CN115984806B - Dynamic detection system for road marking damage - Google Patents
Dynamic detection system for road marking damage Download PDFInfo
- Publication number
- CN115984806B CN115984806B CN202310265394.2A CN202310265394A CN115984806B CN 115984806 B CN115984806 B CN 115984806B CN 202310265394 A CN202310265394 A CN 202310265394A CN 115984806 B CN115984806 B CN 115984806B
- Authority
- CN
- China
- Prior art keywords
- marking
- image
- road marking
- module
- 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.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a dynamic detection system for road marking damage, which comprises an image conversion module, an ROI (region of interest) area extraction module, an image denoising module, a road marking positioning module, a road marking classification module and a road marking damage calculation module, wherein the acquired road marking image is subjected to image conversion to obtain a road marking aerial view, the aerial view image is subjected to ROI area extraction to obtain an interested area in the image, the interested area of the image is subjected to pretreatment, the pretreated image is subjected to road marking positioning, and finally marking damage is calculated according to a road marking classification result, so that the damage detection of straight lines and arrow marks is finally realized, and the dynamic detection system has the advantages of good effect and high accuracy.
Description
Technical Field
The invention relates to the technical field of intelligent traffic research, in particular to a dynamic detection system for road marking damage.
Background
The lane lines are important components of road traffic marking lines and marks, and traffic information such as guidance, restriction, warning and the like is transmitted to traffic participants through lines, arrows, characters and the like so as to ensure the life safety of vehicles and pedestrians. Because it is mainly marked on the road surface, it is subject to sun, rain, wind and rain freezing and impact abrasion of vehicle, so that it is easy to produce breakage and even missing.
Traditional lane line breakage mainly depends on manual visual detection, and shooting is performed to carry out later breakage statistics, so that time is consumed and labor cost is huge; road marking breakage detection based on unmanned aerial vehicle aerial photography cannot detect road marking shielded by buildings such as bridges and tunnels, and the detection range is limited by electric power; the other type is based on the vehicle-mounted camera to detect the damage of the lane lines, only the straight line for detecting the road dividing line is met, and the detection of the arrow indication marked lines is not realized.
Therefore, the invention provides a dynamic road marking damage detection system for simultaneously detecting a line marking and an arrow indication marking based on a vehicle-mounted camera and a single lane, so as to solve the technical problems.
Disclosure of Invention
In order to solve the technical problems, the invention provides a dynamic detection system for road marking damage.
The invention provides a dynamic detection system for road marking breakage, which comprises an image conversion module, a detection module and a detection module, wherein the image conversion module is used for converting a road marking image into a road marking aerial view by utilizing aerial view conversion based on affine projection conversion;
the ROI area extraction module is used for extracting an interested area of the road marking aerial view so as to restrict the detection range of the road marking aerial view;
the image denoising module is used for eliminating noise interference in the road marking aerial view on the premise of retaining image characteristics;
the road marking positioning module is used for extracting marking edges in the road marking aerial view and fitting out the minimum circumscribed rectangle of the marking based on the marking edges;
the road marking classification module is used for classifying the road marking based on the center point coordinate of the minimum circumscribed rectangle of the marking, and the classification category comprises straight line type demarcation marking and arrow type indication marking;
the road marking damage calculation module is used for detecting the damage proportion of arrow indication markings by adopting a filling method based on morphology and detecting the damage proportion of straight line demarcation markings by adopting a pixel metering method.
Preferably, the system further comprises an image acquisition module for acquiring the road marking image by using the vision sensor.
Preferably, the ROI area extraction module includes determining a region of interest in the road marking bird's eye view according to a position of the lane portion in the road marking image and in combination with the driving offset.
Preferably, the image denoising module performs gaussian filtering calculation by using a two-dimensional zero-mean discrete gaussian function as a smoothing filter, obtains weights from each point to a center point, and then performs convolution operation on the weights and the image to perform bilateral filtering treatment, wherein a formula of the two-dimensional zero-mean discrete gaussian function is as follows:
wherein σ represents the size of the gaussian kernel, (x, y represents any coordinate point of the image, G (x, y represents a two-dimensional distribution of gaussian;
the convolution operation formula is:
wherein g (i, j) represents the resulting pixel value of the output point (i, j); (i, j) and (k, l) represent points in the image, respectively, S (i, j) refers to a set of points of a size range of (2n+1) centered on (i, j), f (k, l) represents an original pixel value of the point (k, l), and w (i, j, k, l) represents a weighting coefficient calculated by two gaussian functions.
