CN108229327B - Lane line detection method, device and system based on background reconstruction - Google Patents
Lane line detection method, device and system based on background reconstruction Download PDFInfo
- Publication number
- CN108229327B CN108229327B CN201711284745.5A CN201711284745A CN108229327B CN 108229327 B CN108229327 B CN 108229327B CN 201711284745 A CN201711284745 A CN 201711284745A CN 108229327 B CN108229327 B CN 108229327B
- Authority
- CN
- China
- Prior art keywords
- image
- lane line
- background
- median filtering
- processing
- 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
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method, a device and a system for detecting lane lines based on background reconstruction, which are used for quickly and efficiently detecting the lane lines in a road image. The lane line detection method comprises the following steps: acquiring lane line characteristics of a road image to be detected based on background reconstruction; carrying out image processing on the lane line characteristics to obtain a processing result; and performing Blob analysis on the processing result to confirm the image blocks of the lane lines on the road image to be detected. According to the invention, the background reconstruction is carried out on the road image to be detected, and the lane line characteristics are rapidly obtained, so that the algorithm flow for detecting the road lane line is simpler and more efficient, and the rapid detection of the lane line is realized.
Description
Technical Field
The invention relates to the technical field of digital image processing, in particular to a lane line detection method, a lane line detection device and a lane line detection system based on background reconstruction.
Background
A lane departure warning system and a lane keeping system are two important functional modules of an advanced driving assistance system, wherein the core technology is a lane line detection algorithm. With the vigorous development of unmanned driving technology and advanced driving assistance systems, more and more lane line detection algorithms are emerging. Due to the inherent imaging characteristics of the camera and the influence of external environment light, the gray values of different areas on the collected image are different, and the expression form is that the gray value at the far and near positions of the camera is smaller, and the gray value at the farther position is larger. The background needs to be modeled to extract the lane lines more effectively, however, the traditional lane line detection algorithm has insufficient importance on the background modeling. In addition, the conventional lane line detection algorithm generally extracts characteristic information such as color, edge, texture direction and the like of a lane line in an image, then performs lane line detection by using a least square fitting method or RANSAC (random sample consensus) in combination with Hough transformation and related variant algorithms (such as probability Hough transformation), and finally tracks the lane line by using algorithms such as kalman filtering, particle filtering and the like. The whole algorithm has complex and fussy flow, so that the algorithm processing time is too long.
In view of the above, the present invention is particularly proposed.
Disclosure of Invention
The invention mainly aims to disclose a lane line detection method, a lane line detection device and a lane line detection system based on background reconstruction, which are used for solving the problem that the prior art has complicated and fussy algorithm flow for lane line detection, so that the algorithm processing time is too long.
In order to achieve the above object, according to an aspect of the present invention, a lane line detection method based on background reconstruction is disclosed, and the following technical solution is adopted:
a lane line detection method based on background reconstruction comprises the following steps: acquiring lane line characteristics of a road image to be detected based on background reconstruction; carrying out image processing on the lane line characteristics to obtain a processing result; and performing Blob analysis on the processing result to confirm the image blocks of the lane lines on the road image to be detected.
Further, the obtaining of the lane line characteristics of the road image to be detected based on the background reconstruction includes: carrying out median filtering processing on the region of interest of the road image to be detected to obtain a median filtering image keeping effective edge information; carrying out background modeling processing on the median filtering image to obtain a background image; and subtracting the background image from the median filtering image to obtain the lane line characteristics of the road image to be detected.
Further, performing background modeling processing on the median filtered image to obtain a background image of the median filtered image includes: projecting the median filtering image to the Y axis to obtain a projection curve of the median filtering image; performing baseline drift correction on the projection curve to obtain a background baseline curve of the median filtering image; and generating the background image according to the background baseline curve.
