WO2021185332A1 - 一种基于胎厚测量的轮胎异常变形量的检测方法 - Google Patents

一种基于胎厚测量的轮胎异常变形量的检测方法 Download PDF

Info

Publication number
WO2021185332A1
WO2021185332A1 PCT/CN2021/081636 CN2021081636W WO2021185332A1 WO 2021185332 A1 WO2021185332 A1 WO 2021185332A1 CN 2021081636 W CN2021081636 W CN 2021081636W WO 2021185332 A1 WO2021185332 A1 WO 2021185332A1
Authority
WO
WIPO (PCT)
Prior art keywords
tire
hub
area
image
deformation
Prior art date
Application number
PCT/CN2021/081636
Other languages
English (en)
French (fr)
Inventor
肖梅
黄洪滔
明秀玲
张雷
徐婷
Original Assignee
长安大学
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 长安大学 filed Critical 长安大学
Publication of WO2021185332A1 publication Critical patent/WO2021185332A1/zh

Links

Images

Classifications

    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration using local operators
    • G06T5/30—Erosion or dilatation, e.g. thinning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20024—Filtering details
    • G06T2207/20032—Median filtering
    • 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
    • Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation

Definitions

  • the invention relates to the technical field of automobile appearance detection and diagnosis, in particular to a detection method of abnormal tire deformation based on tire thickness measurement.
  • Automobile tires are rubber products with strong elasticity. They are in direct contact with the road surface and are one of the important parts of the automobile. They play a very important role in driving: bearing the weight of the automobile; together with the automobile suspension, they can alleviate the problem of the automobile when driving. Impact, to ensure the comfort and smoothness of the vehicle; to ensure good adhesion between the wheels and the road surface, and to improve the traction, braking and passing properties of the vehicle.
  • the deformation of the tire during driving will change due to the influence of many factors such as tire pressure, road temperature, load, tire parameters, and road smoothness. Excessive or too small deformation will cause many hidden dangers.
  • tire deformation is large when the vehicle is running, the friction coefficient with the road surface will increase, and the fuel consumption will increase, which will cause the steering wheel to be heavy and easy to run. Accelerate the aging of tires, leading to shoulder wear; the friction between the tire and the ground doubles, the tire temperature rises sharply, the tire becomes soft, the strength decreases, and it is easy to cause a tire blowout when driving at high speed.
  • the grip of the tire will be reduced and the friction and adhesion will be reduced, which will affect the braking performance of the vehicle; the vibration of the vehicle body will affect the comfort of the driver and passengers, and indirectly affect the life of other parts; accelerate the tire
  • the local wear of the central pattern of the tread reduces the tire life. Excessive sidewall tension will cause the carcass to decrease in elasticity, increase the load on the vehicle, and increase the probability of a puncture caused by rupture when a foreign body impacts. It can be seen that the automatic detection of vehicle deformation is very important for reducing tire wear and improving driving safety. However, it is difficult to directly observe the fine deformation rate of tires with the naked eye. In view of this, an automatic detection method of the deformation rate is urgently needed to solve the safety problem caused by abnormal deformation of vehicle tires.
  • the object of the present invention is to provide a method for detecting abnormal tire deformation based on tire thickness measurement, which solves the problem that the existing vehicle tires cannot be detected for abnormal deformation, which causes the driving safety of the vehicle.
  • the invention provides a method for detecting abnormal tire deformation based on tire thickness measurement, which includes the following steps:
  • Step 1 Perform preprocessing on the collected tire image to obtain a tire binary image
  • Step 2 Extract the wheel hub area and the ground area from the tire binary map obtained in step 1, and use them as the initial wheel hub image and the ground area map respectively;
  • Step 3 Perform morphological processing on the initial map of the hub in step 2 to obtain a filling map of the hub;
  • Step 4 Calculate the hub parameters based on the hub filling map obtained in step 3;
  • Step 5 construct a correction circle according to the wheel hub parameters obtained in step 4.
  • Step 6 Use the correction circle obtained in step 5 to update the wheel hub filling map; and calculate the centroid coordinates and virtual radius of the updated wheel hub filling map;
  • Step 7 Process the virtual radius of the hub filling map in step 3 and the virtual radius of the hub filling map updated in step 6, where, when
  • Step 8 Extract the lowest point group of the hub in the updated hub filling map
  • Step 9 Calculate the shortest distance between the lowest point group of the hub and the ground area
  • Step 10 Calculate the actual tire thickness of the tire according to the shortest distance between the lowest point group of the hub and the ground area obtained in step 9;
  • Step 11 Calculate the load deformation rate of the tire according to the actual tire thickness of the tire obtained in step 10;
  • Step 12 Determine the state of tire deformation according to the load deformation rate of the tire obtained in Step 11.
  • step 1 the collected tire image is preprocessed, and the specific method is:
  • S1 Perform grayscale and denoising processing on the collected tire image, and then obtain the preprocessed tire image;
  • S2 Perform binary classification and segmentation on the preprocessed tire image to obtain a preliminary binary image of the tire
  • the collecting device when collecting tire images, is at the same height as the center of the wheel hub, and the shooting angle coincides with the wheel axis.
