Nothing Special   »   [go: up one dir, main page]

CN113420587B - A vehicle active collision avoidance method based on road pothole detection - Google Patents

A vehicle active collision avoidance method based on road pothole detection Download PDF

Info

Publication number
CN113420587B
CN113420587B CN202110508060.4A CN202110508060A CN113420587B CN 113420587 B CN113420587 B CN 113420587B CN 202110508060 A CN202110508060 A CN 202110508060A CN 113420587 B CN113420587 B CN 113420587B
Authority
CN
China
Prior art keywords
pothole
image
boundary
road
point
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
Application number
CN202110508060.4A
Other languages
Chinese (zh)
Other versions
CN113420587A (en
Inventor
袁朝春
吴辛恺
何友国
张厚忠
孙晓强
陈龙
蔡英凤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangsu University
Original Assignee
Jiangsu University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jiangsu University filed Critical Jiangsu University
Priority to CN202110508060.4A priority Critical patent/CN113420587B/en
Publication of CN113420587A publication Critical patent/CN113420587A/en
Application granted granted Critical
Publication of CN113420587B publication Critical patent/CN113420587B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06T7/11—Region-based segmentation
    • 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
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30248—Vehicle exterior or interior
    • G06T2207/30252—Vehicle exterior; Vicinity of vehicle
    • G06T2207/30256—Lane; Road marking
    • 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

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • Traffic Control Systems (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a vehicle active collision avoidance method based on pavement pit detection, which utilizes multi-sensor fusion to detect a front pavement pit, uses algorithms such as graying and multi-scale Retinex to pretreat the front pavement pit, improves the image processing efficiency, utilizes Hough transformation to remove road marking interference in an image, accurately calculates the pit area based on a threshold segmentation theory and a chain code method, synthesizes gradient obstacle and depth step obstacle judgment feasibility, synthesizes the above various information, can more accurately calculate the pit passing limit speed, and reduces traffic accidents caused by the vehicle passing through the pit.

