CN110210451B - Zebra crossing detection method - Google Patents
Zebra crossing detection method Download PDFInfo
- Publication number
- CN110210451B CN110210451B CN201910511680.6A CN201910511680A CN110210451B CN 110210451 B CN110210451 B CN 110210451B CN 201910511680 A CN201910511680 A CN 201910511680A CN 110210451 B CN110210451 B CN 110210451B
- Authority
- CN
- China
- Prior art keywords
- zebra crossing
- image
- point
- region
- points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4053—Super resolution, i.e. output image resolution higher than sensor resolution
Abstract
The invention relates to a zebra crossing detection method, and belongs to the technical field of road environment perception in unmanned driving. S1: detecting the zebra crossing in real time by adopting a camera fixedly installed at the top of the vehicle to obtain an image; s2: preprocessing the image; s3: carrying out back projection transformation on the zebra crossing candidate region, converting the image into a top view, and carrying out feature extraction and detection on the zebra crossing on the basis of the top view; s4: calculating the distance between peak values, and reserving the peak values with the same distance; s5: determining a zebra crossing candidate area; s6: after the zebra crossing candidate area is determined, extracting the zebra crossing feature points; s7: and determining the zebra crossing area. The method can provide the zebra crossing detection of the complex road section for the intelligent vehicle by detecting the histogram peak value and the LSD straight line and combining the black and white mutation point characteristics of the zebra crossing, thereby having wide market prospect and application value.
Description
Technical Field
The invention belongs to the technical field of road environment perception in unmanned driving, and relates to a zebra crossing detection method for an intelligent vehicle.
Background
With the development of society and economic technology, vehicles on traffic roads are more and more, the road environment is more and more complex, the traffic accidents caused by the vehicles are frequent, intelligent traffic is carried out at the same time in order to reduce the occurrence of the situations, and in the intelligent traffic, an intelligent vehicle is a key part of the intelligent traffic, and the occurrence of the intelligent vehicle helps drivers to drive, so that the occurrence frequency of the traffic accidents is reduced. The road zebra crossing is an important safety sign, and for a vehicle driver, the zebra crossing means that the vehicle driver needs to drive carefully and safely pass at a slow speed; for pedestrians, the zebra crossing is a protective strip which safely passes through the road. Thus, zebra crossing detection is an indispensable part of intelligent traffic.
The existing patent about zebra crossing detection is mainly based on a Hough transform and vanishing point detection method, and mainly utilizes visual information and the like to obtain zebra crossing features, and then trains a classifier to detect the zebra crossing features. The method can realize complete visual information at the zebra crossing and obvious contrast between the target and the background, but at a large complex crossing, the zebra crossing cannot be accurately detected due to the interference on the zebra crossing detection caused by the complex road environment.
Based on the problems, the invention provides a zebra crossing detection method with robustness for detection in different time periods.
Disclosure of Invention
In view of this, the present invention provides a zebra crossing detection method for an intelligent vehicle, which provides more reliable and accurate data for unmanned driving of the vehicle by detecting and positioning the zebra crossing, reduces traffic accidents on the zebra crossing, and improves safety of the unmanned vehicle when the unmanned vehicle travels.
In order to achieve the purpose, the invention provides the following technical scheme:
a zebra crossing detection method specifically comprises the following steps:
s1: the array camera is fixedly installed on the roof of the vehicle, the super-resolution module is added, the low-resolution image is processed to obtain a high-resolution image, so that a higher-quality output image is obtained, and the detection of the intelligent vehicle on the zebra crossing is facilitated.
S2: and preprocessing the acquired image, including determining an image region of interest, processing image gray scale, Gaussian smoothing, self-adaptive binarization, edge detection, image segmentation, morphological closed operation and the like.
S3: and carrying out back projection transformation on the image after the region of interest is determined, converting the image into a top view, and extracting and detecting the features of the zebra crossing on the basis of the top view.
S4: and carrying out histogram peak value statistics on the image after the inverse perspective transformation, and reserving the areas with the same peak value spacing.
S5: and (3) solving all straight lines in the image after the inverse perspective transformation by adopting an LSD (linear space-time decomposition) straight line detection algorithm, counting the areas with dense straight lines, performing AND operation on the areas with dense straight lines obtained by counting and the areas with the same histogram peak value space, and determining the result as the zebra crossing candidate area.
