CN113223034A - Road edge detection and tracking method - Google Patents
Road edge detection and tracking method Download PDFInfo
- Publication number
- CN113223034A CN113223034A CN202110535212.XA CN202110535212A CN113223034A CN 113223034 A CN113223034 A CN 113223034A CN 202110535212 A CN202110535212 A CN 202110535212A CN 113223034 A CN113223034 A CN 113223034A
- Authority
- CN
- China
- Prior art keywords
- road
- image
- edge detection
- road image
- edge
- 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
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000003708 edge detection Methods 0.000 title claims abstract description 35
- 230000011218 segmentation Effects 0.000 claims abstract description 17
- 230000009466 transformation Effects 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000001914 filtration Methods 0.000 claims description 16
- 230000002146 bilateral effect Effects 0.000 claims description 6
- 238000001069 Raman spectroscopy Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 9
- 238000000605 extraction Methods 0.000 description 6
- 239000011159 matrix material Substances 0.000 description 6
- 238000001514 detection method Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- 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
- G06T2207/10016—Video; Image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a road edge detection and tracking method, which relates to the technical field of road edge detection and comprises the following specific steps: preprocessing the road image to obtain a road image threshold value segmentation map; performing edge detection on the road image threshold segmentation graph, and obtaining a road image edge detection graph by adopting a Canny edge detection operator; identifying a correct road edge straight line from the road image edge detection graph through Hough transformation with direction constraint and a probability voting method; and tracking the correct road edge straight line, and dynamically delimiting an ROI (region of interest) according to the road edge parameter information. The road edge recognition method can effectively filter interference factors, accurately recognize the road edge, and has the advantages of good real-time performance, high accuracy and the like.
Description
Technical Field
The invention relates to the technical field of road edge detection, in particular to a road edge detection and tracking method.
Background
The original road marking work needs manual real-time control, the working efficiency is low, and the spraying quality is influenced by subjective factors easily. With the continuous development of automation technology and sensor technology, intelligent driving systems begin to enter the career of people, and various sensor technologies are adopted to realize automatic road identification and automatic equipment position adjustment. The machine vision technology is introduced into the field of road marking, the marking work efficiency can be improved, and the loss caused by misoperation of workers is avoided.
Road environment is complicated and changeable, interference conditions such as strong light and shadow can occur in the marking process, and the accuracy and the precision of road edge identification are influenced. When the current road identification technology encounters strong light, shadow or sundry interference, the phenomenon of detection error is easy to occur, and the current road identification technology cannot adapt to the complicated and changeable road marking construction environment. Therefore, how to accurately identify the road edge and track the road edge parameters is a problem to be solved urgently for those skilled in the art.
Disclosure of Invention
In view of the above, the invention provides a road edge detection and tracking method, which solves the problems of inaccurate identification and overlarge deviation of the existing road additional detection and tracking technology in some environments with strong light and shadows, and achieves the technical effects of accurately identifying road edges and tracking road edge parameters.
In order to achieve the purpose, the invention adopts the following technical scheme: a road edge detection and tracking method comprises the following specific steps:
preprocessing the road image to obtain a road image threshold value segmentation map;
performing edge detection on the road image threshold segmentation graph, and obtaining a road image edge detection graph by adopting a Canny edge detection operator;
identifying a correct road edge straight line from the road image edge detection graph through Hough transformation with direction constraint and a probability voting method;
and tracking the correct road edge straight line, and dynamically delimiting an ROI (region of interest) according to the road edge parameter information.
Preferably, the preprocessing the road image specifically includes the following steps:
carrying out weighted average graying on the road image to obtain a grayed image;
extracting an ROI (region of interest) region of the grayed image, and filtering out irrelevant regions to obtain a first road image;
carrying out bilateral filtering on the first road image to obtain a second road image;
and segmenting the second road image by adopting an Otsu threshold value, and separating target information and background information to obtain a road image threshold value segmentation map.
By adopting the technical scheme, the method has the following beneficial technical effects: the ROI area extraction is carried out on the gray-scale image, irrelevant areas in the road image are filtered, the subsequent algorithm operation amount is reduced, and the algorithm operation rate can be greatly improved.
Preferably, the edge detection is performed on the road image threshold segmentation graph through a Canny edge detection operator.