Preferably, the road marking positioning module includes:
the primary edge detection module is used for detecting the mutation of the image edge gray level by utilizing an edge detection operator Sobel to obtain an edge amplitude image;
the edge image one-time enhancement module is used for enhancing the contrast ratio of the edge amplitude image according to the maximum proportion and enhancing the marking edge information according to the gray value of the maximized image;
the secondary edge detection module is used for carrying out secondary edge detection on the image after primary enhancement by utilizing a canny operator;
an edge image secondary enhancement module for enhancing the contrast of the high frequency region (edges and corners) of the image and enhancing the graticule edge information again;
the threshold segmentation module is used for carrying out threshold segmentation on the image with the secondarily enhanced edge;
the morphological processing module is used for carrying out operation on the image so as to eliminate small points and burrs which are obviously isolated around the edge of the marking line;
and the marking fitting and positioning module is used for obtaining the minimum circumscribed rectangle of the marking edge according to the extracted marking edge.
Preferably, the operation on the image includes an open operation or a closed operation.
Preferably, the road marking classification module further comprises marking a coordinate center point of the minimum bounding rectangle as a left straight line type boundary marking within a left third of the image;
marking the coordinate center point of the minimum circumscribed rectangle as a middle arrow indicating marking line within the middle third range of the image;
and marking the coordinate center point of the minimum circumscribed rectangle as a right straight line type demarcation line within the right third range of the image.
Preferably, the method for detecting the breakage proportion of the arrow indicating marking line based on the filling method based on morphology comprises the following operation steps:
extracting a marking area according to the marking outline, performing local binarization operation on the marking area, and calculating a marking area A1;
performing morphological closing operation (namely performing expansion operation and then performing corrosion operation) on the area, and then calculating the area A2 again;
calculating the difference between A1 and A2, and calculating the breakage percentage, wherein the breakage proportion calculation formula is as follows
Preferably, the pixel metering method for detecting the breakage proportion of the straight line demarcation line comprises the following operation steps:
extracting a complete unbroken marking, and calculating the marking area B1;
performing OSTU threshold segmentation on the positioning marked lines to extract the marked lines, and calculating the area B2 of the region;
calculating the difference between B1 and B2, and calculating the breakage percentage, wherein the breakage proportion calculation formula is as follows
Compared with the related art, the dynamic detection system for road marking damage provided by the invention has the following beneficial effects:
according to the invention, the acquired road marking image is subjected to image conversion to obtain the road marking aerial view, the aerial view image is subjected to ROI region extraction to obtain the region of interest in the image, the region of interest of the image is subjected to pretreatment, the pretreated image is subjected to road marking positioning, and finally the marking breakage is calculated according to the road marking classification result, so that the breakage detection of straight lines and arrow marks is finally realized, and the method has the advantages of good effect and high accuracy.
Drawings
FIG. 1 is a schematic image acquisition diagram of a dynamic road marking breakage detection system provided by the invention;
FIG. 2 is a flow chart of a marking breakage detection algorithm of the dynamic road marking breakage detection system provided by the invention;
FIG. 3 is a flow chart of image preprocessing of a dynamic detection system for road marking breakage provided by the invention;
FIG. 4 is a flow chart of the line locating of the dynamic detection system for road line breakage provided by the invention;
FIG. 5 is a flow chart of a calculation of the classification and breakage of a road marking in the dynamic detection system of road marking according to the present invention;
fig. 6 is a flowchart of a dynamic road marking breakage detection system provided by the invention.
Detailed Description
The invention will be further described with reference to the drawings and embodiments.