Description

Active collision avoidance method for vehicle based on pavement pit detection
Technical Field
The invention belongs to the technical field of driving safety, and particularly relates to an active collision avoidance method based on pavement pit detection.
Background
According to the statistical data of the transportation department, the total mileage of the national highway is 501.25 ten thousand kilometers by 2019, and is increased by 54.86 ten thousand kilometers compared with 2014. The highway density 52.21 km/hundred square kilometers is increased by 5.71 km/hundred square kilometers compared with 2014, and the situation of stable growth is presented. And along with the increase of the service life of the road and the increase of traffic volume, the pressure of passenger traffic borne by the road is increased, the road structure is damaged, and a large number of road diseases are generated, wherein the most common road pits are the road pits which are the most easy to influence the driving safety. The specific size information of the road pits is timely and accurately obtained, dangerous working conditions are avoided in advance, and the method has important practical significance for driving safety, but the influence of the road pits on the driving safety of the vehicle is not considered in the existing active collision avoidance algorithm applied to the vehicle.
Disclosure of Invention
The invention discloses an active collision avoidance method based on pavement pit detection, which can accurately predict the current pit passing limit speed of a vehicle and avoid the influence of the pavement pit on the driving safety in the driving process of the vehicle. The specific scheme is as follows:
an active collision avoidance method based on pavement pit detection comprises the following steps:
step 1, pit information is obtained;
Step 2, graying and enhancing the pit image, and removing a road marking area;
Step 3, extracting outline features of the pits of the pavement;
Tracking the pit outline obtained in the step 3 and calculating the pit area;
Step 5, judging the trafficability by combining the information obtained in the step 1 and the pit boundary gradient;
and 6, obtaining the pit passing limit vehicle speed by integrating the information obtained in the steps 1,4 and 5.
Further, in the step 1, pit depth is obtained by utilizing a radar, and a pavement pit image is obtained by utilizing a camera;
further, the weighted average method for performing the graying processing on the image in the step 2 specifically includes:
Gray(i,j)=0.299*R(i,j)+0.578*G(i,j)+0.114*B(i,j)
In the formula, (i, j) represents coordinates of the pixel point, and Gray (i, j) represents a pixel value of the image after gradation.
Further, in the step 2, the enhancement processing of the image adopts an MSR multi-scale algorithm, which specifically comprises the following steps:
s(i,j)=l(i,j)r(i,j)
Where (i, j) denotes the pixel coordinates, l (i, j) denotes the incident illumination image, r (i, j) denotes the reflected illumination image, and s (i, j) denotes the product of the incident illumination and the reflected illumination image.
The luminance component is estimated from the known image s (i, j) using a gaussian convolution function F (i, j), specifically as follows:
l(i,j)=F(i,j)*s(i,j)
wherein, is convolution operation, lambda is normalized coefficient, sigma is scale parameter of Gaussian function.
The MSR multi-scale algorithm is utilized for linear weighted summation, and the specific formula is as follows:
wherein N is the number of Gaussian kernels, usually 3, and
Further, the specific method for extracting the road marking area in the step 2 is as follows:
Based on a Canny edge detection operator, carrying out Hough transformation on a road surface image containing the road marking, and extracting the linear edge of the road marking:
ρ=xcosθ+ysinθ
Wherein ρ is the distance from the straight line to the origin of coordinates, the range ρ epsilon (0, r), r is the length of the diagonal line of the image, θ is the angle between the straight line and the x-axis, and the range θ epsilon (0, 180 °);
And calculating the gray average value of each region formed by the extracted marks, dividing the road mark region by setting a threshold value, and removing the extracted road mark region from the original image to obtain a new road detection region.
Further, the specific method for extracting the outline of the pavement pit in the step 3 comprises the following steps:
And adopting a Canny operator to realize contour extraction. Firstly, calculating an optimal segmentation threshold T k of the image by utilizing calcHist functions and min-MaxLoc functions, then performing binarization processing on the image, and finally obtaining the edge profile of the pit by adopting a median filtering algorithm;
further, the specific method for tracking the pit outline and calculating the pit area in the step 4 is as follows:
And carrying out similarity tracking on the pit boundary by adopting an 8-direction chain code method. Firstly, counting the total number A 1 of pixels at the boundary of the pit of the pavement, then carrying out vectorization setting on boundary points of the pit, then counting the total number A 2 of pixels in the boundary of the pit, and finally calculating the area s of the pit of the pavement;
further, the trafficability in the step5 is comprehensively judged by gradient obstacle and depth step obstacle function;