Preferably, the slope and the coordinates of the right road edge straight line are tracked through a Kalman filter; the ROI area is dynamically defined according to a Raman filter.
Preferably, the grayscale values of the grayed image are:
Gray(i，y)＝0.299*R(i，y)+0.578G(i，y)+0.114B(i，y)；
wherein R (i, y) represents a value of the road image red component, G (i, y) represents a value of the road image green component, and B (i, y) represents a value of the road image blue component.
By adopting the technical scheme, the method has the following beneficial technical effects: and giving different weights to the three color components of the acquired road image, wherein when the weight of R, G, B is respectively 0.299, 0.578 and 0.114, the obtained gray level image is relatively accordant with human vision.
According to the technical scheme, the invention discloses and provides a road edge detection and tracking method, which has the following beneficial technical effects compared with the prior art:
1. the invention uses the road image preprocessing technology, so that the obtained gray level image is relatively in accordance with the human vision, the arithmetic speed of the algorithm and the efficiency of subsequent image processing and feature extraction are improved, and irrelevant information is reduced.
2. The invention uses the technology of extracting the road edge straight line characteristics, improves the accuracy of extracting the actual road condition and reduces the phenomenon of overlarge deviation.
3. The road edge recognition method can effectively filter interference factors, accurately recognize the road edge, and has the advantages of good real-time performance and high accuracy.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic flow diagram of the process of the present invention;
2(a) -2 (b) are graphs of the road edge straight line feature extraction effect of the invention;
FIG. 3 is a Kalman filtering tracking effect diagram of the present invention;
fig. 4 is a diagram illustrating a process of processing a campus road image according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a road edge detection and tracking method, which comprises the following steps as shown in figure 1:
the method comprises the following steps: and preprocessing the road image to obtain a road image threshold segmentation map.
Specifically, the method for preprocessing the road image specifically comprises the following steps:
(1) and carrying out weighted average graying on the road image to obtain a grayed image.
By giving different weights to three color components of the acquired road image, when the weights of R, G, B are respectively 0.299, 0.578 and 0.114, the obtained gray level image is relatively accordant with human vision.
Therefore, the grayscale values of the grayed-out image are:
Gray(i，y)＝0.299*R(i，y)+0.578G(i，y)+0.114B(i，y)；
where R (i, y) represents a road image red component value, G (i, y) represents a road image green component value, and B (i, y) represents a road image blue component value.
(2) And extracting an ROI (region of interest) region of the gray image, and filtering out irrelevant regions to obtain a first road image. The position of the road surface in the road image collected by the vehicle-mounted camera during working is basically unchanged, and invalid information such as sky, trees, green belts and the like always appears above and on two sides of the image. Therefore, the road information needs to be intercepted by the algorithm, subsequent image processing and feature extraction are carried out, irrelevant information is filtered, and the operation rate of the algorithm can be greatly improved.
(3) And carrying out bilateral filtering on the first road image to obtain a second road image.
Bilateral filtering considers the space distance and the pixel similarity of the neighborhood of the image target pixel point at the same time to obtain two different weight value tables, and the noise interference can be filtered by considering two independent weight values at the same time, so that the effect of edge information is ensured.
And f (m, n) is set as the gray value of the target pixel point, and g (x, y) is set as the gray value of the target pixel of the second road image. The bilateral filtering formula is:
and (3) weighted combination of space and value range of neighborhood pixel points:
wherein (x, y) is coordinate value of target pixel point, (m, n) is coordinate value of neighborhood pixel point, sigmadAnd σrIs a convolution smoothing factor.
(4) And segmenting the second road image by adopting an Otsu threshold value, and separating target information and background information to obtain a road image threshold value segmentation map.
Let L gray levels in the grayed-out image be present, where the probability of occurrence of the gray value i is
Wherein N is the total number of pixels of the grayed image; n isiThe number of pixel values with the gray value of i; l is the gray level of the image. The gray value appearing in the road image is sequentially given to a threshold k, and the image is divided into a target area (M)In the background region (B), p (i) is the probability of occurrence of a pixel point with a gray value i, and the probability of occurrence of the target information is:
the probability of occurrence of background information is:
the target area gray values are:
the gray values of the background area are:
second image overall gray level mean value:
the inter-class variance between the target region and the background region is:
σ(x)2＝ωB(x)[μM(x)-μ]2+ωM(x)[μB(x)-μ]2＝ωB(x)ωM(x)[μB(x)-μM(x)]2
and when the inter-class variance between the target area and the background area is maximum, the corresponding gray value is the optimal segmentation threshold value of the road image.