Example 1
In the present embodiment, as shown in the flowchart of fig. 6, the following operations are included:
in the present embodiment, as shown in fig. 2, the dynamic detection system for road marking breakage includes an image conversion module corresponding to an aerial view conversion for converting a road marking image into an aerial view of the road marking by utilizing the aerial view conversion based on affine projection conversion;
an ROI region extraction module corresponding to the ROI region extraction, configured to extract a region of interest of the road-marking aerial view, so as to constrain a detection range of the road-marking aerial view;
the image denoising module is used for eliminating noise interference in the road marking aerial view on the premise of retaining image characteristics;
the road marking positioning module is used for extracting the marking edge in the road marking aerial view and fitting out the minimum circumscribed rectangle of the marking based on the marking edge;
the road marking classification module is used for classifying the road marking based on the center point coordinates of the minimum circumscribed rectangle of the marking edge, and the classification category comprises a straight line type demarcation marking and an arrow type indication marking;
the road marking damage calculation module is used for detecting the damage proportion of the arrow indicating marking by adopting a filling method based on morphology and detecting the damage proportion of the straight line demarcation marking by adopting a pixel metering method;
the system also comprises an image acquisition module corresponding to the input image, which is used for acquiring the road marking image by using the vision sensor;
when the road marking damage dynamic detection system is implemented by combining all modules, as shown in fig. 2, the system comprises the following operation steps:
s101: collecting road marking images by using image collecting equipment, and inputting the road marking images into a detection system; in particular, the image capturing device may be a vision sensor, or any other device capable of capturing an image of a road marking, which is not particularly limited herein.
S102: performing bird' S eye conversion based on affine projection transformation on the road marking image acquired in S101
Specifically, the process of acquiring an image by the image acquisition device is a process of projecting a three-dimensional object in the real world onto a two-dimensional plane, which can be explained according to a pinhole camera model, and is not described herein, but because the object will take on an irregular shape with a near size and a far size (as shown in the left diagram in fig. 1) under the perspective angle, in a transverse area, the road markings have different widths at different positions, and take on a phenomenon of near width and far width (as shown in the left diagram in fig. 1); in the longitudinal region, the actual distance represented by each pixel is different (as shown in the right-hand diagram in fig. 1);
therefore, in the road marking breakage detection according to the present invention, the road marking is converted into the road marking plan view by bird's eye conversion.
S103: ROI region extraction for bird's eye-converted road marking image
In the specific implementation process, the bird's eye view information is analyzed to find that the image contains the useful lane part and also contains objects with complex shapes such as sky, vehicles, trees, pedestrians and the like, so that great interference is caused to detection, and the detection range is restrained by a method of selecting an interested region because the position of the lane part appears in each frame of image is fixed, the information interference outside the lane is eliminated, and the calculated amount in the detection process is reduced;
the specific selection mode is as follows: according to the position of the lane part in the image, the left and right normal offset in the driving process is considered, the offset is left and right offset by half a lane, and the part is reserved as the region of interest.
S104: as shown in FIG. 3, the region of interest in the road marking image is preprocessed to achieve the effects of edge protection and denoising
In an actual driving environment, the influence of factors such as shaking of a camera, rapid movement of a vehicle, environmental changes such as weather and illumination, noise generated in the imaging process of the camera and the like greatly influences the lane line detection effect, so that the interference needs to be eliminated to the greatest extent by using a common image preprocessing technology, and the method comprises Gaussian filtering processing and bilateral filtering processing which are needed to be carried out in sequence;
the Gaussian filter processing mainly suppresses noise of an image under the condition of retaining image characteristics as much as possible, and meanwhile, as the background color of a pavement is relatively simple, but small cracks and pits of concrete or asphalt pavement are more, a two-dimensional zero-mean discrete Gaussian function is adopted as a smoothing filter, and the Gaussian function is as follows:
where σ represents the size of the gaussian kernel, the greater σ, the better the smoothness of the gaussian filter, (x, y represents any coordinate point of the image, and G (x, y represents a two-dimensional distribution of gaussian).