  • step 2 the wheel hub area and the ground area are extracted from the tire binary map obtained in step 1.
  • the specific method is:
  • the connected area block is the hub area, which is used as the initial image of the hub:
  • ba i is the sum of pixels of the i-th connected area in the tire binary image
  • ⁇ 1 is the hub size threshold
  • bc i is the sum of pixels of the circumscribed rectangular template of the i-th connected area
  • Is the rectangularity of the i-th connected region in the tire binary map
  • ⁇ 1 is the rectangularity threshold
  • Ox i is the row coordinate value of the centroid of the i-th connected region
  • ⁇ 1 is the hub position threshold
  • the connected area block in the tire binary map satisfies the following formula at the same time, the connected area block is considered to be a ground area, and it is regarded as a ground area map:
  • step 3 morphological processing is performed on the initial image of the hub in step 2 to obtain a filling image of the hub.
  • the specific method is:
  • the circular structure operator sf is used to perform a closed operation on the initial image of the hub to fill in the missing small area blocks in the initial image of the hub to obtain a filling image of the hub;
  • step 4 calculate the hub parameters according to the hub filling map obtained in step 3.
  • the specific method is:
  • cx and cy are the row and column coordinates of the centroid of the hub area, respectively; r is the virtual radius.
  • step 5 the correction circle is constructed according to the wheel hub parameters obtained in step 4.
  • the specific method is:
  • step 6 use the correction circle obtained in step 5 to update the wheel hub filling map.
  • the specific method is:
  • step 8 extracting the lowest point group of the hub in the updated hub filling map
  • step 9 calculate the shortest distance between the lowest point group of the hub and the ground area according to the following formula:
  • d p is the shortest distance between the pixel point p in the lowest point group and the ground area; x p and y p are the row and column coordinate values of the pixel point p; x q and y q are the row and column coordinate values of the pixel point q .
  • step 10 the actual thickness of the tire is calculated according to the shortest distance between the lowest point group of the hub and the ground area obtained in step 9, and the specific method is:
  • dt is the actual tire thickness of the tire
  • dl is the actual diameter of the hub
  • 2r is the pixel diameter of the hub.
  • step 11 the load deformation rate of the tire is calculated according to the actual tire thickness of the tire obtained in step 10.
  • the specific method is:
  • ⁇ is the load deformation rate of the tire
  • db is the standard tire thickness of the tire
  • db S ⁇
  • S is the tire width
  • ⁇ is the aspect ratio
  • step 12 the state of tire deformation is determined according to the load deformation rate of the tire obtained in step 11, specifically
  • fl is the status mark of the tire deformation
  • ⁇ is the upper threshold of the deformation
  • ⁇ is the lower threshold of deformation
  • dd is the tire thickness value of the new tire measured under standard load and low air pressure
  • dh is the tire thickness value of the new tire measured under standard load and high air pressure.
  • the present invention provides a method for detecting abnormal tire deformation based on tire thickness measurement. Based on the circular characteristics of the wheel hub, the tire deformation rate is calculated using image processing technology, and compared with the deformation rate under standard high pressure and low pressure, it can be determined Judging the state of tire deformation.
  • the method of the present invention can detect abnormal tire deformation when there is no professional tire pressure measurement tool, or when the temperature, load, and road conditions change, thereby improving the driving safety of the vehicle, and is especially suitable for real-time monitoring tires of high-risk vehicles.
  • Figure 1 is a tire image f
  • Figure 2 is a preprocessed tire image g
  • Figure 3 is a preliminary diagram b of tire binary value
  • Figure 4 is a tire binary diagram e
  • Figure 5 is a preliminary view of the wheel hub c
  • Figure 6 is a ground area map R
  • Figure 7 is the wheel hub filling diagram ca.
  • the present invention provides a method for detecting abnormal tire deformation based on tire thickness measurement.
  • the method can be run offline on the client side, and can also be combined with cloud data to achieve online automatic detection; specifically including the following steps:
  • Step 0 Car tire image acquisition. Under the condition that the tire of the vehicle is not removed, use the image acquisition device (or smart phone) to collect the image of the automobile tire. The tire part is located in the upper middle of the image.
  • the collected tire image is represented by the symbol f, and the image size is denoted as M and N, that is, the total number of rows and the total number of columns of the image f are M rows and N columns, respectively, and the total pixels are M ⁇ N.
  • Image preprocessing includes: image gradation, denoising, etc.
  • image gradation can reduce the calculation amount of subsequent processing, there are many methods of gradation processing, and the weighted average method can be used for gradation processing to get a better comparison.
  • Reasonable gray-scale image denoising processing can reduce noise in tire images (such as mud spots and stains, etc.) and improve the accuracy of subsequent processing.
  • Denoising processing can use a median filter for salt and pepper noise.
  • the pre-processed tire image g is shown in FIG. 2.
  • Step 2 Preprocess the binarization of the tire image.
  • the preprocessed tire image g has a sharp contrast, the ground and the wheel are bright colors, and the tires are dark colors. Using the brightness difference, the preprocessed tire image g is divided into two categories, as shown in equation (1), the tire binary initial Figure b.