Step two: and carrying out edge detection on the road image threshold segmentation graph to obtain a road image edge detection graph.
Further, edge detection is carried out on the road image threshold segmentation graph through a Canny edge detection operator.
Step three: and identifying the correct road edge straight line by the road image edge detection graph through Hough transformation with direction constraint and a probability voting method.
Further, Hough line detection is performed on the edge information of the road image edge detection graph, so that all possible edges in the image are detected, and as shown in fig. 2(a) - (b), a road edge line feature extraction effect graph is shown.
It should be noted that, in the embodiment of the present invention, direction constraints are introduced for detecting a straight line on the basis of the conventional Hough transform. Through the observation to road marking job site environment, road marking equipment marking and the visual angle of gathering the image are always located between two road edges, according to priori knowledge and on the basis of observing the research to a large amount of road images, find the angle of road edge always to be located certain angular range. Therefore, in this embodiment, with reference to the pixel coordinate system, the upper left corner of the road image is the origin of coordinates, the right side is the positive direction of the x-axis, and the downward side is the positive direction of the y-axis, the angles of the straight lines extracted by Hough transformation are sequentially calculated, the direction constraints of 15 ° to 75 ° and-15 ° to-75 ° are introduced, and the straight lines which do not satisfy the direction condition are filtered. The boundary range of the right road edge is set to be 15-75 degrees, and the angle range of the left road edge is-15-75 degrees.
Further, after the detected straight lines are subjected to direction constraint, different weight values are given to different line segments according to the length of the line segments, the weight values of the straight lines are given to the intersection points of the corresponding straight lines, and the maximum straight line intersection point is voted to be the maximum road vanishing point.
Let the line segment L in the image1Has a length of L (L)1)，L2Has a length of L (L)2) The extension lines of the two line segments intersect at a point (x)i，yj) If the point is in the statistical voting matrix, the number of votes obtained is: p (x)i，yj)＝l(L1)+l(L2). The line segment intersection point with the highest vote number is the maximum road vanishing point: v (x, y) argmax [ P (x)i，yj)]。
Interference lines in the image through the obtained maximum road vanishing point pairsThe sections are filtered out. Let ScRepresenting a set of line segments constrained by vanishing points, SiFor the line segment collection before filtering, σ is a set inspection threshold, and the road vanishing point constraint is as follows:
Sc＝{Si|||Si-VP||＜σ}
when the distance between the line segment and the road vanishing point obtained by voting is smaller than a set inspection threshold, the line segment is reserved; otherwise, the line segment is deleted.
Step four: and tracking a correct road edge straight line, and dynamically delimiting an ROI (region of interest) according to the road edge parameter information.
In this embodiment, the slope and coordinates of the correct road-edge line are tracked by a kalman filter. Specifically, the Kalman filtering method is:
let k and x1The slope and bottom coordinates of the road edge straight line are respectively, and the observation value matrix of the system is as follows:
z＝(k,x1)Tlet us say Δ k and Δ x1The change rates of the slope of the road edge straight line and the bottom coordinate are respectively, and the state vector of the system in Kalman filtering is set as follows:
x＝(k,x1,Δk,Δx1)T
the system state transition matrix is:
the measurement matrix is set to:
the slope error of the line allowed by the system is +/-5 degrees, the variation error is +/-2, the allowable deviation of the abscissa of the lower end of the line is +/-10 pixels, the variation error is +/-4, and therefore the mean square error matrix is as follows:
the system noise covariance matrix is:
the system measures the noise covariance:
and the Kalman filtering method determines the weight of the observed value and the predicted value through Kalman gain so as to obtain the optimal estimated value of the system. The system continuously performs optimal estimation on the state value of the system through iterative update of the error covariance, so that optimal prediction tracking of the state parameters of the system is realized.