S6: the method comprises the steps of extracting characteristic points of a zebra crossing candidate area, wherein the zebra crossing of an image under inverse perspective transformation is a plurality of parallel black-and-white stripes with almost consistent widths, taking one row at every 10 pixel points for statistics, counting black-and-white pixel mutation points of each row of zebra crossing, and defining a point from 0 mutation to 255 mutation as an up point and a point from 255 mutation to 0 mutation as a down point. And then calculating the distance WIDTH between the adjacent up and down catastrophe points, and if the WIDTH is less than a certain threshold WIDTH, considering the catastrophe points as zebra crossing characteristic points, wherein the threshold WIDTH is the WIDTH between black and white stripes of the zebra crossing of the image.
S7: and determining the zebra crossing area. After the zebra crossing feature points are obtained, the first line feature point top is obtainediOrdinate y ofiI is more than or equal to 1 and less than or equal to n, and the last row of feature points bottomjOrdinate y ofjJ is more than or equal to 1 and less than or equal to n. If n is more than or equal to 3 pairs of yiAnd yjIs within a set threshold, the region is retained and is called keyarea. Traversing the feature points of each effective row in the keyarea, calculating the shortest distance s from each feature point of other effective rows to the feature point of the effective row in the x direction, setting a threshold value, and if the shortest distance s between the feature points in the x direction of the rows is in the range of the threshold value, determining that the region is a zebra crossing region.
The invention has the beneficial effects that:
(1) the method fully utilizes the uniform distribution of bright and dark stripes of the urban zebra crossing, has obvious pixel mutation characteristics, extracts the characteristic points of the zebra crossing, determines the candidate region of the zebra crossing by utilizing a histogram peak value statistical method and an LSD (least squares) linear detection algorithm, and can accurately position the zebra crossing. And finally, determining the final zebra crossing area according to the position relation among the zebra crossing feature points.
(2) The invention adopts various image processing methods to realize the detection of the zebra crossing characteristic points and fully utilizes the obvious characteristics of the zebra crossing. Aiming at images acquired under various illumination conditions which may face, firstly, a super-resolution module is adopted to process images output by a camera, and secondly, an image segmentation method is adopted to effectively segment zebra stripes and a background. And then, a method of expanding firstly and corroding secondly is adopted, so that the problems of noise in the image and holes possibly existing in the zebra crossing are solved, and the interference of the pavement lane lines on the zebra crossing detection is eliminated.
(3) The invention provides more reliable and accurate data for the unmanned driving of the vehicle by detecting and positioning the zebra crossings, reduces traffic accidents on the zebra crossings, and improves the traveling safety of the unmanned vehicle, thereby having wide application value and market prospect.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a flow chart of the zebra crossing detection method of the present invention;
FIG. 2 is a schematic diagram of a zebra crossing break point;
fig. 3 is a schematic diagram of a road adopting the zebra crossing detection method of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Referring to fig. 1 to 3, fig. 1 is a zebra crossing detection method for an intelligent vehicle, which includes the following steps:
step 1: firstly, an array camera is fixedly installed on the top of the vehicle, a super-resolution module is installed, and a low-resolution image is processed to obtain a high-resolution image so as to obtain a higher-quality output image, thereby being beneficial to the detection of the intelligent vehicle on the zebra crossing.
Step 2: preprocessing the acquired image, wherein the preprocessing of the image comprises the following steps:
(1) and determining an image interested region and cutting off the sky part of the image.
(2) And graying the image, and converting the RGB image into a grayscale image by adopting a formula Gray of 0.3R +0.59G + 0.11B.
(3) And smoothing the acquired data image, and operating by adopting a Gaussian filter.
(4) And (4) carrying out binarization processing on the image by adopting a rapid self-adaptive binarization method.
(5) And carrying out edge detection on the image, and extracting gradient features in the x direction and the y direction by adopting a Laplacian operator.
(6) And carrying out watershed segmentation on the image, and separating the zebra crossing from the background in the picture.
(7) And (3) performing expansion and corrosion operations on the image, eliminating small cavities, smoothing the outline of the object, and removing the interference of the road lane lines on the zebra stripes.