The bilateral filtering processing is to optimize each weight calculated by the spatial proximity from each point to the center point in Gaussian filtering (spatial proximity), optimize the weight to be the product of the weight calculated by the spatial proximity and the weight calculated by the pixel value similarity, and then perform convolution operation on the optimized weight and the image, thereby achieving the effect of edge protection and denoising; wherein the convolution operation is
Wherein g (i, j) represents the resulting pixel value of the output point (i, j); (i, j) and (k, l) represent points in the image, respectively, S (i, j) refers to a set of points of a size range of (2n+1) centered on (i, j), f (k, l) represents an original pixel value of the point (k, l), and w (i, j, k, l) represents a weighting coefficient calculated by two gaussian functions.
S105: the road marking is positioned as shown in fig. 4, and specifically comprises the following processing steps:
s201: performing primary edge detection, and completing the step by a primary edge detection module
In this embodiment, specifically, the edge refers to that where the pixel gray level change is most significant on the image, the edge detection Sobel operator detects the edge by using the abrupt change of the image edge gray level, the Sobel operator includes two sets of 3X3 filters, which are sensitive to the edge in the horizontal and vertical directions, respectively, the two direction templates convolve with the image along the X-axis and the Y-axis, respectively, the directions are from top to bottom and from left to right, the center of the template coincides with a certain pixel on the image, and the points around the pixel are multiplied by the coefficients on the template, as shown below, where G (X) and G (Y) represent the gradient values of the image detected by the lateral and longitudinal edges, respectively.
G(X)＝(X3+2X6+X9)-(X1+2X4+X7)
G(Y)＝(X1+2X2+X3)-(X7+2X8+X9)
The magnitude of the gradient value G of each pixel point on the image is calculated by combining the transverse gradient value and the longitudinal gradient value of the point through the following formula:
s202: edge image one-time enhancement
In the embodiment, specifically, after Sobel performs edge detection once, an edge amplitude image is obtained, the contrast of the edge amplitude image is enhanced according to the maximum proportion, and meanwhile, the marking edge information is enhanced according to the gray value of the maximized image; the specific method comprises the following steps: the maximum and minimum values of the pixels are calculated, the pixels are scaled according to the maximum value, and the gray value is stretched to 0-255.
S203: performing secondary edge detection on the image after primary edge enhancement
In this embodiment, specifically, the image after the edge is enhanced once is subjected to secondary edge detection by using a canny operator; the specific method comprises the following steps: firstly carrying out Gaussian blur on an image after primary edge enhancement, then calculating the gradient amplitude and direction of the image after Gaussian blur, and finally carrying out non-maximum suppression on the amplitude image.
S204: performing edge image secondary enhancement on an edge amplitude image obtained after canny secondary edge detection
In this embodiment, specifically, an edge amplitude image is obtained after a canny secondary edge is detected, the contrast of a high-frequency region (edge and corner) of the image is enhanced, and the edge information of the marking is enhanced again to obtain an edge secondary enhanced image; the specific method comprises the following steps: new pixel value = pixel value of the point of the original image-average pixel value of pixel values of all points in the kernel size range, then approximately rounding (which may have a negative value) the result, and adding the value to the pixel value of the original point to obtain the final result to replace the pixel value of the original point.
S205: threshold segmentation of edge secondarily enhanced images
In this embodiment, specifically, threshold segmentation is performed on the image after edge secondary enhancement; the image f (x, y) comprises an object, a background and noise, and a certain threshold T is set to divide the image into two parts: since the pixel groups larger than T and the pixel groups smaller than T are not necessarily simply distributed in two gray scale ranges between the actually obtained image object and the background, two or more thresholds are required to extract the object,
wherein, T1: a lower threshold of gray value representing an image, T2: representing the upper gray value threshold of the image.
S206: morphological processing of thresholded segmented images
In this embodiment, specifically, the thresholded and segmented image has significantly isolated small points and burrs around the edges of the reticle, and is eliminated by performing an open operation on the image. The open operation is a filter based on geometric operation, and comprises two processes of corrosion and expansion, wherein the area of an image target is not obviously changed while small points and burrs are eliminated, and the corrosion is to convolve the image with structural elements (or also called kernels) and has no requirement on the size and shape of the structural elements.