Further, the performing baseline shift correction on the projection curve to obtain a background baseline curve of the median filtered image includes: the preset smoothing condition is expressed by minimizing a least square function with a penalty term, and is represented by an equation (1):
where y is a signal of length m, z is another sequence, and λ is a compromise factor;
wherein, Delta2ziRepresents the second derivative at point i, i.e.:
Δ2zi＝(zi-zi-1)-(zi-1-zi-2)＝zi-2zi-1+zi-2 (2)
the asymmetric least square method introduces a weight vector w on the basis of the formula (1), and obtains a minimized formula (3):
deriving equation system (4) from minimization equation (3):
(W+λDTD)z＝Wy (4)
where W is a diagonal matrix of W, i.e., W ═ diag (W), and D is a matrix of second derivatives of z, i.e., Dz ═ Δ @2z, solving equation (4) yields the background baseline curve equation (5):
z＝(W+λDTD)-1Wy (5)
weight coefficient w in equation (5)iAccording to an asymmetrical selection when yi＞ziWhen wiP and yi＜ziWhen wi＝1-p。
Further, the image processing of the lane line feature to obtain a processing result includes: carrying out global self-adaptive binarization on the lane line characteristics to obtain a binary image; filling holes in the binary image to make up for the defects of the lane lines on the binary image; and performing closing operation and opening operation on the image filled with the hollow hole to obtain the processing result.
According to another aspect of the present invention, a lane line detection apparatus based on background reconstruction is provided, and the following technical solution is adopted:
a lane line detection apparatus based on background reconstruction includes: the acquisition module is used for acquiring the lane line characteristics of the road image to be detected based on background reconstruction; the image processing module is used for carrying out image processing on the lane line characteristics to obtain a processing result; and the confirming module is used for carrying out Blob analysis on the processing result so as to confirm the image blocks of the lane lines on the road image to be detected.
Further, the obtaining module comprises: the median filtering processing module is used for carrying out median filtering processing on the interesting region of the road image to be detected to obtain a median filtering image keeping effective edge information; the background modeling processing module is used for carrying out background modeling processing on the median filtering image to obtain a background image of the median filtering image; and the calculating module is used for subtracting the background image from the median filtering image to obtain the lane line characteristics of the road image to be detected.
Further, the background modeling processing module includes: the projection module is used for projecting the median filtering image to the Y axis to obtain a projection curve of the median filtering image; the correction module is used for carrying out baseline drift correction on the projection curve to obtain a background baseline curve of the median filtering image; and the generating module is used for generating the background image according to the background baseline curve.
The correction module is further configured to: the preset smoothing condition is expressed by minimizing a least square function with a penalty term, and is represented by an equation (1):
where y is a signal of length m, z is another sequence, and λ is a compromise factor;
wherein, Delta2ziRepresents the second derivative at point i, i.e.:
Δ2zi＝(zi-zi-1)-(zi-1-zi-2)＝zi-2zi-1+zi-2 (2)
the asymmetric least square method introduces a weight vector w on the basis of the formula (1), and obtains a minimized formula (3):
deriving equation system (4) from minimization equation (3):
(W+λDTD)z＝Wy (4)
where W is a diagonal matrix of W, i.e., W ═ diag (W), and D is a matrix of second derivatives of z, i.e., Dz ═ Δ @2z, solving equation (4) yields the background baseline curve equation (5):
z＝(W+λDTD)-1Wy (5)
weight coefficient w in equation (5)iAccording to an asymmetrical selection when yi＞ziWhen wiP and yi＜ziWhen wi＝1-p。
Further, the image processing module includes: the binarization module is used for carrying out global self-adaptive binarization on the lane line characteristics to obtain a binary image; the cavity filling module is used for filling cavities in the binary image so as to make up the defects of the lane lines on the binary image; and the operation module is used for performing closed operation and open operation on the image after the cavity filling to obtain the processing result.
According to still another aspect of the present invention, there is provided a driving assistance system, and the following technical solutions are adopted:
a driving assistance system includes the lane line detection device described above.
The method comprises the steps of modeling a background of a road image to be detected, extracting lane line characteristics, and carrying out global self-adaptive binarization on a lane line characteristic map to obtain a binary image; and finally, performing Blob analysis on the binary image, and determining the image blocks which accord with the characteristics of the lane line as the lane line. By the technical scheme, the problem of lane line detection is solved quickly and efficiently.
Drawings
In order to more clearly illustrate the embodiments of the present application or technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings can be obtained by those skilled in the art according to the drawings.