The tracking effect is shown in fig. 3, in the graph, the green line and the blue line are slopes of left and right road edge straight lines obtained by constraint screening of vanishing points by using Hough transformation with constraints, and the red line and the black line are slope changes of the left and right road edges after kalman filtering is introduced. Fig. 3 illustrates that the present invention can realize stable identification and tracking of road edges under different environments.
Further, the ROI area is dynamically defined according to a Raman filter.
The slope and the coordinates of the right road edge straight line are tracked through a Kalman filter, the optimal estimation information of the road edge line is obtained, and then the ROI area of the road image can be dynamically extracted.
This embodiment uses shaded campus roads for the experiment. The road image with the interference elements is adopted for verification, and as shown in fig. 4, in the experiment, the acquired road image is preprocessed through algorithms such as weighted graying, ROI extraction, bilateral filtering, Otsu threshold segmentation and the like in sequence, so that a threshold segmentation graph of the road image is obtained. And detecting road image edge information by adopting a Canny operator, and carrying out road edge straight line detection by Hough transformation. And aiming at the problem that the road edge straight line detected by Hough transformation contains more interference line segments, Hough transformation with direction constraint and probability voting method-based road vanishing point detection are provided, and straight lines which do not meet the conditions are filtered. Aiming at the problem that the identification deviation of a road edge straight line of a certain frame is large in the marking process, a Kalman filtering method is provided for tracking the road edge straight line coordinate information, and the ROI area is dynamically divided according to the road edge information identified last time. As can be seen from FIG. 4, the method for detecting and tracking the road edge can effectively filter out interference factors and accurately identify the road edge, and has the advantages of good real-time performance and high accuracy.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (7)
1. A road edge detection and tracking method is characterized by comprising the following specific steps:
preprocessing the road image to obtain a road image threshold value segmentation map;
performing edge detection on the road image threshold segmentation graph, and obtaining a road image edge detection graph by adopting a Canny edge detection operator;
identifying a correct road edge straight line from the road image edge detection graph through Hough transformation with direction constraint and a probability voting method;
and tracking the correct road edge straight line, and dynamically delimiting an ROI (region of interest) according to the road edge parameter information.
2. The method for detecting and tracking road edges according to claim 1, wherein the preprocessing of the road image specifically comprises the following steps:
carrying out weighted average graying on the road image to obtain a grayed image;
extracting an ROI (region of interest) region of the grayed image, and filtering out irrelevant regions to obtain a first road image;
carrying out bilateral filtering on the first road image to obtain a second road image;
and segmenting the second road image by adopting an Otsu threshold value, and separating target information and background information to obtain a road image threshold value segmentation map.
3. The road edge detection and tracking method according to claim 1, wherein edge detection is performed on the road image threshold segmentation map by a Canny edge detection operator.
4. The road edge detection and tracking method according to claim 1, wherein the slope and coordinates of the correct road edge straight line are tracked through a kalman filter; the ROI area is dynamically defined according to a Raman filter.
5. The method for detecting and tracking road edge according to claim 2, wherein the grayscale values of the grayed image are:
Gray(i，y)＝0.299*R(i，y)+0.578G(i，y)+0.114B(i，y)；
where R (i, y) represents a road image red component value, G (i, y) represents a road image green component value, and B (i, y) represents a road image blue component value.
6. The method for detecting and tracking the road edge according to claim 1, wherein direction constraints of 15 degrees to 75 degrees and-15 degrees to-75 degrees are added to the straight line extracted through the Hough transformation, and the straight line which does not meet the direction condition is filtered.
