CN114926410A - Method for detecting appearance defects of brake disc - Google Patents
Method for detecting appearance defects of brake disc Download PDFInfo
- Publication number
- CN114926410A CN114926410A CN202210473779.3A CN202210473779A CN114926410A CN 114926410 A CN114926410 A CN 114926410A CN 202210473779 A CN202210473779 A CN 202210473779A CN 114926410 A CN114926410 A CN 114926410A
- Authority
- CN
- China
- Prior art keywords
- brake disc
- type
- edge pixel
- index
- pixel 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.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 83
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000001514 detection method Methods 0.000 claims abstract description 24
- 239000013598 vector Substances 0.000 claims abstract description 17
- 230000002950 deficient Effects 0.000 claims description 5
- 150000001875 compounds Chemical class 0.000 claims description 2
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 239000002537 cosmetic Substances 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 2
- 238000013441 quality evaluation Methods 0.000 abstract 1
- 230000011218 segmentation Effects 0.000 description 4
- 238000003708 edge detection Methods 0.000 description 3
- 230000000877 morphologic effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000003754 machining Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- 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
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- 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
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
-
- 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
- G06T2207/30168—Image quality inspection
Abstract
The invention relates to the field of brake disc processing detection, in particular to a brake disc appearance defect detection method, which clusters pixels in an interested area in a surface image of a brake disc to be detected, obtains the overall contrast of each type of pixel according to the category vector of each type of pixel, obtains the saliency of each type of pixel according to the overall contrast, the number of characteristic points and the distance index of each type of pixel, obtains a defect area according to the saliency, utilizes edge pixel in a sliding window to fit a straight line, obtains an angle index of each edge pixel according to the included angle between the fitted straight line and the positive horizontal direction, obtains the confidence to be selected of each edge pixel according to an angle index sequence, selects an optimal edge pixel to fit a candidate circle according to the confidence to be selected, obtains a brake disc shape index according to the candidate circle, obtains the brake disc appearance defect according to the saliency of the defect area and the brake disc shape index for quality evaluation, the method is accurate and efficient.
Description
Technical Field
The application relates to the field of brake disc machining detection, in particular to a method for detecting appearance defects of a brake disc.
Background
The brake disc is a mechanical part with a circular appearance, and is mainly used for braking automobile wheels, mechanical equipment, machine tools and the like. The appearance of the brake disc is inevitable in the manufacturing process, so that a plurality of defects can be caused, and the appearance contour of some defective brake discs can have larger form deviation. At present, most of brake disc production enterprises which finish machining detect the appearance quality of a brake disc mainly by means of manual visual inspection, and the method has the defects that the product detection efficiency is low, the phenomenon of poor product outflow occurs sometimes, and the product quality is difficult to guarantee due to the fact that the objectivity is insufficient, the dependence on the experience and responsibility of an inspector is large, the workload is large, and the like. Some adopt certain automatic means to carry out quality inspection to the brake disc outward appearance, but judge through the mode of comparing with standard sample after scanning the brake disc, the detection mode mainly has the problem such as time consuming, inefficiency and evaluation standard are too single, and the defect pattern changes slightly and just probably can not detect or take place the erroneous judgement, and the mode of this kind of traditional image comparison is not fit for the outward appearance quality inspection of casting blank product.