And step 3: the inverse perspective transformation of the acquired image comprises:
first, the inverse perspective transformation involves subjecting the image to a series of rotational and translational transformations, the overall transformation formula being as follows:
wherein R, T is the rotation and translation matrix of the data from the world coordinate system to the camera coordinate system, f is the focal length of the camera,
and
indicates the number of pixels contained per unit length in the x-direction and the y-direction; u. of0、v0The number of pixels in the horizontal and vertical directions representing the difference between the central pixel coordinate of the image and the pixel coordinate of the origin of the image, respectively, u and v represent the coordinates of the image coordinate system in units of pixels, and ZcIs the Z-axis coordinate value, X, in camera coordinatesW、YW、ZWIs the X, Y and Z values in the world coordinate system.
And through inverse perspective transformation, the zebra stripes are drawn into a plurality of parallel black and white stripes, so that subsequent characteristic points can be conveniently extracted.
And 4, step 4: the statistical histogram peak specifically includes: and carrying out histogram statistics on the image after the inverse perspective, adopting the histogram statistics to count the pixel value of the region, and searching a histogram peak value, wherein the histogram peak value is a white stripe region of the zebra crossing. The distance between the peaks is calculated, all peaks at the same distance are retained, and the region between the first peak and the last peak is retained.
And 5: determining a zebra crossing candidate area, specifically comprising:
(1) and carrying out LSD line detection on the image after inverse perspective. The LSD is a straight line detection algorithm, the purpose of the LSD is to detect a local straight outline in an image, and the method can obtain a straight line detection result with higher precision in a shorter time. After all the straight line segments in the region are counted, the angle theta between each line segment and the x direction of the image is calculated, and all the line segments with the same theta are counted. Starting from the origin, traversing all line segments containing the same theta, counting the times that the vertical distance q between the next line segment and the previous line segment is smaller than a distance threshold p, if the times n is larger than m, extracting the region between the first line segment and the last line segment, wherein m is the threshold of the number of line segments in the defined region.
(2) And (5) performing AND operation on the area of the histogram peak value statistics and the area extracted by the straight line detection, and determining the result as a zebra crossing candidate area.
And 6: extracting the characteristic points of the zebra crossing, and obtaining the relevant parameters of the characteristic points for the zebra crossing detection, wherein the method comprises the following steps:
(1) and processing the zebra crossing candidate area. The image zebra crossing after reverse perspective is a plurality of parallel black and white stripes with almost the same width, the specific characteristic point extraction method is to take one row at every 10 pixel points for statistics, and count black and white pixel mutation points of each row of zebra crossing, wherein the point of the pixel value from 0 mutation to 255 mutation is defined as an up mutation point, and the point of the pixel value from 255 mutation to 0 mutation is defined as a down mutation point as shown in fig. 2. And traversing each pixel point of each row, if the pixel value of each pixel point and the pixel values of the next five continuous points are 0 or 255, determining that the pixel points are not the catastrophe points, and otherwise, if the current pixel point is different from the next five continuous points, determining that the current pixel point is the catastrophe point.
(2) Calculating the distance WIDTH between the up catastrophe point and the down catastrophe point of each line, if the WIDTH is smaller than a certain threshold WIDTH (the threshold WIDTH is the WIDTH between black and white stripes of the zebra crossing of the image), regarding the catastrophe point as a zebra crossing candidate characteristic point, defining the effective line of the behavior, and if the number of the effective lines is more than or equal to 3, defining the point on the effective line as the characteristic point.
And 7: determining a zebra crossing area, specifically comprising:
(1) after the zebra crossing feature points are obtained, the first line feature point top is obtainediOrdinate y ofiI is more than or equal to 1 and less than or equal to n, and the last row of feature points bottomjOrdinate y ofjJ is more than or equal to 1 and less than or equal to n. If n is more than or equal to 3 pairs of yiAnd yjIs within a set threshold, the region is retained and is called keyarea.
(2) And traversing the feature points of each effective row in the keyarea region, and calculating the shortest distance s from each feature point of other effective rows to the feature point of the effective row in the x direction.