7. The method for detecting and tracking road edges according to claim 1, wherein the probability voting method comprises the following specific steps:
different line segments are endowed with different weights according to the length of the line segments, the weights are endowed with the intersection points of corresponding straight lines, and the maximum straight line intersection point, namely the maximum road vanishing point, is voted;
when the distance between the line segment and the maximum road vanishing point is smaller than a set threshold value, the line segment is reserved;
and when the distance between the line segment and the maximum road vanishing point is larger than a set threshold value, deleting the line segment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110535212.XA CN113223034A (en) | 2021-05-17 | 2021-05-17 | Road edge detection and tracking method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110535212.XA CN113223034A (en) | 2021-05-17 | 2021-05-17 | Road edge detection and tracking method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113223034A true CN113223034A (en) | 2021-08-06 |
Family
ID=77092459
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110535212.XA Pending CN113223034A (en) | 2021-05-17 | 2021-05-17 | Road edge detection and tracking method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113223034A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113793354A (en) * | 2021-08-20 | 2021-12-14 | 航天晨光股份有限公司 | Self-adaptive straight line detection method based on Hough transform |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105893949A (en) * | 2016-03-29 | 2016-08-24 | 西南交通大学 | Lane line detection method under complex road condition scene |
| WO2018023916A1 (en) * | 2016-08-01 | 2018-02-08 | 北京大学深圳研究生院 | Shadow removing method for color image and application |
| CN109583435A (en) * | 2019-01-24 | 2019-04-05 | 安徽理工大学 | A kind of adaptive ballot method for detecting lane lines |
| CN112001216A (en) * | 2020-06-05 | 2020-11-27 | 商洛学院 | Automobile driving lane detection system based on computer |
-
2021
- 2021-05-17 CN CN202110535212.XA patent/CN113223034A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105893949A (en) * | 2016-03-29 | 2016-08-24 | 西南交通大学 | Lane line detection method under complex road condition scene |
| WO2018023916A1 (en) * | 2016-08-01 | 2018-02-08 | 北京大学深圳研究生院 | Shadow removing method for color image and application |
| CN109583435A (en) * | 2019-01-24 | 2019-04-05 | 安徽理工大学 | A kind of adaptive ballot method for detecting lane lines |
| CN112001216A (en) * | 2020-06-05 | 2020-11-27 | 商洛学院 | Automobile driving lane detection system based on computer |
Non-Patent Citations (2)
| Title |
|---|
| 邱东等: "基于改进概率霍夫变换的车道线快速检测方法", 《计算机技术与发展》 * |
| 陈涛等: "基于优先像素与卡尔曼滤波追踪的车道线检测", 《汽车工程》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113793354A (en) * | 2021-08-20 | 2021-12-14 | 航天晨光股份有限公司 | Self-adaptive straight line detection method based on Hough transform |
| CN113793354B (en) * | 2021-08-20 | 2023-08-15 | 航天晨光股份有限公司 | Self-adaptive straight line detection method based on Hough transform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110866903B (en) | Ping-pong ball identification method based on Hough circle transformation technology | |
| CN108921176B (en) | Pointer instrument positioning and identifying method based on machine vision | |
| CN107679520B (en) | Lane line visual detection method suitable for complex conditions | |
| CN107330376B (en) | Lane line identification method and system | |
| CN109325935B (en) | Power transmission line detection method based on unmanned aerial vehicle image | |
| CN107909081B (en) | Method for quickly acquiring and quickly calibrating image data set in deep learning | |
| CN108921813B (en) | Unmanned aerial vehicle detection bridge structure crack identification method based on machine vision | |
| CN109993099A (en) | A kind of lane line drawing recognition methods based on machine vision | |
| CN101853333A (en) | Method for picking marks in medical robot navigation positioning images | |
| CN106778551A (en) | A kind of fastlink and urban road Lane detection method | |
| CN114758322B (en) | Road quality detection system based on machine identification | |
| CN112528868B (en) | Illegal line pressing judgment method based on improved Canny edge detection algorithm | |
| CN110298344A (en) | A kind of positioning of instrument knob and detection method based on machine vision | |
| CN112883824A (en) | Finger vein feature recognition device for intelligent blood sampling and recognition method thereof | |
| CN111476804A (en) | Method, device and equipment for efficiently segmenting carrier roller image and storage medium | |
| CN115841633A (en) | Power tower and power line associated correction power tower and power line detection method | |
| CN107516315B (en) | Tunneling machine slag tapping monitoring method based on machine vision | |
| CN111768455A (en) | Image-based wood region and dominant color extraction method | |
| CN115018785A (en) | Hoisting steel wire rope tension detection method based on visual vibration frequency identification | |
| CN113223034A (en) | Road edge detection and tracking method | |
| CN117197700B (en) | Intelligent unmanned inspection contact net defect identification system | |
| CN107977608B (en) | Method for extracting road area of highway video image | |
| CN113053164A (en) | Parking space identification method using look-around image | |
| CN113298725A (en) | Correction method for superposition error of ship icon image | |
| Gupta et al. | Robust lane detection using multiple features |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210806 |