  • x and y are the row and column coordinates of a certain pixel in the image, both are integer values, and 1 ⁇ x ⁇ M; 1 ⁇ y ⁇ N; T is the image segmentation threshold, which can be determined by the maximum between-class method .
  • the initial binary value diagram b of the tire is shown in FIG. 3.
  • Step 3 Morphological processing of tire binary graph. Due to the influence of noise and shadows, in the initial image b of the tire binary value, there are still holes in areas such as the wheel hub and the ground, which are not conducive to subsequent processing, so morphological processing should be performed on it first. Specifically, it includes: firstly use a small structure operator se to perform a closed operation on the tire binary initial map b (that is, first expand and then corrode), and then perform the hole filling operation to obtain the tire binary map e.
  • the structure operator se selects a circular structure operator with a smaller radius (radius 5) to prevent the non-hub area from connecting the wheel hub area.
  • the tire binary map e is shown in Figure 4. Shown.
  • Step 4 Extract the wheel hub area of the tire binary map e.
  • the wheel hub has the characteristics of being round, larger in size, and located in the upper part of the image in the tire image. Based on the characteristics of the wheel hub, the wheel hub area in the tire binary map e is extracted. Assuming that there are H connected area blocks in the tire binary map e, if a connected area block satisfies the formula (2-4) at the same time, then the connected area block is considered to be a hub area, and this area block is extracted to obtain the initial wheel hub image c.
  • the value of ⁇ 1 is 0.2, the value of ⁇ 1 is 0.7, and the value of ⁇ 1 is 0.65;
  • Step 5 Extract the ground area of the tire binary map e.
  • the ground area in the tire image is large in size and located at the bottom of the image. Based on this, if a connected area block in the tire binary map e satisfies equation (5-6) at the same time, then the connected area block is considered to be a ground area, and the ground The area picture is R.
  • Step 6 Use the larger circular structure operator sf to perform a closed operation on the initial image c of the hub (that is, first expansion and then corrosion calculation) to fill in the missing small area blocks in the initial image c of the hub to obtain the hub filling image ca .
  • Step 7 Calculate the parameters of the wheel hub.
  • the hub filling map ca only the hub area is marked as 1, and the pixel value of the remaining area is 0.
  • the parameter calculation process of the hub area is shown in equation (7-9).
  • cx and cy are the row and column coordinates of the centroid of the hub area, respectively; r is the virtual radius.
  • cx and cy are 337 and 369, respectively, and the virtual radius r is 216.3.
  • Step 8 Construct a correction circle.
  • the purpose of constructing the correction circle is to accurately segment the hub area.
  • the parameters that need to be determined for the correction circle are the center coordinates (cx', cy') and the radius r'.
  • Step 9 Update the wheel filling diagram ca. Use the correction circle to remove the noise area in ca, that is, set the pixel values outside the correction circle to 0, and the pixel values inside the correction circle remain unchanged.
  • Step 10 Calculate the centroid coordinates (cx, cy) and virtual radius r of the updated wheel hub filling map ca.
  • the calculation formula is as shown in (7-9).
  • Step 11 If
  • ⁇ represents the radius threshold, and the value is 0.001-0.1.
  • the radius threshold ⁇ takes a value of 0.05.
  • Step 12 Extract the lowest point group of the wheel hub.
  • the updated wheel hub filling map ca all the pixels with the largest row coordinates are the lowest point group of the wheel hub, denoted by W.
  • Step 13 Calculate the shortest distance between the lowest point group of the hub and the ground area.
  • d p is the shortest distance between the pixel point p in the lowest point group and the ground area; x p and y p are the row and column coordinates of the pixel point p; x q and y q are the row, Column coordinate value.
  • the shortest distance between the lowest point group and the ground area is shown in Table 2.
  • Step 14 Calculate the unit pixel size
  • dl is the actual diameter of the hub known from the specifications and dimensions; 2r is the pixel diameter of the hub.
  • dl is 15 inches, that is, 38.1 cm, so
  • Step 15 Calculate the actual tire thickness of the tire.
  • the average of the shortest distance of the lowest point group of the wheel hub is the minimum tire thickness of the tire.
  • dt is the actual tire thickness of the tire.
  • dt is 5.1382.
  • Step 16 Calculate the standard tire thickness of the tire. Calculate the standard tire thickness of the tire according to the standard size of the tire (tyre width and aspect ratio).
  • S is the tire width and ⁇ is the aspect ratio.
  • the tire width S is 195 cm
  • Step 17 Calculate the load deformation rate ⁇ of the tire.
  • the load deformation rate of the tire ⁇ 0.3992.
  • Step 18 Determine the abnormal state of tire deformation. Determine the state of tire deformation based on the deformation rate: normal or abnormal.
  • ⁇ and ⁇ are respectively the upper and lower thresholds of deformation, which are determined by the standard low pressure It is determined by the tire thickness of a car under high pressure, and its calculation formula is shown in (16-17).
  • dd is the tire thickness value of the new tire measured under standard load and low air pressure
  • dh is the tire thickness value of the new tire measured under standard load and high air pressure.
  • Step 19 The algorithm ends.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Geometry (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Tires In General (AREA)