(3) And setting a threshold value t, if the shortest distance s in the x direction among a plurality of lines of feature points is less than t and the number of the lines is more than or equal to 3, determining that the region is a zebra crossing region, and if not, traversing the next feature point of the first effective line.
The invention adopts various image processing methods to realize the detection of the zebra crossing characteristic points and fully utilizes the obvious characteristics of the zebra crossing. Aiming at images acquired under various illumination conditions which may be faced, the method firstly adopts a super-resolution module to process the image output by a camera, and secondly adopts an image segmentation method to effectively segment the zebra crossing and the background. The method adopts a morphological closed operation method, firstly expands and then corrodes, is used for solving the problems of noise points in the image and possible holes of the zebra crossing, and eliminates the interference of the lane lines on the road surface to the zebra crossing detection. Finally, the invention also adopts a method for combining the histogram peak value and LSD line detection to determine the candidate area of the zebra crossing. The method provides more reliable and accurate data for unmanned driving of the vehicle by detecting and positioning the zebra crossing, reduces traffic accidents on the zebra crossing, and improves the traveling safety of the unmanned vehicle, thereby having wide application value and market prospect.
Finally, although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that various changes and modifications may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (3)
1. A zebra crossing detection method is characterized by comprising the following steps:
s1: the array camera is fixedly arranged on the roof of the vehicle, the super-resolution module is arranged, and the low-resolution image is processed to recover a high-resolution image so as to obtain a higher-quality output image;
s2: preprocessing the acquired image, extracting an image region of interest, and performing image segmentation and expansion corrosion;
s3: carrying out back projection transformation on the zebra crossing candidate area, converting the image into a top view, and carrying out feature extraction and detection on the zebra crossing on the basis of the top view;
s4: counting the peak value of the image histogram in the region of interest, calculating the distance between peak values, and reserving the peak values with the same distance;
s5: adopting an LSD (line Segment detector) line detection segmentation algorithm to obtain all lines in the region of interest, counting the regions with dense lines, performing AND operation with the histogram peak region, and determining a zebra crossing candidate region; the method specifically comprises the following steps:
s51: detecting all straight lines in the region of interest by adopting an LSD (line Segment detector) algorithm, counting all straight lines with an included angle theta with the x direction of the image, judging the distance q between the adjacent straight lines, and keeping the straight lines with the distance q within a threshold value p;
s52: performing AND operation on the regions in the first straight line and the last straight line which are counted and the region determined by the peak value of the statistical histogram, and determining a zebra crossing candidate region;
s6: after the zebra crossing candidate area is determined, extracting the zebra crossing feature points; the method specifically comprises the following steps:
s61: processing the zebra crossing candidate area: taking one row from every 10 pixel points for statistics, counting black and white pixel mutation points of each row of zebra stripes, defining the point of a pixel value from 0 mutation to 255 as an up mutation point, and the point from 255 mutation to 0 as a down mutation point, traversing each pixel point of each row, if the pixel value of the pixel point is the same as the pixel values of the next five continuous points, determining that the pixel points are not mutation points, otherwise, if the current pixel point is different from the next five continuous points, determining that the point is a mutation point;
s62: calculating the distance WIDTH between the up catastrophe point and the down catastrophe point adjacent to each row, if the WIDTH is less than a certain threshold WIDTH, regarding the catastrophe point as a zebra crossing candidate characteristic point, wherein the threshold WIDTH is the WIDTH between black and white stripes of the zebra crossing in the image, and defining an effective row of the behavior, and if the number of the effective rows is more than or equal to 3, defining a point on the effective row as a characteristic point; wherein the threshold width is the width between black and white stripes of the zebra crossing in the image;
s7: after the zebra crossing characteristic points are obtained, determining zebra crossing areas according to the characteristic that the black and white stripes of the zebra crossings are uniformly distributed and the relation of the positions of the characteristic points; the method specifically comprises the following steps:
s71: after the zebra crossing feature points are obtained, finding out a first row feature point topiOrdinate y ofiI is more than or equal to 1 and less than or equal to n, and the last row of feature points bottomjOrdinate y ofjJ is more than or equal to 1 and less than or equal to n; if n is more than or equal to 3 pairs of yiAnd yjIf the distance is within the range of the set threshold value, the region is reserved and is called keyarea;
s72: traversing the feature points of each row in the keyarea area, and calculating the shortest distance s from each feature point of other rows to the feature point of the current row in the x direction;
s73: and setting a threshold value t, if the shortest distance s in the x direction among the characteristic points of multiple lines is less than t and the number of the lines is more than or equal to 3, determining that the region is a zebra crossing region, and if not, traversing the next characteristic point of the first effective line.