Further, the pit passing limit vehicle speed v l in step 6 is composed of the pit area s, the depth d, the trafficability f ΔT, the current vehicle speed v, the vertical acceleration absolute value maximum value |a z-max |, and the weighted acceleration root mean square value Determining;
the invention has the beneficial effects that:
The active safety obstacle avoidance method for the automobile based on the road surface pit detection detects the road surface pit by utilizing multi-sensor fusion, carries out pretreatment by utilizing algorithms such as graying and multi-scale Retinex, improves the image processing efficiency, removes road marking interference in the image by utilizing Hough transformation, accurately calculates the pit area based on a threshold segmentation theory and a chain code method, synthesizes gradient obstacle and depth step obstacle to judge trafficability, and can more accurately calculate the pit passing limit speed by synthesizing the various information, thereby reducing traffic accidents caused by the vehicle passing through the pit.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a geometric model of pit depth measurement;
FIG. 3 is a schematic diagram of a grade barrier.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
The invention is described in further detail below with reference to the drawings and examples.
As shown in fig. 1, the invention provides an active collision avoidance method for an automobile based on pavement pit detection, which comprises the following steps:
step 1, pit information is obtained;
The single-point ranging laser radar is arranged on a front bumper of the vehicle, the pitch angle is beta, and the single-point ranging laser radar is used for calculating the depth d of the pit, as shown in fig. 2, and the specific calculation formula is as follows:
Wherein, Indicating the distance at which the laser spot reaches the deepest part of the pit,The distance between the laser point of the single-point ranging laser radar and the road surface is a fixed value.
Step 2, carrying out graying and enhancement treatment on the image, and extracting and removing a road marking area;
2.1 The image is subjected to graying treatment by adopting a weighted average method, and the specific formula is as follows:
Gray(i,j)=0.299*R(i,j)+0.578*G(i,j)+0.114*B(i,j)
where (i, j) represents coordinates of the pixel point, and Gray (i, j) represents a pixel value of the image after graying.
2.2 On the basis of Retinex theory, using an MSR multi-scale algorithm to carry out weighted summation on different scales, wherein the specific formula is as follows:
s(i,j)=l(i,j)r(i,j)
where (i, j) denotes the pixel coordinates, l (i, j) denotes the incident illumination image, r (i, j) denotes the reflected illumination image, and s (i, j) denotes the product of the incident illumination and the reflected illumination image.
The luminance component is estimated from the known image s (i, j) using a gaussian convolution function F (i, j), specifically as follows:
l(i,j)=F(i,j)*s(i,j)
wherein, is convolution operation, lambda is normalized coefficient, sigma is scale parameter of Gaussian function.
The multi-scale Retinex algorithm is utilized to carry out weighted summation on different scales, and the specific formula is as follows:
Wherein r MSR (i, j) represents a reflected illumination image calculated by an MSR algorithm, F n (i, j) represents a Gaussian convolution function, N is the number of Gaussian kernels, and is always 3
2.3 Based on Canny edge detection operator, carrying out Hough transformation linear detection on the image, extracting a road marking area and removing the road marking area from the original image, wherein the specific method comprises the following steps:
ρ=xcosθ+ysinθ
Wherein ρ is the distance from the straight line to the origin of coordinates, the range is ρ ε (0, r), r is the diagonal length of the image, θ is the angle between the straight line and the x-axis, and the range is θ ε (0, 180 °).
And calculating the gray average value of each region formed by the extracted marked lines. Setting a threshold T, if the gray level mean value difference between the areas is smaller than the threshold T, defining different areas as the same area, and accurately dividing the road marking area according to the characteristic that the gray level value of the road marking area is higher than that of a common road surface. And then subtracting the extracted road marking area from the original image to obtain a new road detection area.
Step 3, extracting outline features of the pits of the pavement;
the contour extraction of the pavement pits is realized by adopting a Canny operator, and the specific steps are as follows:
3.1 Using calcHist functions to obtain a gray level histogram corresponding to the image and determining a gray level range where two peaks are located, wherein the specific formula is as follows:
Where l represents the pixel mean value of the whole image, and m and n are the length and width of the pixels of the image, respectively.
3.2 Using the min-MaxLoc function to calculate the pixel maximum and the gray values t l and t r in the left and right side peak ranges, respectively. The code is as follows:
void minMaxLoc(InputArray src,CV_OUT double*minVal,CV_OUT double*maxVal＝0,CV_OUT Point*minLoc＝0,CV_OUT Point*maxLoc＝0,InputArray mask＝noArray());