Abstract

一种基于胎厚测量的轮胎异常变形量的检测方法，包括以下步骤：基于轮毂的圆形特性，利用图像处理的技术计算轮胎变形率，与标准高压、低压下的变形率做比较，可以判定轮胎变形的状态判定；该方法能够实现对轮胎的异常变形进行检测，提高车辆的行车安全，尤其适用于高危车辆的实时监控轮胎。

Description

一种基于胎厚测量的轮胎异常变形量的检测方法 技术领域
本发明涉及汽车外观检测诊断的技术领域，特别是一种基于胎厚测量的轮胎异常变形量的检测方法。
背景技术
汽车轮胎是强弹性的橡胶制品，它和路面直接接触，是汽车的重要部件之一，在行驶中起着非常重要的作用：承受着汽车的重量；和汽车悬架共同来缓和汽车行驶时的冲击，保证车辆的舒适性和平顺性；保证车轮和路面有良好的附着性，提高汽车的牵引性、制动性和通过性。
轮胎行驶中变形量会受胎压、路面温度、载重量、轮胎参数和路面平整度等诸多因素的影响而发生变化，变形量过大或过小都会引发诸多的隐患。车辆行驶中轮胎变形量较大时，与路面的摩擦系数增大，油耗上升，进而造成方向盘沉重，易跑偏；轮胎的胎壁及胎体长时间受到挤压和形变(屈挠)，会加速轮胎的老化，导致胎肩磨损；轮胎与地面的摩擦成倍增加，胎温急剧升高，轮胎***，强度下降，高速行驶时容易引发爆胎等等。若轮胎变形量较小时，轮胎的抓地力的减少、摩擦附着力降低，影响车辆制动性能；车体震动大，会影响司乘人员的舒适性，且间接影响其他零部件的寿命；加速轮胎胎面中央花纹的局部磨损，使轮胎寿命下降；胎壁张力过大，会导致胎体弹性下降，车辆受到的负荷增大，异物撞击时出现破裂导致爆胎的概率大大增加。可见，车辆的变形量的自动检测，对于减缓轮胎磨损和提高行车安全等非常重要。然而，轮胎细微的变形率很难用肉眼直接观测到，鉴于此，亟需一种变形率的自动检测方法，解决车辆轮胎异常变形带来的安全问题。
发明内容
本发明的目的在于提供一种基于胎厚测量的轮胎异常变形量的检测方法，解决了现有的车辆轮胎异常变形无法进行检测，导致车辆的行车安全存在问题。
为了达到上述目的，本发明采用的技术方案是：
本发明提供的一种基于胎厚测量的轮胎异常变形量的检测方法，包括以下步骤：
步骤1，对采集得到的轮胎图像进行预处理，得到轮胎二值图；
步骤2，在步骤1中得到的轮胎二值图上提取轮毂区域和地面区域，分别作为轮毂初图和地面区图；
步骤3，对步骤2中的轮毂初图进行形态处理，得到轮毂填补图；
步骤4，根据步骤3中得到的轮毂填补图计算轮毂参数；
步骤5，根据步骤4中得到的轮毂参数构建矫正圆；
步骤6，利用步骤5中得到的矫正圆更新轮毂填补图；并计算更新后的轮毂填补图的形心坐标和虚拟半径；
步骤7，将步骤3中的轮毂填补图的虚拟半径和步骤6中更新后的轮毂填补图的虚拟半径进行处理，其中，当|r'-r|≤τ，转入步骤8，否则，转入步骤5；
步骤8，在更新后的轮毂填补图中提取轮毂的最低点群；
步骤9，计算轮毂最低点群和地面区域之间的最短距离；
步骤10，根据步骤9得到的轮毂最低点群和地面区域之间的最短距离，计算轮胎的实际胎厚；
步骤11，根据步骤10中得到的轮胎的实际胎厚计算轮胎的载重变形率；
步骤12，根据步骤11中得到的轮胎的载重变形率判断轮胎变形的状态。
优选地，步骤1中，对采集得到的轮胎图像进行预处理，具体方法是：
S1，对采集得到的轮胎图像进行灰度化及去噪处理，之后得到预处理轮胎图像；
S2，对预处理轮胎图像进行二分类分割，得到轮胎二值初图；
S3，对得到的轮胎二值初图进行形态处理，得到轮胎二值图；
其中，在采集轮胎图像时，采集设备与轮毂中心同高度，拍摄角度与轮轴重合。
优选地，步骤2中，在步骤1中得到的轮胎二值图上提取轮毂区域和地面区域，具体方法是：
设定轮胎二值图上有H个连通区域块，若其中一个连通区域块同时满足下式，则该连通区域块为轮毂区域，作为轮毂初图：
Figure PCTCN2021081636-appb-000001
Figure PCTCN2021081636-appb-000002
Figure PCTCN2021081636-appb-000003
其中，ba i为轮胎二值图中第i个连通区域的像素总和；α 1为轮毂尺寸阈值；bc i为第i个连通区域的外接矩形模板的像素总和；
Figure PCTCN2021081636-appb-000004
为轮胎二值图中第i个连通区域的矩形度；β 1为矩形度阈值；Ox i为第i个连通区域的形心的行坐标值；γ 1为轮毂位置阈值；
若轮胎二值图中某个连通区域块同时满足下式，则认为该连通区域块为地面区域，作为地面区图：
Figure PCTCN2021081636-appb-000005
Figure PCTCN2021081636-appb-000006
优选地，步骤3中，对步骤2中的轮毂初图进行形态处理，得到轮毂填补图，具体方法是：
利用圆形结构算子sf对轮毂初图进行闭运算，以填补轮毂初图中漏检的小区域块，得到轮毂填补图；
步骤4中，根据步骤3中得到的轮毂填补图计算轮毂参数，具体方法是：
设定轮毂填补图中的轮毂区域的像素值为1，其余区域的像素值为0；根据下式计算轮毂区域的参数：
Figure PCTCN2021081636-appb-000007
Figure PCTCN2021081636-appb-000008
Figure PCTCN2021081636-appb-000009
其中，cx和cy分别为轮毂区域的形心行和列坐标值；r为虚拟半径。
优选地，步骤5中，根据步骤4中得到的轮毂参数构建矫正圆，具体方法是：
确定矫正圆的圆心坐标(cx’，cy’)和半径r’，其中，cx’＝cx；cy’＝cy；r’＝r；
步骤6中，利用步骤5中得到的矫正圆更新轮毂填补图，具体方法是：
利用矫正圆去除轮毂填补图中的噪声区域，得到更新后的轮毂填补图；
优选地，步骤8中，在更新后的轮毂填补图中提取轮毂的最低点群，具体方法是：
在更新后的轮毂填补图中，将行坐标为最大值的所有像素点作为轮毂最低点群；
步骤9中，根据下式计算轮毂最低点群和地面区域之间的最短距离：
Figure PCTCN2021081636-appb-000010
