (p, q) rank Hermite square equally can be melted into one-dimensional cascade to calculate.
D) orthogonal Gauss-Hermite square (OGHM)
The multinomial function base of the individual event function base of geometric moment, Hermite and Legendre all show one at the edge of window
Individual big jump (discontinuous). in order to preferably portray the local feature of signal, especially for noise signal, we are from smooth
Orthogonal moment .Gauss function is widely adopted as smoothing kernel. and therefore, orthogonal Hermite square is therefore suggested.
A given Gaussian function g (t, σ), the orthogonal Gauss-Hermiten rank square of signal s (x), Mn(s, S (x)), fixed
Justice is:
Wherein Bn(t)=g (t, σ) Pn(t),σ is that Gauss standard is poor, Pn(t)
It is Hermite polynomial function, Pn=Hn(t/ σ),
In order to calculate orthogonal Gauss-Hermite n rank square, we can use fast algorithm:
M0(x,s(m)(x))=g (x, σ) * s(m)(x)(9)
And
Mn(x,s(m)(x))=2 (n-1) Mn-2(x,s(m)(x))+2σMn-1(x,s(m+1)(x)) to n >=2 (11)
Wherein s(m)(x)=dm/dxms(x),s(0)X ()=s (x), " * " represents convolution. special circumstances:
M0(x, s (x))=g (x, σ) * s (x),Two-dimensional quadrature Gauss-Hermite rank square
(OGHM) it is defined as:
Wherein
Hp,q(u/ σ, v/ σ)=Hp(u/σ)Hq(v/σ),
I (x, y) is piece image.
In order to calculate Mp,q, we can be realized with one-dimensional series connection:
The property of 2.OGHM
In order to apply n rank orthogonal Gauss-Hermite square (OGHM) in rim detection, we demonstrate that their several property
Matter.
From a Gaussian wave filter, Wo Menyou:
For n rank orthogonal Gauss-Hermite square (OGHM), we can prove that following property:
Property 1
To given Gaussian function g (t, σ) and image I (x, y), have:
Here aiOnly relevant with σ.
Prove:We use mathematical induction.
To n=0, M0(x, I (x, y))=g (x, σ) * I (x, y) is it is clear that give a0=σ0=1.
To n=1, M1(x, I (x, y))=2 σ g(1)(x, σ) * I (x, y), this provides a0=0;a1=2 σ.
If to n≤k (k is integer k >=2), (16) proposition is correct, we demonstrate that proposition is also correct to n=k+1.
Because Mk+1(x, I (x, y))=2kMk-1(x,I(x,y))+2σMk(x,I(1)(x,y)),k≥1,
Therefore, according to formula (17) and formula (18), Wo Menyou:
This property illustrates the smoothing kernel of n rank orthogonal Gauss-Hermite squareIt is high
The linear combination of n order derivative, M are arrived in the 0 of this functionn(x, I (x, y)) is the linear of the derivative of gaussian filtering image on X- domain
Combination.
Property 2
To given Gaussian function g (t, σ) and image I (x, y)
Prove:Equally use mathematical induction.
This property illustrates even order orthogonal Gauss-Hermite square Mn(x, I (x, y)) is the gaussian filtering on X- domain
The linear combination of the even order derivative of image, and odd order orthogonal Gauss-Hermite square Mn(x, I (x, y) are the Gausses on X- domain
The linear combination of the odd order derivative of filtering image.
This is an important properties, and in our work, we will use it for edge inspection according to this property
Survey.
Property 3
Given Gaussian function g (t, σ) and image I (x, y) and its squareSo to maskFor, it is in area
Between n different zero crossing is existed on (- ∞, ∞).
Prove slightly.Fn(t, σ) is also the template of OGHM.
3 use OGHM to detect traffic mark
(1) Cleaning Principle
According to formula (16), (20) and (21), because OGHM is the convolution of the derivative of gaussian filtering on X- domain for the image
Linear combination, or perhaps the linear combination of the different X- differential of filtering image;And differential can be used for rim detection detection.
Because odd order OGHM is the linear combination of image convolution of odd order derivative of gaussian filtering on X- domain, in other words
It is the linear combination of the different odd order differential of filtering image, therefore odd order OGHM can be used for the traffic mark symbol in detection image
Number edge. contain Gaussian smoothing due in OGHM, thus compared with classic differential method, it have stronger anti-noise ability.