3.3 First, an initial segmentation threshold T 0 of the image is calculated, the image is segmented into a target part and a background part, then the gray average values T l-m and T r-m of the two parts are obtained, and a new threshold T k+1 can be obtained from the gray average values of the two parts. The specific formula is as follows:
where Q (i, j) is the gray scale of any point of the image, N (i, j) is the number of pixels of pixel (i, j), and T k is the threshold.
The final image optimal segmentation threshold T k can be derived from satisfying equation T k＝Tk+1.
3.4 Binarization processing is carried out on the image, so that the gray values of all the pixel points in the pavement pit area are 255 (white), and the gray values of the external pixel points are 0 (black).
3.5 Using a median filtering algorithm to obtain the edge profile of the pit:
yi＝med{fi-n,...,fi-1,fi,fi+1,...,fi+n}
Where f i is the median value of the order from small to large and replaces the gray value of the specified point.
Tracking the pit outline obtained in the step 3 and calculating the pit area;
and carrying out similarity tracking on pit boundaries by adopting an 8-direction chain code method, wherein the chain code of the curve is 57670013234.
4.1 Dividing the binarized pavement pit image into a plurality of columns, tracking the pit edge from the upper left corner in reverse time in order from small to large, and obtaining the maximum number at the tail end as the total number A 1 of boundary pixels.
4.2 Defining a vector link from the upper boundary point (N-1) to the current boundary point (N) as a front vector, a vector link code from the current boundary point (N) to the lower boundary point (N+1) as a rear vector, and marking the vector direction and the vector value by referring to the 8-direction link code, wherein the center pixel point of the link code is the current boundary point.
4.3 The pixel points in the boundary are sequentially subjected to vector analysis in the order from small to large, and the front vector and the rear vector of each boundary point are recorded. If the pixel on the right side of the current boundary point is the boundary inner point, the number of pixels between the current pixel point and the adjacent next boundary point is calculated, the expression is N i+1-Ni -1, and the sum of the number of pixels is the total number of pixels in the boundary A 2. Where N i is the column value of the current boundary point and N i+1 is the column value of the next boundary point. And if the judgment condition shown in the formula is met, defining the point as a boundary inner point:
bv+.8 with fv=5, or fv <3 and |fv-bv| >4, or fv >5 and |fv-bv| <4
Where fv is the front vector and bv is the rear vector.
4.4 Road surface pit area s is obtained by A 1,A2, u:
Wherein A 1 is the total number of pixels at the boundary of the pavement pit, A 2 is the total number of pixels in the boundary of the pavement pit, and u is the nominal pixel equivalent.
Step 5, judging the trafficability by combining the information obtained in the step 1 and the pit boundary gradient;
Road hole boundary gradient β exceeding threshold β max (β max in the present embodiment takes 0.236) is regarded as a gradient obstacle, and as shown in fig. 3, road hole boundary gradient β is obtained:
the barrier function is as follows:
wherein h z is the vehicle ground clearance and b is the vehicle wheelbase.
The depth d of the pavement pit exceeds the ground clearance h z of the automobile to be regarded as a depth step obstacle, the automobile cannot pass, and the obstacle function is as follows:
Comprehensive gradient boundary obstacle and depth step obstacle can judge vehicle trafficability:
fΔ＝max(fΔ(β),fΔ(d))
If the passability influence factor f Δ is larger than 1, the situation that the boundary obstacle of the pit on the road exceeds the limit working condition is indicated, the pit road is defined as a non-passable area, and otherwise, the road can pass through.
Wherein f ΔT is a trafficability determination factor.
Step 6, obtaining the pit passing limit speed by integrating the information obtained in the steps 1,4 and 5;
Fitting vehicle vertical acceleration maximum value |a z-max | and weighted acceleration root mean square value using fifth order polynomials
|az-max|＝0.0064s5-0.0568s4+1560.5d4+0.1895s3+851.58d3-0.0001v3-0.2918s2-243.30d2+0.0033v2+0.2036s+19.572d-0.0441v
The pit passing limit vehicle speed v l is composed of pit area s, depth d, trafficability f ΔT, current vehicle speed v, vertical acceleration absolute value maximum value |a z-max | and weighted acceleration root mean square valueAnd (3) determining:
Firstly, judging whether the road surface of the pit can pass through by using the trafficability judging factor f ΔT, and if the road surface of the pit can pass through, calculating the range of the pit passing limit speed v l by using the known pit area s and the known pit depth d, wherein the specific method is as follows:
Wherein a z-limit is the maximum absolute vertical acceleration of the vehicle, Is the maximum weighted acceleration.
The above list of detailed descriptions is only specific to practical embodiments of the present invention, and they are not intended to limit the scope of the present invention, and all equivalent manners or modifications that do not depart from the technical scope of the present invention should be included in the scope of the present invention.