S207: line edge fitting positioning
In this embodiment, specifically, according to the extracted edges of the marks, the minimum circumscribed rectangles of the left-side marks, the middle marks and the right-side mark detection areas are respectively obtained, and the center point coordinates of the minimum circumscribed rectangles of the marks are obtained.
S106: coordinate-based reticle classification
In this embodiment, specifically, the reticle is divided into a straight line type demarcation reticle and an arrow type indication reticle based on the center point coordinates of the minimum circumscribed rectangle of the reticle; the coordinate center point belongs to a left straight line type boundary line and a mark in the left third of the image, the coordinate center point belongs to a middle arrow type indication mark in the middle third of the image, and the coordinate center point belongs to a right straight line type boundary line and a mark in the right third of the image; the image coordinates divide the reticle into a linear type boundary reticle and an arrow type indication reticle, and are calculated based on different types of reticle classifications.
S107: respectively carrying out damage calculation on different kinds of marked lines
In this embodiment, specifically, as shown in fig. 5, the marked lines are indicated by arrows, and most cases are internal wear, which is expressed as: more cracks or holes appear. Aiming at the internal abrasion type marking, a marking abrasion detection algorithm is designed, and the algorithm mainly adopts morphological operation and marking physical morphological characteristics to detect damage, so that a filling method based on morphology is adopted;
the marking damage detection is carried out by comparing the marking area difference before and after morphological operation, and the specific operation of the detection algorithm is as follows:
extracting a marking area according to the marking outline, performing local binarization operation on the marking area, and calculating a marking area A1;
performing morphological closing operation (namely performing expansion operation and then performing corrosion operation) on the area, and then calculating the area A2 again;
calculating the difference between A1 and A2, and calculating the breakage percentage, wherein the breakage proportion calculation formula is as follows
For the straight line demarcation marked line, the main characteristics of the broken line type demarcation marked line are fracture and edge corrosion, the cut-off type breakage is mainly characterized in that the breakage occurs at a certain position in the middle of the road marked line due to certain reasons, so that the original marked line is divided into two or more sections, and the edge corrosion marked line is mainly characterized in that the width of the marked line is narrowed due to the abrasion of the edge, so that a pixel metering method is adopted;
the reticle breakage detection is performed by comparing the standard perfect reticle with the extracted reticle area difference, the detection algorithm is specifically operated as follows, as shown in figure 5,
and extracting the complete unbroken marked line, and calculating the marked line area B1.
And (3) carrying out OSTU threshold segmentation on the positioning marked lines to extract the marked lines, and calculating the area B2.
Calculating the difference between B1 and B2, and calculating the breakage percentage, wherein the breakage proportion calculation formula is as follows:
example two
Unlike the first embodiment, the following is: morphological processing of thresholded segmented images
In the embodiment, specifically, the thresholded and segmented image has obviously isolated small points and burrs around the edge of the marking line, and the small points and burrs are eliminated by performing a closed operation on the image; the closed operation is opposite to the open operation, and the closed operation is to expand the image and then corrode the image, and then perform the open operation on the marked line image and then perform the closed operation to remove the marked line edge interference.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes or direct or indirect application in other related technical fields are included in the scope of the present invention.
Claims (7)
1. A dynamic road marking breakage detection system, comprising:
an image conversion module for converting the road marking image into a road marking aerial view by utilizing aerial view conversion based on affine projection conversion;
the ROI area extraction module is used for extracting an interested area of the road marking aerial view so as to restrict the detection range of the road marking aerial view;
the image denoising module is used for eliminating noise interference in the road marking aerial view on the premise of retaining image characteristics;
the road marking positioning module is used for extracting marking edges in the road marking aerial view and fitting out the minimum circumscribed rectangle of the marking based on the marking edges;
the road marking classification module is used for classifying the road marking based on the center point coordinate of the minimum circumscribed rectangle of the marking, and the classification category comprises straight line type demarcation marking and arrow type indication marking;
the road marking damage calculation module is used for detecting the damage proportion of arrow indication markings by adopting a filling method based on morphology and detecting the damage proportion of straight line demarcation markings by adopting a pixel metering method;
the morphological filling method comprises the following operation steps of:
extracting a marking area according to the marking outline, performing local binarization operation on the marking area, and calculating a marking area A1;
performing morphological closing operation on the region, and then calculating the region area A2 again;
calculating the difference between A1 and A2, and calculating the breakage percentage, wherein the breakage proportion calculation formula is as follows
The pixel metering method for detecting the damage proportion of the straight line demarcation line comprises the following operation steps:
extracting a complete unbroken marking, and calculating the marking area B1;
performing OSTU threshold segmentation on the positioning marked lines to extract the marked lines, and calculating the area B2 of the region; calculating the difference between B1 and B2, and calculating the breakage percentage, wherein the breakage proportion calculation formula is as follows
2. The dynamic road marking breakage detection system of claim 1, further comprising an image acquisition module for acquiring road marking images using the vision sensor.