Claims (1)
1. a kind of traffic mark detection method based on orthogonal Gauss-Hermit square it is characterised in that:It comprises the following steps:
Step one:Obtain image I (x, y) in interest region;
Step 2:The coloured image of acquisition is converted into gray scale (256 grades) image f (x, y). setting 3 colouring components that (x, y) put is
(R, G, B), then
F (x, y)=0.3R (x, y)+0.59G (x, y)+0.11B (x, y), this method directly invoke built-in function I (x, y)=
rgb2gray(I);
Step 3:Calculate OGHM with Y-direction respectively in X direction, i.e. the square in first calculated level directionWherein n=2k+1, k=0,1,2 ..., the subsequent vertical direction that calculates
Square, that is,a2i+1It is constant;In our algorithm, σ is taken as 5;Specifically real
In existing, first moment is replaced by we with the first-order difference of filtered image, third moment with filtered image first-order difference with
Linear group of third order difference to replace entirely;Third moment is also only used in our experiment;
Step 4:With Otus method, image is split;If the total pixel of image is N, gray level is L, and gray scale is the pixel of i
For NiIndividual;W (k) and u (k) is made to represent that 0 arrives probability and the average gray that the pixel of gray level k occurs respectively, Set as being divided into M pixel class C againj(there is M-1 thresholding 0≤t1<t2<...<tM-1≤ L-1), will scheme
As being divided into M pixel class Cj,Cj∈[tj-1+1,tj], j=1 ..., M, t0=0, tM=L-1. then CjThe probability w occurringj, averagely
Gray scale and variance are respectively:wj=w (tj)-w(tj-1),Then
Interior class variance isInter-class varianceTherefore
Dividing method is to make the threshold values group { t that inter- object distance is minimum and between class distance is maximum1,t2,t3,...,tM-1As M threshold values
Optimal threshold values group;As M=2, that is, it is divided into 2 classes, now for maximizing σ2=ω0(μ-μ0)2+ω1(μ-μ1)2Obtain image to divide
Cut thresholding T;And with T, binaryzation is implemented with image, that is,
Step 5:Quote morphologic filtering and remove noise, first carry out a computingThen carry out opening operation X B, here
B is taken as 8 connections a little.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610805781.0A CN106447651A (en) | 2016-09-07 | 2016-09-07 | Traffic sign detection method based on orthogonal Gauss-Hermite moment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610805781.0A CN106447651A (en) | 2016-09-07 | 2016-09-07 | Traffic sign detection method based on orthogonal Gauss-Hermite moment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106447651A true CN106447651A (en) | 2017-02-22 |
Family
ID=58164780
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610805781.0A Pending CN106447651A (en) | 2016-09-07 | 2016-09-07 | Traffic sign detection method based on orthogonal Gauss-Hermite moment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106447651A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113888721A (en) * | 2021-10-08 | 2022-01-04 | 西北工业大学 | Three-dimensional image feature design method based on integer-fractional order orthogonal polynomial |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101303729A (en) * | 2008-07-01 | 2008-11-12 | 山东大学 | Novel method for detecting fingerprint singularity |
CN103218616A (en) * | 2013-05-05 | 2013-07-24 | 西安电子科技大学 | Image outline characteristic extraction method based on Gauss-Hermite special moment |
CN104123549A (en) * | 2014-07-30 | 2014-10-29 | 中国人民解放军第三军医大学第二附属医院 | Eye positioning method for real-time monitoring of fatigue driving |
-
2016
- 2016-09-07 CN CN201610805781.0A patent/CN106447651A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101303729A (en) * | 2008-07-01 | 2008-11-12 | 山东大学 | Novel method for detecting fingerprint singularity |
CN103218616A (en) * | 2013-05-05 | 2013-07-24 | 西安电子科技大学 | Image outline characteristic extraction method based on Gauss-Hermite special moment |
CN104123549A (en) * | 2014-07-30 | 2014-10-29 | 中国人民解放军第三军医大学第二附属医院 | Eye positioning method for real-time monitoring of fatigue driving |
Non-Patent Citations (4)
Title |
---|
吴有富 等: "Traffic object detections and its action analysis", 《PATTERN RECOGNITION LETTERS》 * |
孔令建 等: "基于正交Gaussian-Hermite矩的边缘检测方法", 《广西大学学报:自然科学版》 * |
强俊 等: "基于正交Gaussian-Hermite矩的运动检测在去隔行中的研究", 《安徽工程大学学报》 * |
谢鹏 等: "基于Otsu阈值彩色图像分割方法研究", 《SCIENCE & TECHNOLOGY INFORMATION》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113888721A (en) * | 2021-10-08 | 2022-01-04 | 西北工业大学 | Three-dimensional image feature design method based on integer-fractional order orthogonal polynomial |
CN113888721B (en) * | 2021-10-08 | 2022-09-16 | 西北工业大学 | Three-dimensional image feature design method based on integer-fractional order orthogonal polynomial |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Bilal et al. | Real-time lane detection and tracking for advanced driver assistance systems | |
Andrade et al. | A novel strategy for road lane detection and tracking based on a vehicle’s forward monocular camera | |
Kaur et al. | Lane detection techniques: A review | |
Nieto et al. | Road environment modeling using robust perspective analysis and recursive Bayesian segmentation | |
Huang et al. | Vehicle detection and inter-vehicle distance estimation using single-lens video camera on urban/suburb roads | |
AU2020102039A4 (en) | A high-precision multi-targets visual detection method in automatic driving scene | |
Ming et al. | Vehicle detection using tail light segmentation | |
CN103927526A (en) | Vehicle detecting method based on Gauss difference multi-scale edge fusion | |
CN105260712A (en) | Method and system for detecting pedestrian in front of vehicle | |
CN107392139A (en) | A kind of method for detecting lane lines and terminal device based on Hough transformation | |
CN202134079U (en) | Unmanned vehicle lane marker line identification and alarm device | |
CN105426868A (en) | Lane detection method based on adaptive region of interest | |
Li et al. | A lane marking detection and tracking algorithm based on sub-regions | |
You et al. | Lane detection algorithm for night-time digital image based on distribution feature of boundary pixels | |
CN114973199A (en) | Rail transit train obstacle detection method based on convolutional neural network | |
CN113297939B (en) | Obstacle detection method, obstacle detection system, terminal device and storage medium | |
Rahaman et al. | Lane detection for autonomous vehicle management: PHT approach | |
CN106447651A (en) | Traffic sign detection method based on orthogonal Gauss-Hermite moment | |
CN114724113B (en) | Road sign recognition method, automatic driving method, device and equipment | |
CN107452230B (en) | Obstacle detection method and device, terminal equipment and storage medium | |
CN116311113A (en) | Driving environment sensing method based on vehicle-mounted monocular camera | |
CN105416290A (en) | Method and device for detecting distances between vehicles | |
Yuan et al. | A robust vanishing point estimation method for lane detection | |
Zakaria et al. | Fully convolutional neural network for Malaysian road lane detection | |
Mir et al. | Automated speed limit identification for efficient driving system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170222 |
|
RJ01 | Rejection of invention patent application after publication |