2. The zebra crossing detection method of claim 1, wherein the preprocessing of the image in step S2 comprises the following steps:
s21: determining an image interested area, and cutting off the sky part of the image;
s22: graying the image;
s23: smoothing the acquired data image;
s24: carrying out binarization processing on the image;
s25: carrying out edge detection on the image, and extracting gradient features in the x and y directions;
s26: carrying out marker-based watershed segmentation on the image, and separating the zebra crossing from the background in the image;
s27: the image is firstly expanded and then corroded, small cavities are eliminated, the outline of an object is smoothed, and the interference of a road lane line on the zebra crossing is eliminated.
3. The method according to claim 1, wherein in step S4, the statistical image histogram peak specifically includes: performing histogram statistics on the image after inverse perspective, and adopting the histogram to count the pixel value of the image area, and searching a histogram peak value, wherein the histogram peak value is a white stripe area of the zebra crossing; the distance between the peaks is calculated, all peaks at the same distance are retained, and the region between the first peak and the last peak is retained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910511680.6A CN110210451B (en) | 2019-06-13 | 2019-06-13 | Zebra crossing detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910511680.6A CN110210451B (en) | 2019-06-13 | 2019-06-13 | Zebra crossing detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110210451A CN110210451A (en) | 2019-09-06 |
| CN110210451B true CN110210451B (en) | 2022-07-08 |
Family
ID=67792666
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910511680.6A Active CN110210451B (en) | 2019-06-13 | 2019-06-13 | Zebra crossing detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110210451B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807771B (en) * | 2019-10-31 | 2022-03-22 | 长安大学 | Defect detection method for road deceleration strip |
| CN111445510A (en) * | 2020-03-24 | 2020-07-24 | 杭州东信北邮信息技术有限公司 | Method for detecting straight line in image |
| CN111507287B (en) * | 2020-04-22 | 2023-10-24 | 山东省国土测绘院 | Method and system for extracting road zebra crossing corner points in aerial image |
| CN112069282B (en) * | 2020-09-08 | 2021-07-02 | 湖北亿咖通科技有限公司 | Road zebra crossing generation method, electronic device and storage medium |
| CN112215123B (en) * | 2020-10-09 | 2022-10-25 | 腾讯科技（深圳）有限公司 | Target detection method, device and storage medium |
| CN112800989A (en) * | 2021-02-02 | 2021-05-14 | 上海博泰悦臻电子设备制造有限公司 | Method and device for detecting zebra crossing |
| CN113344953B (en) * | 2021-04-21 | 2023-07-04 | 中国计量大学 | Machine vision tidal bore flow velocity measurement method based on unmanned aerial vehicle |
| CN113435350A (en) * | 2021-06-30 | 2021-09-24 | 浙江大华技术股份有限公司 | Traffic marking detection method, device, equipment and medium |
| CN114724118B (en) * | 2022-06-08 | 2022-09-27 | 苏州魔视智能科技有限公司 | Zebra crossing detection method, device, equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488976A (en) * | 2013-09-17 | 2014-01-01 | 北京联合大学 | Stop mark real-time detection and distance measurement method based on intelligent driving |
| CN103488975A (en) * | 2013-09-17 | 2014-01-01 | 北京联合大学 | Zebra crossing real-time detection method based in intelligent driving |
| CN103593671A (en) * | 2013-11-25 | 2014-02-19 | 中国航天科工集团第三研究院第八三五七研究所 | Wide-range lane line vision inspection method based on collaboration of three cameras |
| CN104742912A (en) * | 2013-12-27 | 2015-07-01 | 比亚迪股份有限公司 | Lane deviation detection method and device |
| CN105718870A (en) * | 2016-01-15 | 2016-06-29 | 武汉光庭科技有限公司 | Road marking line extracting method based on forward camera head in automatic driving |
| CN106909916A (en) * | 2017-03-20 | 2017-06-30 | 南京大学 | A kind of method based on cell phone platform quick detection with identification pavement zebra stripes |