Claims (2)

1.一种基于路面坑洞检测的车辆主动避撞方法，其特征在于，包括如下步骤：1. A vehicle active collision avoidance method based on road pothole detection, characterized in that it comprises the following steps: 步骤1：获取坑洞信息；Step 1: Get pothole information; 步骤2：对坑洞图像进行灰度化和增强处理，提取并去除道路标线区域；Step 2: Grayscale and enhance the pothole image, extract and remove the road marking area; 步骤3：提取路面坑洞轮廓特征；Step 3: Extract the contour features of road potholes; 步骤4：跟踪步骤3得到的坑洞轮廓并计算坑洞面积；Step 4: Track the pothole contour obtained in step 3 and calculate the pothole area; 步骤5：判断可通行性；Step 5: Determine feasibility; 步骤6：根据步骤1、4、5得到的信息得出过坑极限车速；Step 6: Determine the maximum speed for crossing a pit according to the information obtained in steps 1, 4, and 5; 步骤2中对图像进行灰度化处理采用加权平均法，具体为：In step 2, the image is grayed using a weighted average method, specifically: Gray(i,j)＝0.299*R(i,j)+0.578*G(i,j)+0.114*B(i,j)Gray(i,j)＝0.299*R(i,j)+0.578*G(i,j)+0.114*B(i,j) 式中，(i,j)表示像素点的坐标，Gray(i,j)表示灰度化后图像的像素值；In the formula, (i, j) represents the coordinates of the pixel point, and Gray(i, j) represents the pixel value of the grayscale image; 步骤2中对图像进行增强处理采用MSR多尺度算法，具体为：In step 2, the image is enhanced using the MSR multi-scale algorithm, specifically: 设s(i,j)＝l(i,j)r(i,j)Let s(i,j)＝l(i,j)r(i,j) 式中，(i,j)表示像素坐标，l(i,j)表示入射光照图像，r(i,j)表示反射光照图像，s(i,j)表示入射光照和反射光照图像乘积；Where (i, j) represents the pixel coordinates, l(i, j) represents the incident illumination image, r(i, j) represents the reflected illumination image, and s(i, j) represents the product of the incident illumination and reflected illumination images; 利用高斯卷积函数F(i,j)从已知图像s(i,j)中估算亮度分量，具体公式如下：The Gaussian convolution function F(i,j) is used to estimate the brightness component from the known image s(i,j). The specific formula is as follows: l(i,j)＝F(i,j)*s(i,j)l(i,j)＝F(i,j)*s(i,j) 其中，*为卷积运算，λ为归一化系数，σ为高斯函数的尺度参数；Among them, * is the convolution operation, λ is the normalization coefficient, and σ is the scale parameter of the Gaussian function; 利用MSR多尺度算法进行线性加权求和，具体公式为：The MSR multi-scale algorithm is used for linear weighted summation. The specific formula is: 其中，rMSR(i,j)表示通过MSR算法计算得到的反射光照图像，Fn(i,j)表示高斯卷积函数，N为高斯核个数，取3，且 Where r MSR (i, j) represents the reflected illumination image calculated by the MSR algorithm, F n (i, j) represents the Gaussian convolution function, N is the number of Gaussian kernels, which is 3, and 步骤2中的提取并去除道路标线区域的具体方法为：The specific method for extracting and removing the road marking area in step 2 is: 基于Canny边缘检测算子对包含道路标线的路面图像进行Hough变换，提取道路标线的直线边缘：Based on the Canny edge detection operator, the road surface image containing the road markings is subjected to Hough transform to extract the straight edges of the road markings: ρ＝xcosθ+ysinθρ＝xcosθ+ysinθ 其中，ρ为直线到坐标原点的距离，范围为ρ∈(0,r)，r为图像对角线长度，θ为直线与x轴间的夹角，范围为θ∈(0,180°)；Among them, ρ is the distance from the straight line to the origin of the coordinate system, ranging from ρ∈(0,r), r is the length of the image diagonal, and θ is the angle between the straight line and the x-axis, ranging from θ∈(0,180°); 计算提取到的标线所构成的每个区域的灰度均值，设定阈值T，如果区域间的灰度均值差小于阈值T，则将不同区域定义为同一区域，根据道路标线区域的灰度值高于普通路面的特点，可精确分割出道路标线区域；然后在原始图像减去提取出的道路标线区域得到新的道路检测区域；Calculate the grayscale mean of each area formed by the extracted road markings, set a threshold T, and if the grayscale mean difference between areas is less than the threshold T, define different areas as the same area. Based on the characteristic that the grayscale value of the road marking area is higher than that of the ordinary road surface, the road marking area can be accurately segmented; then subtract the extracted road marking area from the original image to obtain a new road detection area; 步骤3中提取路面坑洞轮廓采用Canny算子实现轮廓提取，具体方法为：In step 3, the Canny operator is used to extract the contour of the pothole on the road. The specific method is as follows: 3.1，利用calcHist函数得到图像对应的灰度直方图并确定两波峰所在的灰度范围，具体为：3.1, use the calcHist function to obtain the grayscale histogram corresponding to the image and determine the grayscale range of the two peaks, specifically: 其中，l表示整张图像的像素均值，m，n分别为图像的像素的长度和宽度；Among them, l represents the pixel mean of the entire image, m and n are the