其中，d p为最低点群中像素点p和地面区域的最短距离；x p和y p为像素点p的行、列坐标值；x q和y q为像素点q的行、列坐标值。
优选地，步骤10中，根据步骤9得到的轮毂最低点群和地面区域之间的最短距离，计算轮胎的实际胎厚，具体方法是：
Figure PCTCN2021081636-appb-000011
其中，dt为轮胎的实际胎厚；
Figure PCTCN2021081636-appb-000012
为单位像素尺寸，
Figure PCTCN2021081636-appb-000013
dl为轮毂的实际直径；2r为轮毂的像素直径。
优选地，步骤11中，根据步骤10中得到的轮胎的实际胎厚计算轮胎的载重变形率，具体方法是：
Figure PCTCN2021081636-appb-000014
其中，ξ为轮胎的载重变形率；db为轮胎的标准胎厚，db＝S·μ，S为胎宽，μ为扁平比。
优选地，步骤12中，根据步骤11中得到的轮胎的载重变形率判断轮胎变形的状态，具体
方法是：
Figure PCTCN2021081636-appb-000015
其中，fl为轮胎变形量的状态标记，fl＝1表示轮胎变形量为异常，fl＝0表示轮胎变形量为正常；κ为变形量上阈值，
Figure PCTCN2021081636-appb-000016
λ为变形量下阈值，
Figure PCTCN2021081636-appb-000017
dd为新轮胎在标准载重、低气压下测得的胎厚值；dh为新轮胎在标准载重、高气压下测得的胎厚值。
与现有技术相比，本发明的有益效果是：
本发明提供的一种基于胎厚测量的轮胎异常变形量的检测方法，基于轮毂的圆形特性，利用图像处理的技术计算轮胎变形率，与标准高压、低压下的变形率做比较，可以判定轮胎变形的状态判定。本发明方法可以实现没有专业胎压测量工具，或气温、载重和路面情况等发生变化时，对轮胎的异常变形进行检测，提高车辆的行车安全，尤其适用于高危车辆的实时监控轮胎。
附图说明
图1是轮胎图像f；
图2是预处理轮胎图像g；
图3是轮胎二值初图b；
图4是轮胎二值图e；
图5是轮毂初图c；
图6是地面区图R；
图7是轮毂填补图ca。
具体实施方式
下面结合附图，对本发明进一步详细说明。
本发明提供的一种基于胎厚测量的轮胎异常变形量的检测方法，该方法能够在客户端离线运行，也可以和云数据结合，实现线上自动检测；具体包括以下步骤：
步骤0：汽车轮胎图像采集。在车辆轮胎不拆卸的情形下，用图像采集设备(或智能手机)采集汽车轮胎的图像，采集设备应与轮毂中心同高度，使拍摄角度与轮轴重合，要求拍摄到完整的轮胎图像，并使轮胎部分位于图像的中上部。采集到的轮胎图像用符号f表示，图像大小记为M和N，即图像f的总行数和总列数分别为M行和N列，总像素点为M×N。
本实施例中，M＝773，N＝604，轮胎图像f如图1所示。
步骤1：图像预处理。图像预处理包括：图像灰度化、去噪等处理，图像灰度化可以降低后续处理的计算量，灰度化处理的方法有很多，可采用加权平均法进行灰度化处理，以得到较合理的灰度图像；去噪处理能减少轮胎图像中噪声(如泥点和污渍等)，提高后续处理的精度，去噪处理可以采用针对椒盐噪声的中值滤波器。轮胎图像f经过预处理后得到预处理轮胎图像g。
实施例中，预处理轮胎图像g如图2所示。
步骤2：预处理轮胎图像的二值化。预处理轮胎图像g具有鲜明的对比度，地面和轮毂为高亮色，轮胎为暗色，利用亮度的对差异对预处理轮胎图像g进行二分类分割，如式(1)所示，得到轮胎二值初图b。
Figure PCTCN2021081636-appb-000018
其中，x和y是图像中某像素点的行、列坐标值，均为整数值，且，1≤x≤M；1≤y≤N；T为图像分割阈值，可由最大类间法确定得到。
本实施例中，轮胎二值初图b如图3所示。
步骤3：轮胎二值图的形态处理。由于受噪声和阴影等的影响，轮胎二值初图b中，轮毂和地面等区域还存在空洞，不利于后续的处理，故应先对其执行形态处理。具体包括：先用小的结构算子se对轮胎二值初图b进行闭运算(即先膨胀、后腐蚀运算)，接着进行填空洞操作，得到轮胎二值图e。
本实施例中，考虑到轮毂的形状，结构算子se选用半径较小(半径为5)的圆形结构算子，以防止将非轮毂区域和轮毂区域相连，轮胎二值图e如图4所示。
步骤4：提取轮胎二值图e的轮毂区域。轮毂在轮胎图像中具有为圆形、尺寸较大、位于图像上部等特征，基于轮毂特征提取轮胎二值图e中的轮毂区域。假定轮胎二值图e中有H个连通区域块，若某连通区域块同时满足式(2-4)，则认为该连通区域块为轮毂区域，提取该区域块得到轮毂初图c。
Figure PCTCN2021081636-appb-000019
Figure PCTCN2021081636-appb-000020
Figure PCTCN2021081636-appb-000021
其中，ba i为轮胎二值图e中第i个连通区域的像素总和；α 1为轮毂尺寸阈值，通常取值为0.18-0.5；bc i为第i个连通区域的外接矩形模板(外接矩形及其内部像素值均为1)的像素总和；
Figure PCTCN2021081636-appb-000022
为轮胎二值图e中第i个连通区域的矩形度；β 1为矩形度阈值，通常为0.70-0.85；Ox i为第i个连通区域的形心的行坐标值；γ 1为轮毂位置阈值，取值为0.6-0.75。
本实施例中，α 1取值为0.2，β 1取值为0.7，γ 1取值为0.65；轮毂区域i＝l的各参数如下：ba l＝144843，bc l＝189570，Ox l＝369，有：
Figure PCTCN2021081636-appb-000023
和
Figure PCTCN2021081636-appb-000024
均符合式(2-4)中的条件，提取的轮毂初图c如图5所示。
步骤5：提取轮胎二值图e的地面区域。轮胎图像中的地面区域尺寸较大、位于图像的底部，基于此，若轮胎二值图e中某连通区域块同时满足式(5-6)，则认为该连通区域块为地面区域，记地面区图为R。
Figure PCTCN2021081636-appb-000025
Figure PCTCN2021081636-appb-000026