3. The dynamic road marking breakage detection system according to claim 2, wherein the ROI area extraction module includes determining the region of interest in the road marking bird's eye view in combination with the driving offset based on the position of the lane portion in the road marking image.
4. The dynamic detection system for road marking breakage according to claim 3, wherein the image denoising module comprises a step of performing gaussian filtering calculation by using a two-dimensional zero-mean discrete gaussian function as a smoothing filter, obtaining weights from each point to a center point, and performing convolution operation on the weights and the image to perform bilateral filtering processing, wherein the formula of the two-dimensional zero-mean discrete gaussian function is as follows:
wherein σ represents the size of the gaussian kernel, (x, y represents any coordinate point of the image, G (x, y represents a two-dimensional distribution of gaussian;
the convolution operation formula is:
wherein g (i, j) represents the resulting pixel value of the output point (i, j); (i, j) and (k, l) represent points in the image, respectively, S (i, j) refers to a set of points of a size range of (2n+1) centered on (i, j), f (k, l) represents an original pixel value of the point (k, l), and w (i, j, k, l) represents a weighting coefficient calculated by two gaussian functions.
5. The dynamic road marking breakage detection system of claim 4, wherein said road marking locating module comprises:
the primary edge detection module is used for detecting the mutation of the image edge gray level by utilizing an edge detection operator Sobel to obtain an edge amplitude image;
the edge image one-time enhancement module is used for enhancing the contrast ratio of the edge amplitude image according to the maximum proportion and enhancing the marking edge information according to the gray value of the maximized image;
the secondary edge detection module is used for carrying out secondary edge detection on the image after primary enhancement by utilizing a canny operator;
the edge image secondary enhancement module is used for enhancing the contrast of a high-frequency region of the image and enhancing the marking edge information again;
the threshold segmentation module is used for carrying out threshold segmentation on the image with the secondarily enhanced edge;
the morphological processing module is used for carrying out operation on the image so as to eliminate small points and burrs which are obviously isolated around the edge of the marking line;
and the marking fitting and positioning module is used for obtaining the minimum circumscribed rectangle of the marking edge according to the extracted marking edge.
6. The dynamic road marking breakage detection system according to claim 5, wherein said operation of operating on the image includes an open operation or a closed operation.