length and width of the image pixels respectively; 3.2，利用min-MaxLoc函数分别计算左侧和右侧波峰范围内的像素最大值及其灰度值tl和tr；3.2, use the min-MaxLoc function to calculate the maximum pixel value and grayscale value t l and t r within the left and right peak ranges respectively; 3.3，计算图像初始分割阈值T0，将图像分割出目标和背景两部分，然后求出两部分的灰度均值tl-m和tr-m，则新阈值Tk+1可由两部分的灰度均值得到，具体公式如下：3.3, calculate the initial image segmentation threshold T 0 , segment the image into two parts, the target and the background, and then find the grayscale mean t lm and t rm of the two parts. The new threshold T k+1 can be obtained from the grayscale mean of the two parts. The specific formula is as follows: 其中，Q(i,j)是图像任意一点的灰度，N(i,j)是像素点(i,j)的像素数，Tk是阈值；Among them, Q(i,j) is the grayscale of any point in the image, N(i,j) is the number of pixels at the pixel point (i,j), and Tk is the threshold; 最终图像最佳分割阈值Tk可由满足等式Tk＝Tk+1得到；The optimal segmentation threshold T k of the final image can be obtained by satisfying the equation T k =T k+1 ; 3.4，对图像进行二值化处理，得到路面坑洞区域内所有像素点的灰度值都为255，外部像素点灰度值都为0；3.4, binarize the image to obtain that the grayscale value of all pixels in the pothole area is 255, and the grayscale value of the external pixels is 0; 3.5，采用中值滤波算法得到坑洞的边缘轮廓：3.5, use the median filter algorithm to get the edge contour of the pothole: yi＝med{fi-n,...,fi-1,fi,fi+1,...,fi+n}y i =med{f in ,...,f i-1 , fi ,f i+1 ,...,f i+n } 其中，fi为从小到大排序的中值并代替指定点的灰度值；Among them, fi is the median value sorted from small to large and replaces the gray value of the specified point; 步骤4中跟踪坑洞轮廓并计算坑洞面积的具体方法如下：The specific method of tracking the pothole contour and calculating the pothole area in step 4 is as follows: 采用8方向链码法对坑洞边界进行相似性跟踪，具体地：The 8-directional chain code method is used to track the similarity of pothole boundaries. Specifically: 4.1，将二值化后的路面坑洞图像分成多列，从左上角开始按逆时针对坑洞边缘以从小到大顺序进行跟踪，结尾处的最大编号即为边界像素总数A1；4.1, the binarized pothole image is divided into multiple columns, and the pothole edges are tracked from the upper left corner in reverse order from small to large, and the largest number at the end is the total number of boundary pixels A 1 ; 4.2，按步骤4.1中的边界点序号，定义上一边界点坑洞边界点矢量化设置：定义上一边界点(N-1)到当前边界点(N)的矢量链路为前位矢量；当前边界点(N)到下一边界点(N+1)的矢量链码为后位矢量，矢量方向和矢量值参照8方向链码标记，且链码中心像素点为当前边界点；4.2, according to the boundary point sequence number in step 4.1, define the vector setting of the previous boundary point pothole boundary point: define the vector link from the previous boundary point (N-1) to the current boundary point (N) as the front vector; the vector chain code from the current boundary point (N) to the next boundary point (N+1) is the back vector, the vector direction and vector value refer to the 8-direction chain code mark, and the center pixel of the chain code is the current boundary point; 4.3，将边界内的像素点按照从小到大的顺序依次进行矢量分析，记录每个边界点的前位矢量和后位矢量；若当前边界点的右侧像素是边界内点，则计算当前像素点与相邻下一个边界点之间的像素个数，表达式为Ni+1-Ni-1，像素个数的总和就是边界内像素总数A2；其中，Ni为当前边界点的列值，Ni+1为下一个边界点的列值；满足公式所示的判定条件则定义该点为边界内点：4.3, perform vector analysis on the pixel points within the boundary in ascending order, and record the previous vector and the next vector of each boundary point; if the right pixel of the current boundary point is a point within the boundary, calculate the number of pixels between the current pixel point and the next adjacent boundary point, the expression is Ni +1 - Ni -1, and the sum of the number of pixels is the total number of pixels within the boundary A2 ; where Ni is the column value of the current boundary point, and Ni+1 is the column value of the next boundary point; if the judgment condition shown in the formula is met, the point is defined as a point within the boundary: bv≠8，同时fv＝5，或fv<3且|fv-bv|>4，或fv>5且|fv-bv|<4bv≠8, and fv＝5, or fv<3 and |fv-bv|>4, or fv>5 and |fv-bv|<4 其中，fv为前位矢量，bv为后位矢量；Among them, fv is the front position vector, bv is the back position vector; 4.4，路面坑洞面积s由A1，A2，u得到：4.4, the pothole area s is obtained from A 1 , A 2 , and u: 其中，A1是路面坑洞边界像素总数，A2是路面坑洞边界内像素总数，u是标定像素当量；Among them, A1 is the total number of pixels at the boundary of the pothole, A2 is the total number of pixels within the boundary of the pothole, and u is the calibration pixel