Fig. 1 is a flowchart of a lane line detection method based on background reconstruction according to an embodiment of the present invention;
FIG. 2 is an original image and an ROI of a road image to be detected according to an embodiment of the present invention;
fig. 3 is an image obtained by median filtering a ROI region of a road image to be detected according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating a background curve and a baseline thereof according to an embodiment of the present invention;
FIG. 5 is a background image generated from a background baseline according to an embodiment of the present invention;
fig. 6 is a flowchart illustrating a lane line detection method according to an embodiment of the present invention;
FIG. 7 is a diagram illustrating a median filtered image minus a background image according to an embodiment of the present invention;
FIG. 8 is a binarized representation of a lane line feature map according to an embodiment of the present invention;
FIG. 9 is an image of the binarized image after hole filling, closing operation and opening operation according to the embodiment of the present invention;
FIG. 10 is a schematic diagram of a binary image after performing Blob analysis according to an embodiment of the present invention;
fig. 11 is a schematic structural diagram of a lane line detection apparatus based on background reconstruction according to an embodiment of the present invention.
Detailed Description
The embodiments of the invention will be described in detail below with reference to the drawings, but the invention can be implemented in many different ways as defined and covered by the claims.
Fig. 1 is a flowchart of a lane line detection method based on background reconstruction according to an embodiment of the present invention.
Referring to fig. 1, a lane line detection method based on background reconstruction includes:
s101: acquiring lane line characteristics of a road image to be detected based on background reconstruction;
s103: carrying out image processing on the lane line characteristics to obtain a processing result;
s105: and performing Blob analysis on the processing result to confirm the image blocks of the lane lines on the road image to be detected.
In the technical scheme of this embodiment, in step S101, the road image to be detected is a road image obtained by image capturing devices such as a monocular camera and a binocular camera. First, a region of interest, i.e. an ROI region, is set from a road image to be detected, and usually the lower half of the image is taken, as shown in fig. 2, where fig. 2 is an original image and the ROI region of the road image to be detected according to an embodiment of the present invention. The image is median filtered, noise is removed, and simultaneously, the image edge information is effectively maintained, the template generally takes a window of 3 × 3 or 5 × 5, as shown in fig. 3, fig. 3 is the image after median filtering of the ROI region of the road image to be detected according to the embodiment of the present invention. And then, obtaining the lane line characteristics of the road image to be detected by reconstructing the background of the template region. In step S103, performing image processing on the lane line feature to obtain a processing result; specifically, global self-adaptive binarization is carried out on the lane line characteristic image to obtain a binary image; and further, filling holes in the binary image to make up for the defects of damaged and worn lane lines, and performing closing operation and opening operation on the image after hole filling to remove some burrs and miscellaneous blocks. In step S105, Blob analysis is performed on the processing results obtained in step S103 to determine which are to confirm the image blocks of the lane lines on the road image to be detected.
In the technical solution of the above embodiment, the lane line feature image is a target to be acquired, and in order to achieve the purpose of acquiring the lane line feature image, a background model needs to be constructed first, and then the background model is compared with the current image, so as to obtain a foreground target. The method comprises the steps of modeling a background of a road image to be detected, extracting lane line characteristics, and carrying out global self-adaptive binarization on a lane line characteristic map to obtain a binary image; and finally, performing Blob analysis on the binary image, and determining the image blocks which accord with the characteristics of the lane line as the lane line. By the technical scheme, the problem of lane line detection is solved quickly and efficiently.
As a preferred embodiment, the obtaining of the lane line feature of the road image to be detected based on the background reconstruction includes:
carrying out median filtering processing on the region of interest of the road image to be detected to obtain a median filtering image keeping effective edge information; carrying out background modeling processing on the median filtering image to obtain a background image of the median filtering image; and subtracting the background image from the median filtering image to obtain the lane line characteristics of the road image to be detected.
In the technical solution of this embodiment, performing median filtering on the region of interest of the road image to be detected is a common technical means in the art, and redundant description is not repeated here. In the process of median filtering, median filtering is carried out on the image, the edge information of the image is effectively kept while noise is removed, and a median filtering template generally takes 3 × 3 or 5 × 5. Modeling the median filtering image to obtain a background image, specifically:
projecting the median filtering image to the Y axis to obtain a projection curve of the median filtering image; the baseline shift correction is performed on the projection curve, and algorithms including, but not limited to, Asymmetric Least Squares (AsLS), SNIP, and the like are used to obtain a background baseline curve of the median filtered image, as shown in fig. 4, where fig. 4 is a schematic diagram of the background curve and a baseline thereof according to an embodiment of the present invention. The background image is further generated according to the background baseline curve in fig. 4, the background image is shown in fig. 5, and fig. 5 is the background image generated by the background baseline according to the embodiment of the present invention.