本实施例中，地面区域i＝j的各参数如下：ba j＝128001，Ox j＝684，有：
Figure PCTCN2021081636-appb-000027
和
Figure PCTCN2021081636-appb-000028
符合式(5-6)中的条件，提取的地面区图为R如图6所示。
步骤6：利用较大的圆形结构算子sf对轮毂初图c进行闭运算(即先膨胀、后腐蚀运算)，以填补轮毂初图c中漏检的小区域块，得到轮毂填补图ca。
本实施例中，选用半径为35的圆形结构算子sf，得到的轮毂填补图ca如图7所示。
步骤7：计算轮毂的参数。轮毂填补图ca中只有轮毂区域被标记为1，其余区域像素值为 0，轮毂区域的参数计算过程如式(7-9)所示。
Figure PCTCN2021081636-appb-000029
Figure PCTCN2021081636-appb-000030
Figure PCTCN2021081636-appb-000031
其中，cx和cy分别为轮毂区域的形心行、列坐标值；r为虚拟半径。
本实施例中，cx和cy分别为337和369，虚拟半径r为216.3。
步骤8：构建矫正圆。构建矫正圆的目的是为了精确分割出轮毂区域，矫正圆需要确定的参数有圆心坐标(cx’，cy’)和半径r’，矫正圆参数的初始化或更新策略为：cx’＝cx,cy’＝cy和r’＝r。
步骤9：更新轮毂填补图ca。用矫正圆去除ca中的噪声区域，即将位于矫正圆之外的像素值均置为0，位于矫正圆内的像素值保持不变。
步骤10：计算更新的轮毂填补图ca的形心坐标(cx,cy)和虚拟半径r，其计算式如(7-9)所示。
步骤11：若|r'-r|≤τ，转入步骤12；否则，转入步骤8。其中，τ表示半径阈值，取值分别为0.001-0.1。
本实施例中，半径阈值τ取值为0.05。经过21次步骤8到11的循环迭代，r’＝212.4075，r＝212.4367，可知|r'-r|＝0.0292<0.05＝τ。
步骤12：提取轮毂的最低点群。经过更新后的轮毂填补图ca中，行坐标最大的所有像素 点为轮毂最低点群，记W。
本实施例中，轮毂最低点群W的行列坐标值如表1所示。
表1轮毂最低点群W的坐标值
序号 1 2 3 4 5 6 7 8 9 10 11 12
行坐标值 577 577 577 577 577 577 577 577 577 577 577 577
列坐标值 321 322 323 324 325 326 327 328 329 330 331 332
序号 13 14 15 16 17 18 19 20 21 22 23 24
行坐标值 577 577 577 577 577 577 577 577 577 577 577 577
列坐标值 333 334 335 336 337 338 339 340 341 342 343 344
序号 25 26 27 28 29 30 31 32 33 34 35  
行坐标值 577 577 577 577 577 577 577 577 577 577 577  
列坐标值 345 346 347 348 349 350 351 352 353 354 355  
步骤13：计算轮毂最低点群和地面区域的最短距离。
Figure PCTCN2021081636-appb-000032
其中，d p为最低点群中像素点p和地面区域的最短距离；x p和y p为像素点p的行、列坐标值；x q和y q为地面区域中像素点q的行、列坐标值。
本实施例中，最低点群和地面区域的最短距离如表2所示。
表2最低点群和地面区域的最短距离
序号 1 2 3 4 5 6 7 8 9 10 11 12
最短距离 29 29 29 29 29 29 29 29 29 29 29 29
序号 13 14 15 16 17 18 19 20 21 22 23 24
最短距离 29 29 29 29 28 28 28 28 28 28 28 29
序号 25 26 27 28 29 30 31 32 33 34 35  
最短距离 29 29 29 29 29 29 28 28 28 28 28  
步骤14：计算单位像素尺寸
Figure PCTCN2021081636-appb-000033
Figure PCTCN2021081636-appb-000034
其中，dl为根据规格尺寸获知的轮毂的实际直径；2r为轮毂的像素直径。
本实施例中，dl为15英寸，即38.1cm，得
Figure PCTCN2021081636-appb-000035
步骤15：计算轮胎的实际胎厚。轮毂最低点群最短距离的平均值即为轮胎的最小胎厚。
Figure PCTCN2021081636-appb-000036
其中，dt为轮胎的实际胎厚。
本实施例中，dt为5.1382。
步骤16：计算轮胎的标准胎厚。根据轮胎的规范尺寸(胎宽和扁平比)，计算轮胎的标准胎厚。
db＝S·μ       (13)
其中，S为胎宽，μ为扁平比。
本实施例中，轮胎的胎宽S为195cm，扁平比尺寸μ为：65％，可知db＝12.87cm。
步骤17：计算轮胎的载重变形率ξ。
Figure PCTCN2021081636-appb-000037
本实施例中，轮胎的载重变形率ξ＝0.3992。
步骤18：判定轮胎变形的异常状态。依据变形率判定轮胎变形的状态：正常或异常。
Figure PCTCN2021081636-appb-000038
其中，fl为轮胎变形量的状态标记，为二值变量，fl＝1表示轮胎变形量为异常，fl＝0表示轮胎变形量为正常；κ和λ分别变形量上、下阈值，由标准低压和高压下的汽车胎厚决定，其计算式如(16-17)所示。
在标准载重时，测量新轮胎在标准低压和高压下的胎厚，这是一个基准值，当测定好后，直接查询即可。
Figure PCTCN2021081636-appb-000039
Figure PCTCN2021081636-appb-000040
其中，dd为新轮胎在标准载重、低气压下测得的胎厚值；dh为新轮胎在标准载重、高气压下测得的胎厚值。
本实施例中，测量的κ和λ分别为0.75和0.85，由于ξ<κ，故fl＝0。
步骤19：算法结束。