equivalent; 步骤5中的可通行性判断是步骤1得到的信息和坑洞边界坡度判断得出，具体地：The passability judgment in step 5 is obtained based on the information obtained in step 1 and the slope judgment of the pothole boundary. Specifically: 当俯仰角β超过路面坑洞边界坡度的阈值βmax时视为坡度障碍，车辆不可通过，求得路面坑洞边界坡度的阈值βmax如下所示：When the pitch angle β exceeds the threshold β max of the road pothole boundary slope, it is considered a slope obstacle and the vehicle cannot pass. The threshold β max of the road pothole boundary slope is calculated as follows: 障碍函数设计如下：The barrier function is designed as follows: 其中，hz是车辆离地间隙，b是车辆轴距；Where, h z is the vehicle ground clearance, b is the vehicle wheelbase; 路面坑洞深度d超过汽车的离地间隙hz视为深度阶跃障碍，车辆不可通行，障碍函数设计如下：The depth d of the pothole on the road exceeds the ground clearance hz of the car, which is considered a deep step obstacle and the vehicle cannot pass through. The obstacle function is designed as follows: 综合坡度与深度阶跃障碍可判断车辆可通行性：The comprehensive slope and depth of step obstacles can determine vehicle passability: f△＝max(f△(β),f△(d))f △ =max(f △ (β),f △ (d)) 若可通过性影响因子f△大于1则表示此时路面坑洞边界障碍超过极限工况，定义该坑洞路面为不可通过区域，否则可以通过；If the passability influencing factor f △ is greater than 1, it means that the boundary obstacle of the pothole on the road surface exceeds the limit working condition, and the pothole road surface is defined as an impassable area, otherwise it can be passed; 其中，f△T为可通过性判断因子；Among them, f △T is the passability judgment factor; 步骤6的实现包括：The implementation of step 6 includes: 首先利用五次多项式拟合车辆垂向加速度绝对值最大值|az-max|和加权加速度均方根值 First, the fifth-order polynomial is used to fit the maximum absolute value of the vehicle vertical acceleration |a z-max | and the weighted acceleration root mean square value az-max|＝0.0064s5-0.0568s4+1560.5d4+0.1895s3+851.58d3-0.0001v3-0.2918s2 a z-max |＝0.0064s 5 -0.0568s 4 +1560.5d 4 +0.1895s 3 +851.58d 3 -0.0001v 3 -0.2918s 2 -243.30d2+0.0033v2+0.2036s+19.572d-0.0441v-243.30d 2 +0.0033v 2 +0.2036s+19.572d-0.0441v 过坑极限车速vl由坑洞面积s、深度d、可通行性f△T、当前车速v、垂向加速度绝对值最大值|az-max|和加权加速度均方根值等决定：The maximum speed v l for passing a pothole is determined by the pothole area s, depth d, passability f △T , current vehicle speed v, maximum absolute value of vertical acceleration |a z-max | and weighted RMS value of acceleration Waiting for decision: 先利用可通过性判断因子f△T判断是否可以通过坑洞路面，若可以通行，则利用已知坑洞面积s和坑洞深度d计算出过坑极限车速vl的范围，具体方法如下：First, use the passability judgment factor f △T to determine whether the pothole road surface can be passed. If it can be passed, use the known pothole area s and pothole depth d to calculate the range of the maximum speed v l for passing the pothole. The specific method is as follows: 其中，az-limit为车辆最大绝对垂向加速度，为最大加权加速度。Where a z-limit is the maximum absolute vertical acceleration of the vehicle, is the maximum weighted acceleration. 2.根据权利要求1所述的一种基于路面坑洞检测的车辆主动避撞方法，其特征在于，步骤1的实现包括：坑洞图像由安装在车辆上的CCD工业相机获取；坑洞深度由安装在车辆前保险杠上的单点测距激光雷达获取，设俯仰角为β，则坑洞深度d按如下公式计算：2. According to the method of active vehicle collision avoidance based on road pothole detection in claim 1, it is characterized in that the implementation of step 1 includes: the pothole image is obtained by a CCD industrial camera installed on the vehicle; the pothole depth is obtained by a single-point ranging laser radar installed on the front bumper of the vehicle, and the pitch angle is β, and the pothole depth d is calculated according to the following formula: 其中，表示激光点到达坑洞最深处时的距离，表示单点测距激光雷达的激光点到达路面的距离，为定值。in, Indicates the distance when the laser point reaches the deepest part of the pit. It indicates the distance from the laser point of the single-point ranging laser radar to the road surface, which is a constant value.
CN202110508060.4A 2021-05-10 2021-05-10 A vehicle active collision avoidance method based on road pothole detection Active CN113420587B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110508060.4A CN113420587B (en) 2021-05-10 2021-05-10 A vehicle active collision avoidance method based on road pothole detection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110508060.4A CN113420587B (en) 2021-05-10 2021-05-10 A vehicle active collision avoidance method based on road pothole detection