As a preferred embodiment, the specific operation manner of performing the drift correction on the baseline by using the asymmetric least square algorithm is as follows:
the Asymmetric Least Squares (AsLS) fitting idea was derived from the Whittaker smoother. The Whittaker smoother is based on the idea that: let y be a signal of length m and z be another sequence, and satisfy the following two preset smoothing conditions: (1) smoothing; (2) faithful to y. These two conditions can be represented by minimizing a least squares function with a penalty term:
the first term in the parentheses of the above formula represents the approximation degree of the fitting function and the original data, the second term is used for ensuring the smoothness of the fitting function, and lambda is a compromise factor and plays a role in balancing the approximation degree and the smoothness. Delta2ziRepresents the second derivative at point i, i.e.:
Δ2zi＝(zi-zi-1)-(zi-1-zi-2)＝zi-2zi-1+zi-2 (2)
the asymmetric least square method introduces a weight vector w on the basis of the formula (1):
minimizing equation (3) can derive the following system of equations:
(W+λDTD)z＝Wy (4)
where W is a diagonal matrix of W, i.e., W ═ diag (W), and D is a matrix of second derivatives of z, i.e., Dz ═ Δ @2z. Solving equation (4) yields the estimated baseline:
z＝(W+λDTD)-1Wy (5)
weight coefficient w in equation (5)iAccording to an asymmetrical selection when yi＞ziWhen wiP and yi＜ziWhen wi1-p. Generally, p is a very small value, and the value range of p is 0.001-0.1; thus for yi＞ziThe weight of which is small, and for yi＜ziThe point of (2) is weighted very much, which is the basis of the "asymmetric least squares" expression. λ is typically a large value in the range of 102～109. Since the optimization goal is a convex function, the iterative process converges quickly. In practical solutions, convergence can be achieved by generally 5 to 10 iterations.
Fig. 6 is a specific flowchart of a lane line detection method according to an embodiment of the present invention, where fig. 6 shows a specific embodiment of the lane line detection method, and with reference to fig. 6, the lane line detection method specifically includes:
step 1: collecting a road image;
step 2: setting a region of interest;
and step 3: median filtering;
and 4, step 4: modeling a background;
and 5: subtracting the background image from the median filtered image;
step 6: carrying out global self-adaptive binarization to obtain a binary image;
and 7: filling a cavity;
and 8: performing closed operation and open operation to remove the parasitic lines;
and step 9: and performing Blog analysis to determine the final lane line image block.
More specifically, the road image is collected in the step 1, namely the road image to be detected is acquired; setting an interested region in step 2, selecting the lower half part of the road image to be detected as the interested region, specifically describing the median filtering and background modeling in steps 3 and 4 in the above embodiment, subtracting the background image obtained by background modeling from the median filtered image in step 5, as shown in fig. 7, and fig. 7 is a schematic diagram of the median filtered image minus the background image in the embodiment of the present invention. In steps 6 to 8, image processing is performed on the image obtained by subtracting the background image from the median filtered image, specifically including performing global adaptive binarization on the lane line feature image to obtain a binary image, and referring to fig. 8, fig. 8 is a flowchart of a method for removing black noise from the road image to be detected according to the embodiment of the present invention. After the binary image is obtained, the binary image is subjected to hole filling to make up for damages and the defects of the worn lane lines, and the image after the hole filling is subjected to closing operation and opening operation to remove some burrs and miscellaneous blocks, specifically referring to fig. 9, which is an image of the binary image after the hole filling, closing operation and opening operation in the embodiment of the present invention. Finally, Blob analysis is performed on the image obtained after the processing, and it is determined which image blocks are image blocks of the lane line, referring to fig. 10, fig. 10 is a schematic diagram of a binary image subjected to Blob analysis according to an embodiment of the present invention.
The embodiment provides a specific implementation scheme of lane line detection based on a background modeling scheme, then a background image is subtracted from a median filtered image to extract features of a lane line, a series of image processing is performed on the lane line, and image blocks which accord with the features of the lane line are determined as the lane line. By the technical scheme, the problem of lane line detection is solved quickly and efficiently.
Fig. 11 is a schematic structural diagram of a lane line detection apparatus based on background reconstruction according to an embodiment of the present invention.