| CN108090401A (en) * | 2016-11-23 | 2018-05-29 | 株式会社理光 | Line detecting method and line detection device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2291836B1 (en) * | 2008-05-21 | 2018-09-12 | ADC Automotive Distance Control Systems GmbH | Driver assistance system for preventing a vehicle colliding with pedestrians |
-
2019
- 2019-06-13 CN CN201910511680.6A patent/CN110210451B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488976A (en) * | 2013-09-17 | 2014-01-01 | 北京联合大学 | Stop mark real-time detection and distance measurement method based on intelligent driving |
| CN103488975A (en) * | 2013-09-17 | 2014-01-01 | 北京联合大学 | Zebra crossing real-time detection method based in intelligent driving |
| CN103593671A (en) * | 2013-11-25 | 2014-02-19 | 中国航天科工集团第三研究院第八三五七研究所 | Wide-range lane line vision inspection method based on collaboration of three cameras |
| CN104742912A (en) * | 2013-12-27 | 2015-07-01 | 比亚迪股份有限公司 | Lane deviation detection method and device |
| CN105718870A (en) * | 2016-01-15 | 2016-06-29 | 武汉光庭科技有限公司 | Road marking line extracting method based on forward camera head in automatic driving |
| CN108090401A (en) * | 2016-11-23 | 2018-05-29 | 株式会社理光 | Line detecting method and line detection device |
| CN106909916A (en) * | 2017-03-20 | 2017-06-30 | 南京大学 | A kind of method based on cell phone platform quick detection with identification pavement zebra stripes |
Non-Patent Citations (2)
| Title |
|---|
| ZebraRecognizer:Efficient and Precise Localization of Pedestrian Crossings;Dragan Ahmetovic;《IEEEXplore》;20141208;全文 * |
| 基于图像去雾和图像检测的交通信息提取算法;徐军等;《实验室研究与探索》;20151215(第12期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110210451A (en) | 2019-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110210451B (en) | Zebra crossing detection method | |
| CN105373794B (en) | A kind of licence plate recognition method | |
| CN102708356B (en) | Automatic license plate positioning and recognition method based on complex background | |
| CN101334836B (en) | License plate positioning method incorporating color, size and texture characteristic | |
| CN109101924B (en) | Machine learning-based road traffic sign identification method | |
| Aly | Real time detection of lane markers in urban streets | |
| KR101403876B1 (en) | Method and Apparatus for Vehicle License Plate Recognition | |
| CN105718870A (en) | Road marking line extracting method based on forward camera head in automatic driving | |
| CN105005771A (en) | Method for detecting full line of lane based on optical flow point locus statistics | |
| CN110619279B (en) | Road traffic sign instance segmentation method based on tracking | |
| CN109448000B (en) | Segmentation method of traffic direction sign image | |
| CN105488454A (en) | Monocular vision based front vehicle detection and ranging method | |
| CN104700072A (en) | Lane line historical frame recognition method | |
| CN107895151A (en) | Method for detecting lane lines based on machine vision under a kind of high light conditions | |
| CN106815583B (en) | Method for positioning license plate of vehicle at night based on combination of MSER and SWT | |
| CN104239867A (en) | License plate locating method and system | |
| CN108921076B (en) | Pavement crack disease self-adaptive constant false alarm detection method based on image | |
| CN103324958B (en) | Based on the license plate locating method of sciagraphy and SVM under a kind of complex background | |
| CN109635722B (en) | Automatic identification method for high-resolution remote sensing image intersection | |
| CN103279755A (en) | Vehicle bottom shadow characteristics-based rapid license plate positioning method | |
| CN111695565A (en) | Automobile mark accurate positioning method based on road barrier fuzzy image | |
| CN107918775B (en) | Zebra crossing detection method and system for assisting safe driving of vehicle | |
| CN113378690A (en) | In-road irregular parking identification method based on video data | |
| CN103440785A (en) | Method for rapid lane departure warning | |
| CN110443142B (en) | Deep learning vehicle counting method based on road surface extraction and segmentation |
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 |