Disclosure of Invention
The invention provides a method for detecting appearance defects of a brake disc, which solves the problems of low detection efficiency and insufficient objectivity of the brake disc and adopts the following technical scheme:
acquiring a surface image of a brake disc to be detected;
clustering interested region pixel points of a surface image of a brake disc to be detected, and constructing a category vector of each type of pixel points according to a coordinate mean value and a gray mean value of each type of pixel points and a hue mean value, a saturation mean value and a brightness mean value of the type of pixel points in an HSV space;
obtaining the overall contrast of each type of pixel points according to the category vector of each type of pixel points;
obtaining feature points in each type of pixel points by utilizing angular point detection;
obtaining the distance index of each type of pixel points according to the distance between the center of each type of pixel points and the center of the neighborhood type pixel points;
obtaining the significance of each type of pixel points according to the overall contrast, the number of the characteristic points and the distance index of each type of pixel points, and taking the pixel point type with the significance greater than a threshold value as a defect area;
obtaining edge pixel points of a brake disc, performing clockwise sliding on the edge pixel points by using a sliding window, fitting straight lines to the edge pixel points in the sliding window, and obtaining an angle index of each edge pixel point according to an included angle between the fitted straight lines and the positive horizontal direction;
obtaining an angle index sequence of each edge pixel point according to the angle index of each edge pixel point and the angle indexes of the adjacent pixel points in front of and behind the edge pixel point;
obtaining the confidence coefficient to be selected of each edge pixel point according to the variance of the angle index sequence of each edge pixel point and the maximum value and the minimum value of the angle index sequence;
selecting an optimal edge pixel point according to the confidence coefficient to be selected of each edge pixel point, fitting a candidate circle through the optimal edge pixel point, and obtaining a brake disc morphological index according to the candidate circle;
and obtaining an appearance defect index of the brake disc according to the significance of the defect area and the shape index of the brake disc, and evaluating the quality of the appearance defect of the brake disc to be detected by using the defect index.
The method for calculating the overall contrast of each type of pixel point comprises the following steps:
calculating Euclidean distances between the category vector of each type of pixel point and the category vectors of other types of pixel points;
and taking the sum of the Euclidean distances as the integral contrast of the pixel points.
The method for calculating the significance of each type of pixel point comprises the following steps:
in the formula, X i Significance, Q, of the ith pixel i The number of feature points in the i-th pixel, L i Is a distance index of the i-th type pixel point, C i The overall contrast of the i-th pixel is obtained.
The method for acquiring the angle index sequence of each edge pixel point comprises the following steps:
obtaining an angle index of each edge pixel point;
obtaining angle indexes of z pixel points in front of the edge pixel point and angle indexes of z pixel points behind the edge pixel point, wherein z is an empirical value;
and taking the 2z +1 angle index as an angle index sequence of each edge pixel point.
The method for calculating the confidence to be selected of each edge pixel point comprises the following steps:
in the formula (I), the compound is shown in the specification,
is the confidence coefficient, sigma, to be selected of the edge pixel point c c Is the variance, gamma, of the angle index sequence of the edge pixels c max Is the maximum value, gamma, in the angle index sequence of the edge pixel point c min And the minimum value in the angle index sequence of the edge pixel point c is obtained.
The selection method of the optimal edge pixel point comprises the following steps:
and sequencing the confidence degrees to be selected of all the edge pixel points in a descending order, and taking the first S edge pixel points as the optimal edge pixel points, wherein S is an empirical value.
The calculation method of the brake disc form index comprises the following steps:
in the formula, U is the number of the pixel points at the contour edge of the brake disc, delta is the contour form index of the brake disc to be detected, and x u ，y u The coordinates of the edge pixel points are shown, a and b are the coordinates of the center of a fitting circle, and r is the radius of the fitting circle.
The method for calculating the appearance defect index of the brake disc comprises the following steps:
wherein A is the number of defective regions, X a And Q is the significance of the a-th defect area, and is an index of the apparent defect of the brake disc to be detected.
The method for evaluating the quality of the appearance defects of the brake disc to be detected by using the defect indexes comprises the following steps:
and when the appearance defect index Q of the brake disc to be detected is greater than 0.5, the appearance defect condition of the brake disc to be detected is serious, and the surface of the brake disc to be detected needs to be processed again.