Claims (9)

  1. 一种基于胎厚测量的轮胎异常变形量的检测方法，其特征在于，包括以下步骤：
    步骤1，对采集得到的轮胎图像进行预处理，得到轮胎二值图；
    步骤2，在步骤1中得到的轮胎二值图上提取轮毂区域和地面区域，分别作为轮毂初图和地面区图；
    步骤3，对步骤2中的轮毂初图进行形态处理，得到轮毂填补图；
    步骤4，根据步骤3中得到的轮毂填补图计算轮毂参数；
    步骤5，根据步骤4中得到的轮毂参数构建矫正圆；
    步骤6，利用步骤5中得到的矫正圆更新轮毂填补图；并计算更新后的轮毂填补图的形心坐标和虚拟半径；
    步骤7，将步骤3中的轮毂填补图的虚拟半径和步骤6中更新后的轮毂填补图的虚拟半径进行处理，其中，当|r'-r|≤τ，转入步骤8，否则，转入步骤5；
    步骤8，在更新后的轮毂填补图中提取轮毂的最低点群；
    步骤9，计算轮毂最低点群和地面区域之间的最短距离；
    步骤10，根据步骤9得到的轮毂最低点群和地面区域之间的最短距离，计算轮胎的实际胎厚；
    步骤11，根据步骤10中得到的轮胎的实际胎厚计算轮胎的载重变形率；
    步骤12，根据步骤11中得到的轮胎的载重变形率判断轮胎变形的状态。
  2. 根据权利要求1所述的一种基于胎厚测量的轮胎异常变形量的检测方法，其特征在于，步骤1中，对采集得到的轮胎图像进行预处理，具体方法是：
    S1，对采集得到的轮胎图像进行灰度化及去噪处理，之后得到预处理轮胎图像；
    S2，对预处理轮胎图像进行二分类分割，得到轮胎二值初图；
    S3，对得到的轮胎二值初图进行形态处理，得到轮胎二值图；
    其中，在采集轮胎图像时，采集设备与轮毂中心同高度，拍摄角度与轮轴重合。
  3. 根据权利要求1所述的一种基于胎厚测量的轮胎异常变形量的检测方法，其特征在于，步骤2中，在步骤1中得到的轮胎二值图上提取轮毂区域和地面区域，具体方法是：
    设定轮胎二值图上有H个连通区域块，若其中一个连通区域块同时满足下式，则该连通区域块为轮毂区域，作为轮毂初图：
    Figure PCTCN2021081636-appb-100001
    Figure PCTCN2021081636-appb-100002
    Figure PCTCN2021081636-appb-100003
    其中，ba i为轮胎二值图中第i个连通区域的像素总和；α 1为轮毂尺寸阈值；bc i为第i个连通区域的外接矩形模板的像素总和；
    Figure PCTCN2021081636-appb-100004
    为轮胎二值图中第i个连通区域的矩形度；β 1为矩形度阈值；Ox i为第i个连通区域的形心的行坐标值；γ 1为轮毂位置阈值；
    若轮胎二值图中某个连通区域块同时满足下式，则认为该连通区域块为地面区域，作为地面区图：
    Figure PCTCN2021081636-appb-100005
    Figure PCTCN2021081636-appb-100006
  4. 根据权利要求1所述的一种基于胎厚测量的轮胎异常变形量的检测方法，其特征在于，步骤3中，对步骤2中的轮毂初图进行形态处理，得到轮毂填补图，具体方法是：
    利用圆形结构算子sf对轮毂初图进行闭运算，以填补轮毂初图中漏检的小区域块，得到轮毂填补图；
    步骤4中，根据步骤3中得到的轮毂填补图计算轮毂参数，具体方法是：
    设定轮毂填补图中的轮毂区域的像素值为1，其余区域的像素值为0；根据下式计算轮毂 区域的参数：
    Figure PCTCN2021081636-appb-100007
    Figure PCTCN2021081636-appb-100008
    Figure PCTCN2021081636-appb-100009
    其中，cx和cy分别为轮毂区域的形心行和列坐标值；r为虚拟半径。
  5. 根据权利要求1所述的一种基于胎厚测量的轮胎异常变形量的检测方法，其特征在于，步骤5中，根据步骤4中得到的轮毂参数构建矫正圆，具体方法是：
    确定矫正圆的圆心坐标(cx’，cy’)和半径r’，其中，cx’＝cx；cy’＝cy；r’＝r；
    步骤6中，利用步骤5中得到的矫正圆更新轮毂填补图，具体方法是：
    利用矫正圆去除轮毂填补图中的噪声区域，得到更新后的轮毂填补图。
  6. 根据权利要求1所述的一种基于胎厚测量的轮胎异常变形量的检测方法，其特征在于，步骤8中，在更新后的轮毂填补图中提取轮毂的最低点群，具体方法是：
    在更新后的轮毂填补图中，将行坐标为最大值的所有像素点作为轮毂最低点群；
    步骤9中，根据下式计算轮毂最低点群和地面区域之间的最短距离：
    Figure PCTCN2021081636-appb-100010
    其中，d p为最低点群中像素点p和地面区域的最短距离；x p和y p为像素点p的行、列坐标值；x q和y q为像素点q的行、列坐标值。
  7. 根据权利要求1所述的一种基于胎厚测量的轮胎异常变形量的检测方法，其特征在于， 步骤10中，根据步骤9得到的轮毂最低点群和地面区域之间的最短距离，计算轮胎的实际胎厚，具体方法是：
    Figure PCTCN2021081636-appb-100011
    其中，dt为轮胎的实际胎厚；
    Figure PCTCN2021081636-appb-100012
    为单位像素尺寸，
    Figure PCTCN2021081636-appb-100013
    dl为轮毂的实际直径；2r为轮毂的像素直径。
  8. 根据权利要求1所述的一种基于胎厚测量的轮胎异常变形量的检测方法，其特征在于，步骤11中，根据步骤10中得到的轮胎的实际胎厚计算轮胎的载重变形率，具体方法是：
    Figure PCTCN2021081636-appb-100014
    其中，ξ为轮胎的载重变形率；db为轮胎的标准胎厚，db＝S·μ，S为胎宽，μ为扁平比。
  9. 根据权利要求1所述的一种基于胎厚测量的轮胎异常变形量的检测方法，其特征在于，步骤12中，根据步骤11中得到的轮胎的载重变形率判断轮胎变形的状态，具体方法是：
    Figure PCTCN2021081636-appb-100015
    其中，fl为轮胎变形量的状态标记，fl＝1表示轮胎变形量为异常，fl＝0表示轮胎变形量为正常；κ为变形量上阈值，
    Figure PCTCN2021081636-appb-100016
    λ为变形量下阈值，
    Figure PCTCN2021081636-appb-100017
    dd为新轮胎在标准载重、低气压下测得的胎厚值；dh为新轮胎在标准载重、高气压下测得的胎厚值。
PCT/CN2021/081636 2020-03-20 2021-03-18 一种基于胎厚测量的轮胎异常变形量的检测方法 WO2021185332A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010203370.0A CN111429505B (zh) 2020-03-20 2020-03-20 一种基于胎厚测量的轮胎异常变形量的检测方法
CN202010203370.0 2020-03-20