Referring to fig. 11, a lane line detection apparatus based on background reconstruction includes: the acquisition module 20 is used for acquiring lane line characteristics of the road image to be detected based on background reconstruction; the processing module 40 is configured to perform image processing on the lane line feature to obtain a processing result; and the confirming module 60 is configured to perform Blob analysis on the processing result to confirm the lane line image blocks on the road image to be detected.
Preferably, the obtaining module 20 includes: a median filtering processing module (not shown in the figure) for performing median filtering processing on the region of interest of the road image to be detected to obtain a median filtering image keeping effective edge information; a background modeling processing module (not shown in the figure) for performing background modeling processing on the median filtering image to obtain the background image; and the calculating module (not shown) is used for subtracting the background image from the median filtering image to obtain the lane line characteristics of the road image to be detected.
Preferably, the background modeling processing module includes: a projection module (not shown in the figure) for projecting the median filtered image to the Y axis to obtain a projection curve of the median filtered image; a correction module (not shown) for performing baseline wander correction on the projection curve to obtain a background baseline curve of the median filtered image; a generating module (not shown in the figure) for generating the background image according to the background baseline curve.
Preferably, the correction module is further configured to: the preset smoothing condition is expressed by minimizing a least square function with a penalty term, and is represented by an equation (1):
where y is a signal of length m, z is another sequence, and λ is a compromise factor;
wherein, Delta2ziRepresents the second derivative at point i, i.e.:
Δ2zi＝(zi-zi-1)-(zi-1-zi-2)＝zi-2zi-1+zi-2 (2)
the asymmetric least square method introduces a weight vector w on the basis of the formula (1), and obtains a minimized formula (3):
deriving equation system (4) from minimization equation (3):
(W+λDTD)z＝Wy (4)
where W is a diagonal matrix of W, i.e., W ═ diag (W), and D is a matrix of second derivatives of z, i.e., Dz ═ Δ @2z, solving equation (4) yields the background baseline curve equation (5):
z＝(W+λDTD)-1Wy (5)
weight in formula (5)Number wiAccording to an asymmetrical selection when yi＞ziWhen wiP and yi＜ziWhen wi＝1-p。
Preferably, the processing module 40 comprises: a binarization module (not shown in the figure) for performing global self-adaptive binarization on the lane line characteristics to obtain a binary image; a hole filling module (not shown in the figure) for filling holes in the binary image to make up for the defects of the lane lines on the binary image; and the operation module (not shown) is used for performing closing operation and opening operation on the image after the cavity is filled to obtain the processing result.
The driving assistance system provided by the invention comprises the lane line detection device.
The method comprises the steps of modeling a background of a road image to be detected, extracting lane line characteristics, and carrying out global self-adaptive binarization on a lane line characteristic map to obtain a binary image; and finally, performing Blob analysis on the binary image, and determining the image blocks which accord with the characteristics of the lane line as the lane line. By the technical scheme, the problem of lane line detection is solved quickly and efficiently.
While certain exemplary embodiments of the present invention have been described above by way of illustration only, it will be apparent to those of ordinary skill in the art that the described embodiments may be modified in various different ways without departing from the spirit and scope of the invention. Accordingly, the drawings and description are illustrative in nature and should not be construed as limiting the scope of the invention.
Claims (8)
1. A lane line detection method based on background reconstruction is characterized by comprising the following steps:
acquiring lane line characteristics of a road image to be detected based on background reconstruction;
carrying out image processing on the lane line characteristics to obtain a processing result;
performing Blob analysis on the processing result to confirm the image blocks of the lane lines on the road image to be detected;
the method for acquiring the lane line characteristics of the road image to be detected based on background reconstruction comprises the following steps:
carrying out median filtering processing on the region of interest of the road image to be detected to obtain a median filtering image keeping effective edge information;
carrying out background modeling processing on the median filtering image to obtain a background image;
and subtracting the background image from the median filtering image to obtain the lane line characteristics of the road image to be detected.
2. The lane line detection method of claim 1, wherein performing background modeling processing on the median filtered image to obtain a background image comprises:
projecting the median filtering image to the Y axis to obtain a projection curve of the median filtering image;
performing baseline drift correction on the projection curve to obtain a background baseline curve of the median filtering image;
and generating the background image according to the background baseline curve.