Publications (2)

Publication Number Publication Date
CN113420587A CN113420587A (en) 2021-09-21
CN113420587B true CN113420587B (en) 2024-12-13

Family

ID=77712196

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110508060.4A Active CN113420587B (en) 2021-05-10 2021-05-10 A vehicle active collision avoidance method based on road pothole detection

Country Status (1)

Country Link
CN (1) CN113420587B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114721376A (en) * 2022-03-16 2022-07-08 山西维度空间信息科技有限公司 A road condition identification method, device, equipment and medium for an unmanned minecart
CN114708567B (en) * 2022-06-06 2022-09-06 济南融瓴科技发展有限公司 Road surface hollow detection and avoidance method and system based on binocular camera
CN115268455A (en) * 2022-08-08 2022-11-01 广州高新兴机器人有限公司 Method, system and storage medium for reducing robot obstacle detection false detection rate
CN115131360A (en) * 2022-09-02 2022-09-30 海纳云物联科技有限公司 Road pit detection method and device based on 3D point cloud semantic segmentation
CN118424320A (en) * 2024-07-05 2024-08-02 石家庄铁道大学 A method for road condition recognition and vehicle path planning on unstructured roads

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111414853A (en) * 2020-03-19 2020-07-14 宁波职业技术学院 Automatic driving method, device, computer equipment and storage medium based on machine vision
CN111688690A (en) * 2019-03-12 2020-09-22 现代自动车株式会社 Device and method for avoiding vehicle collapse

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101608924B (en) * 2009-05-20 2011-09-14 电子科技大学 Method for detecting lane lines based on grayscale estimation and cascade Hough transform
CN109919883B (en) * 2018-12-03 2021-06-22 南京三宝科技股份有限公司 Traffic video data acquisition method based on gray level conversion
CN111986120A (en) * 2020-09-15 2020-11-24 天津师范大学 Low-illumination image enhancement optimization method based on frame accumulation and multi-scale Retinex

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111688690A (en) * 2019-03-12 2020-09-22 现代自动车株式会社 Device and method for avoiding vehicle collapse
CN111414853A (en) * 2020-03-19 2020-07-14 宁波职业技术学院 Automatic driving method, device, computer equipment and storage medium based on machine vision

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
4.1.4坑槽面积的计算;张丰梁;《CNKI优秀硕士学位论文全文库》;20160615(第第06期期);30、31页 *

Also Published As

Publication number Publication date
CN113420587A (en) 2021-09-21

Similar Documents

Publication Publication Date Title
CN113420587B (en) A vehicle active collision avoidance method based on road pothole detection
CN113370977B (en) Intelligent vehicle forward collision early warning method and system based on vision
CN103324930B (en) A license plate character segmentation method based on gray histogram binarization
CN110210451B (en) A zebra crossing detection method
CN109800752B (en) Automobile license plate character segmentation and recognition algorithm based on machine vision
CN107066986A (en) A kind of lane line based on monocular vision and preceding object object detecting method
US8385601B2 (en) In-vehicle white line recognition apparatus
CN105701844B (en) Barrier or shadow detection method based on color characteristic
CN106951879A (en) Multi-feature fusion vehicle detection method based on camera and millimeter wave radar
CN110298307B (en) A real-time detection method for abnormal parking based on deep learning
JP4616046B2 (en) VEHICLE IMAGE PROCESSING SYSTEM, VEHICLE IMAGE PROCESSING METHOD, VEHICLE IMAGE PROCESSING PROGRAM, AND VEHICLE
CN110298216A (en) Vehicle deviation warning method based on lane line gradient image adaptive threshold fuzziness
CN104658011A (en) Intelligent transportation moving object detection tracking method
CN118015002B (en) Traffic engineering road condition visual detection method and system
CN110263635B (en) Marker detection and identification method based on structural forest and PCANet
KR101483742B1 (en) Lane Detection method for Advanced Vehicle
CN107480646B (en) A method for detecting abnormal motion in vehicle video based on binocular vision
CN114299247B (en) Road traffic signs and markings rapid detection and troubleshooting methods
CN106778661A (en) A kind of express lane line detecting method based on morphological transformation and adaptive threshold
CN110659552A (en) Tramcar obstacle detection and alarm method
CN113033363A (en) Vehicle dense target detection method based on deep learning
CN106803066A (en) A kind of vehicle yaw angle based on Hough transform determines method
CN115856936A (en) Ground obstacle identification method based on laser radar
Lu et al. Monocular multi-kernel based lane marking detection
CN112802348B (en) A Traffic Flow Counting Method Based on Hybrid Gaussian Model

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