7. The dynamic road marking breakage detection system according to claim 1, wherein the road marking classification module further comprises marking the coordinate center point of the minimum bounding rectangle as a left straight line type boundary marking within the left third of the image;
marking the coordinate center point of the minimum circumscribed rectangle as a middle arrow indicating marking line within the middle third range of the image;
and marking the coordinate center point of the minimum circumscribed rectangle as a right straight line type demarcation line within the right third range of the image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310265394.2A CN115984806B (en) | 2023-03-20 | 2023-03-20 | Dynamic detection system for road marking damage |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310265394.2A CN115984806B (en) | 2023-03-20 | 2023-03-20 | Dynamic detection system for road marking damage |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115984806A CN115984806A (en) | 2023-04-18 |
| CN115984806B true CN115984806B (en) | 2023-06-13 |
Family
ID=85968562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310265394.2A Active CN115984806B (en) | 2023-03-20 | 2023-03-20 | Dynamic detection system for road marking damage |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115984806B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118097642A (en) * | 2024-04-19 | 2024-05-28 | 山东冠通交通设施有限公司 | Traffic engineering marking line clearing quality assessment method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105891220A (en) * | 2016-06-02 | 2016-08-24 | 昆山市交通工程试验检测中心有限公司 | Pavement marker line defect detecting device and detecting method thereof |
| CN114037970A (en) * | 2021-11-19 | 2022-02-11 | 中国重汽集团济南动力有限公司 | Sliding window-based lane line detection method, system, terminal and readable storage medium |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4811201B2 (en) * | 2005-12-06 | 2011-11-09 | 日産自動車株式会社 | Runway boundary line detection apparatus and runway boundary line detection method |
| CN108647593A (en) * | 2018-04-26 | 2018-10-12 | 东华大学 | Unmanned plane road surface breakage classification and Detection method based on image procossing and SVM |
| JP7358762B2 (en) * | 2019-04-02 | 2023-10-11 | トヨタ自動車株式会社 | Road anomaly detection device, road anomaly detection method, and road anomaly detection program |
| JP7380824B2 (en) * | 2020-02-20 | 2023-11-15 | 日本電信電話株式会社 | Vehicle state estimation method, vehicle state estimation device, and vehicle state estimation program |
| CN112396041B (en) * | 2021-01-19 | 2021-04-06 | 四川京炜数字科技有限公司 | Road marking alignment system based on image recognition |
| CN114241436A (en) * | 2021-10-09 | 2022-03-25 | 淮阴工学院 | Lane line detection method and system for improving color space and search window |
| CN115619972A (en) * | 2022-10-26 | 2023-01-17 | 北京工业大学 | Road marking abrasion identification and evaluation method based on point cloud data |
-
2023
- 2023-03-20 CN CN202310265394.2A patent/CN115984806B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105891220A (en) * | 2016-06-02 | 2016-08-24 | 昆山市交通工程试验检测中心有限公司 | Pavement marker line defect detecting device and detecting method thereof |
| CN114037970A (en) * | 2021-11-19 | 2022-02-11 | 中国重汽集团济南动力有限公司 | Sliding window-based lane line detection method, system, terminal and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115984806A (en) | 2023-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111145161B (en) | Pavement crack digital image processing and identifying method | |
| CN109615611B (en) | Inspection image-based insulator self-explosion defect detection method | |
| CN109784344B (en) | Image non-target filtering method for ground plane identification recognition | |
| CN107330376B (en) | Lane line identification method and system | |
| CN105260699B (en) | A kind of processing method and processing device of lane line data | |
| CN109657632B (en) | Lane line detection and identification method | |
| CN109325935B (en) | Power transmission line detection method based on unmanned aerial vehicle image | |
| CN105447512B (en) | A kind of detection method and device for the beauty defect that essence slightly combines | |
| CN108038416B (en) | Lane line detection method and system | |
| CN104134209B (en) | A kind of feature extracting and matching method and system in vision guided navigation | |
| Li et al. | Nighttime lane markings recognition based on Canny detection and Hough transform | |
| CN108921813B (en) | Unmanned aerial vehicle detection bridge structure crack identification method based on machine vision | |
| CN102938057B (en) | A kind of method for eliminating vehicle shadow and device | |
| CN102693423A (en) | Method for precise positioning of license plate in strong light conditions | |
| CN108171695A (en) | A kind of express highway pavement detection method based on image procossing | |
| CN107220976B (en) | Highway positioning method for aerial highway image | |
| CN108198417B (en) | A kind of road cruising inspection system based on unmanned plane | |
| CN202134079U (en) | Unmanned vehicle lane marker line identification and alarm device | |
| CN106815583A (en) | A kind of vehicle at night license plate locating method being combined based on MSER and SWT | |
| CN109635737A (en) | Automobile navigation localization method is assisted based on pavement marker line visual identity | |
| Li et al. | Road markings extraction based on threshold segmentation | |
| CN115984806B (en) | Dynamic detection system for road marking damage | |
| Liu et al. | Lane line detection based on mask R-CNN | |
| Wei et al. | Detection of lane line based on Robert operator | |
| CN109800641B (en) | Lane line detection method based on threshold value self-adaptive binarization and connected domain analysis |
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 |