3. The lane line detection method of claim 2, wherein the performing baseline shift correction on the projection curve to obtain a background baseline curve of the median filtered image comprises:
the preset smoothing condition is expressed by minimizing a least square function with a penalty term, and is represented by an equation (1):
where y is a signal of length m, z is another sequence, and λ is a compromise factor;
wherein, Delta2ziRepresents the second derivative at point i, i.e.:
Δ2zi＝(zi-zi-1)-(zi-1-zi-2)＝zi-2zi-1+zi-2 (2)
the asymmetric least square method introduces a weight vector w on the basis of the formula (1), and obtains a minimized formula (3):
deriving equation system (4) from minimization equation (3):
(W+λDTD)z＝Wy (4)
where W is a diagonal matrix of W, i.e., W ═ diag (W), and D is a matrix of second derivatives of z, i.e., Dz ═ Δ @2z, solving equation (4) yields equation (5) for the background baseline curve:
z＝(W+λDTD)-1Wy (5)
weight coefficient w in equation (5)iAccording to an asymmetrical selection when yi＞ziWhen wiP and yi＜ziWhen wi＝1-p。
4. The method according to claim 1, wherein the image processing of the lane line feature to obtain a processing result comprises:
carrying out global self-adaptive binarization on the lane line characteristics to obtain a binary image;
filling holes in the binary image to make up for the defects of the lane lines on the binary image;
and performing closing operation and opening operation on the image filled with the hollow hole to obtain the processing result.
5. A lane line detection device based on background reconstruction is characterized by comprising:
the acquisition module is used for acquiring the lane line characteristics of the road image to be detected based on background reconstruction;
the processing module is used for carrying out image processing on the lane line characteristics to obtain a processing result;
the confirming module is used for carrying out Blob analysis on the processing result so as to confirm the lane line image blocks on the road image to be detected; the acquisition module includes:
the median filtering processing module is used for carrying out median filtering processing on the interesting region of the road image to be detected to obtain a median filtering image keeping effective edge information;
the background modeling processing module is used for carrying out background modeling processing on the median filtering image to obtain the background image;
and the computing module is used for subtracting the background image from the median filtering image to obtain the lane line characteristics of the road image to be detected.
6. The lane line detection apparatus of claim 5, wherein the background modeling processing module comprises:
the projection module is used for projecting the median filtering image to the Y axis to obtain a projection curve of the median filtering image;
the correction module is used for carrying out baseline drift correction on the projection curve to obtain a background baseline curve of the median filtering image;
and the generating module is used for generating the background image according to the background baseline curve.
7. The lane line detection apparatus of claim 6, wherein the correction module is further to:
the preset smoothing condition is expressed by minimizing a least square function with a penalty term, and is represented by an equation (1):
where y is a signal of length m, z is another sequence, and λ is a compromise factor;
wherein, Delta2ziRepresents the second derivative at point i, i.e.:
Δ2zi＝(zi-zi-1)-(zi-1-zi-2)＝zi-2zi-1+zi-2 (2)
the asymmetric least square method introduces a weight vector w on the basis of the formula (1), and obtains a minimized formula (3):
deriving equation system (4) from minimization equation (3):
(W+λDTD)z＝Wy (4)
where W is a diagonal matrix of W, i.e., W ═ diag (W), and D is a matrix of second derivatives of z, i.e., Dz ═ Δ @2z, solving equation (4) yields the background baseline curve equation (5):
z＝(W+λDTD)-1Wy (5)
weight coefficient w in equation (5)iAccording to an asymmetrical selection when yi＞ziWhen wiP and yi＜ziWhen wi＝1-p。
8. A driving assist system characterized by comprising the lane line detection apparatus according to any one of claims 5 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711284745.5A CN108229327B (en) | 2017-12-07 | 2017-12-07 | Lane line detection method, device and system based on background reconstruction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711284745.5A CN108229327B (en) | 2017-12-07 | 2017-12-07 | Lane line detection method, device and system based on background reconstruction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108229327A CN108229327A (en) | 2018-06-29 |
| CN108229327B true CN108229327B (en) | 2021-05-18 |