Publications (1)

Publication Number Publication Date
WO2021185332A1 true WO2021185332A1 (zh) 2021-09-23

Family

ID=71548517

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/081636 WO2021185332A1 (zh) 2020-03-20 2021-03-18 一种基于胎厚测量的轮胎异常变形量的检测方法

Country Status (2)

Country Link
CN (1) CN111429505B (zh)
WO (1) WO2021185332A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116205923A (zh) * 2023-05-06 2023-06-02 威海锐鑫丰金属科技有限公司 基于x-ray的汽车轮毂内部缺陷无损检测方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111429505B (zh) * 2020-03-20 2023-03-28 长安大学 一种基于胎厚测量的轮胎异常变形量的检测方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105172494A (zh) * 2015-09-18 2015-12-23 浙江宇视科技有限公司 一种轮胎安全检测方法及轮胎安全检测***
CN109003258A (zh) * 2018-06-15 2018-12-14 广东工业大学 一种高精度亚像素圆形零件测量方法
US20190339110A1 (en) * 2018-05-02 2019-11-07 Konica Minolta, Inc. Overload Detection Processing Apparatus, Overload Detection System and Computer-Readable Recording Medium Storing Program
CN110553594A (zh) * 2018-05-31 2019-12-10 柯尼卡美能达株式会社 图像处理装置、过载检测装置、过载检测***以及介质
CN111383225A (zh) * 2020-03-20 2020-07-07 长安大学 一种对比标准变形率的轮胎异常状态的检测方法
CN111429505A (zh) * 2020-03-20 2020-07-17 长安大学 一种基于胎厚测量的轮胎异常变形量的检测方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004038794A (ja) * 2002-07-05 2004-02-05 Toyota Motor Corp 画像処理装置、及び画像処理方法
CN101856947A (zh) * 2010-06-09 2010-10-13 孟祥厚 有胎斜挡圈兼爆胎轮的汽车轮毂
CN105021126A (zh) * 2015-06-18 2015-11-04 中国计量学院 基于机器视觉技术货车侧防护栏安装尺寸测量***及方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105172494A (zh) * 2015-09-18 2015-12-23 浙江宇视科技有限公司 一种轮胎安全检测方法及轮胎安全检测***
US20190339110A1 (en) * 2018-05-02 2019-11-07 Konica Minolta, Inc. Overload Detection Processing Apparatus, Overload Detection System and Computer-Readable Recording Medium Storing Program
CN110553594A (zh) * 2018-05-31 2019-12-10 柯尼卡美能达株式会社 图像处理装置、过载检测装置、过载检测***以及介质
CN109003258A (zh) * 2018-06-15 2018-12-14 广东工业大学 一种高精度亚像素圆形零件测量方法
CN111383225A (zh) * 2020-03-20 2020-07-07 长安大学 一种对比标准变形率的轮胎异常状态的检测方法
CN111429505A (zh) * 2020-03-20 2020-07-17 长安大学 一种基于胎厚测量的轮胎异常变形量的检测方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116205923A (zh) * 2023-05-06 2023-06-02 威海锐鑫丰金属科技有限公司 基于x-ray的汽车轮毂内部缺陷无损检测方法

Also Published As

Publication number Publication date
CN111429505A (zh) 2020-07-17
CN111429505B (zh) 2023-03-28

Similar Documents

Publication Publication Date Title
WO2021185332A1 (zh) 一种基于胎厚测量的轮胎异常变形量的检测方法
JP5111505B2 (ja) タイヤ摩耗推定方法
JP5620268B2 (ja) タイヤ摩耗推定方法及びタイヤ摩耗推定装置
CN106740829B (zh) 基于聚类分析双半挂汽车行驶稳定性自动识别与预警***
US8532872B2 (en) Tire pressure adjustment
US20060076095A1 (en) Detection device for decreased tire pressures and method thereof
CN104470732A (zh) 用于估计轮胎的胎纹深度的方法和装置
US9087263B2 (en) Vision based pedestrian and cyclist detection method
CN109649091A (zh) 基于计算机视觉的汽车轮胎监测***
WO2020071249A1 (ja) タイヤ摩耗推定方法
CN105793687A (zh) 轮胎异常磨损的估计方法和估计设备
CN109774717B (zh) 轮胎摩擦力调控方法及***
KR102025132B1 (ko) 영상이미지 분석 기반의 타이어 공기압 측정 시스템 및 방법
US20220055634A1 (en) Method for estimating the pressure of the tyres of a vehicle
US12031277B2 (en) Pavement anti-skid performance evaluation method and device based on envelope feature
US20090261961A1 (en) Method and apparatus for detecting decrease in tire air pressure and program for determining decrease in tire air pressure
CN112996676B (zh) 用于确定施加至轮胎的载荷的方法
CN109918792B (zh) 一种基于计算机的轮胎不平衡量的动力学模拟***及方法
CN111383225B (zh) 一种对比标准变形率的轮胎异常状态的检测方法
CN113008985B (zh) 一种利用轮胎/路面噪声评价道路路面构造的方法
JP6863101B2 (ja) タイヤの減圧の検出装置
WO2020044739A1 (ja) タイヤ摩耗検知方法及びタイヤ摩耗検知装置
CN115179693B (zh) 一种车辆水滑的预警方法
CN115100510B (zh) 一种轮胎磨损度识别方法
CN113506254B (zh) 跟踪轮胎胎面磨损方法、装置、设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21770697

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21770697

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 21770697

Country of ref document: EP

Kind code of ref document: A1