Family
ID=62653282
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711284745.5A Active CN108229327B (en) | 2017-12-07 | 2017-12-07 | Lane line detection method, device and system based on background reconstruction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108229327B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108983791A (en) * | 2018-09-10 | 2018-12-11 | 贵州大学 | A kind of intelligence vehicle control |
| CN110163109B (en) * | 2019-04-23 | 2021-09-17 | 浙江大华技术股份有限公司 | Lane line marking method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000011133A (en) * | 1998-06-19 | 2000-01-14 | Toshiba Corp | Device and method for detecting moving object |
| CN101916516A (en) * | 2010-08-13 | 2010-12-15 | 深圳市豪恩汽车电子装备股份有限公司 | Lane deviation recognition method |
| CN103383733A (en) * | 2013-05-16 | 2013-11-06 | 浙江智尔信息技术有限公司 | Lane video detection method based on half-machine study |
| CN103996030A (en) * | 2014-05-23 | 2014-08-20 | 奇瑞汽车股份有限公司 | Lane line detection method |
| CN105654064A (en) * | 2016-01-25 | 2016-06-08 | 北京中科慧眼科技有限公司 | Lane line detection method and device as well as advanced driver assistance system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4416039B2 (en) * | 2008-03-19 | 2010-02-17 | 日本電気株式会社 | Striped pattern detection system, striped pattern detection method, and striped pattern detection program |
-
2017
- 2017-12-07 CN CN201711284745.5A patent/CN108229327B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000011133A (en) * | 1998-06-19 | 2000-01-14 | Toshiba Corp | Device and method for detecting moving object |
| CN101916516A (en) * | 2010-08-13 | 2010-12-15 | 深圳市豪恩汽车电子装备股份有限公司 | Lane deviation recognition method |
| CN103383733A (en) * | 2013-05-16 | 2013-11-06 | 浙江智尔信息技术有限公司 | Lane video detection method based on half-machine study |
| CN103996030A (en) * | 2014-05-23 | 2014-08-20 | 奇瑞汽车股份有限公司 | Lane line detection method |
| CN105654064A (en) * | 2016-01-25 | 2016-06-08 | 北京中科慧眼科技有限公司 | Lane line detection method and device as well as advanced driver assistance system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108229327A (en) | 2018-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108230264B (en) | Single image defogging method based on ResNet neural network | |
| CN108665487B (en) | Transformer substation operation object and target positioning method based on infrared and visible light fusion | |
| CN109859171B (en) | Automatic floor defect detection method based on computer vision and deep learning | |
| CN110349207B (en) | Visual positioning method in complex environment | |
| CN109377555B (en) | Method for extracting and identifying three-dimensional reconstruction target features of foreground visual field of autonomous underwater robot | |
| JP2021528784A (en) | Sky filter method for panoramic images and mobile terminals | |
| CN109685045B (en) | Moving target video tracking method and system | |
| Akarsu et al. | A fast and adaptive road defect detection approach using computer vision with real time implementation | |
| CN111145209A (en) | Medical image segmentation method, device, equipment and storage medium | |
| CN105741319B (en) | Improvement visual background extracting method based on blindly more new strategy and foreground model | |
| CN110555868A (en) | method for detecting small moving target under complex ground background | |
| CN113780110A (en) | Method and device for detecting weak and small targets in image sequence in real time | |
| CN108229327B (en) | Lane line detection method, device and system based on background reconstruction | |
| CN103761768A (en) | Stereo matching method of three-dimensional reconstruction | |
| CN107871315B (en) | Video image motion detection method and device | |
| CN113936210A (en) | Anti-collision method for tower crane | |
| KR102120453B1 (en) | Method and Apparatus for Generating Scene Situation Information of Video Using Differentiation of Image Feature and Supervised Learning | |
| CN116205907A (en) | Decorative plate defect detection method based on machine vision | |
| Radzi et al. | Extraction of moving objects using frame differencing, ghost and shadow removal | |
| JP2020006788A (en) | Construction limit determination device | |
| CN117541652A (en) | Dynamic SLAM method based on depth LK optical flow method and D-PROSAC sampling strategy | |
| Matkan et al. | Spatial analysis for outlier removal from LiDAR data | |
| Pavitha et al. | Perception system in autonomous vehicle: A study on contemporary and forthcoming technologies for object detection in autonomous vehicles | |
| CN112084957A (en) | Mobile target retention detection method and system | |
| Sun et al. | Spot segmentation and verification based on improve marker controlled watershed transform |
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 |