The invention has the beneficial effects that: acquiring a brake disc surface image to be detected, clustering pixels in an interested area of the brake disc surface image to be detected, constructing a category vector of each type of pixels, acquiring the overall contrast of each type of pixels according to the category vector, acquiring the significance of each type of pixels according to the overall contrast, the number of characteristic points and distance indexes of each type of pixels, acquiring a defect area according to the significance, performing clockwise sliding on edge pixels by using a sliding window, fitting straight lines to the edge pixels in the sliding window, acquiring an angle index of each edge pixel according to an included angle between the fitted straight line and a horizontal positive direction, acquiring an angle index sequence of each edge pixel, acquiring a confidence to be selected of each edge pixel according to the angle index sequence, selecting a best edge pixel to be fitted candidate circle according to the confidence to be selected, and acquiring a brake disc form index according to the candidate circle, and obtaining the appearance defect index of the brake disc according to the significance of the defect area and the form index of the brake disc, and evaluating the quality of the appearance defect of the brake disc to be detected by using the defect index, wherein the method is accurate and efficient.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for detecting appearance defects of a brake disc according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
An embodiment of a method for detecting appearance defects of a brake disc according to the present invention is shown in fig. 1, and includes:
the method comprises the following steps: acquiring a surface image of a brake disc to be detected;
the method comprises the steps of collecting images of a brake disc to be detected through image collection equipment, detecting and analyzing the appearance of the brake disc according to the defect conditions, and extracting surface characteristic parameters to serve as reference data of defect detection of the brake disc.
The acquisition equipment carries out image acquisition on the brake disc to be detected as reference data for detecting the defects of the brake disc.
And acquiring the brake disc to be detected through image acquisition equipment to acquire image data to be analyzed, and detecting the appearance defect condition of the brake disc to be detected. The image acquisition equipment comprises a camera, a light source, a detection table and the like, an implementer can set the equipment according to actual conditions, the camera is arranged right above the brake disc to be detected, and the overlooking visual angle acquires an orthographic image of the surface of the brake disc and serves as basic image data for subsequent brake disc detection.
The method considers that the surface of the automobile brake disc has some inherent holes, characters of various types and the like, and reduces the attention of irrelevant areas in order to avoid the influence of the inherent grooves, holes and the like on the surface defect detection of the brake disc.
In this embodiment, the semantic segmentation network is used to detect the inherent information (grooves, holes, characters, etc.) on the surface of the brake disc, and obtain a semantic segmentation effect map of the inherent information on the surface of the brake disc, where the tag data of the semantic segmentation network is: setting the pixel value of a pixel point in an inherent characteristic region on the surface of the brake disc to be 1, setting the pixel values of other background pixel points to be 0, training the network through label data and image data, and taking a loss function supervised by network training as a cross entropy loss function; and performing pixel value negation on the acquired brake disc inherent characteristic region semantic segmentation effect graph to acquire a corresponding mask image, wherein the pixel values of pixels in the inherent characteristic region on the surface of the brake disc of the mask image are 0, and the pixel values of other pixels are 1. And multiplying the mask image and the brake disc surface image to be detected, and taking the obtained brake disc surface image after processing as a reference image for extracting the brake disc surface features, wherein the reference image surface comprises an ROI (region of interest) extracted from the brake disc surface features.
Step two: clustering interested region pixel points of a surface image of a brake disc to be detected, and constructing a category vector of each type of pixel points according to a coordinate mean value and a gray mean value of each type of pixel points and a hue mean value, a saturation mean value and a brightness mean value of the type of pixel points in an HSV space; obtaining the overall contrast of each type of pixel points according to the category vector of each type of pixel points;
the step aims to classify the pixel points of the ROI area to obtain the characteristic information of each type of pixel points.
Firstly, graying a brake disc reference image to obtain a corresponding gray image, and carrying out cluster analysis on the gray image based on gray values of pixel points to obtain N categories of pixel points.
Wherein, the category vector of each type of pixel point is as follows:
v i ＝[H i ,S i ,V i ,x i ,y i ,g i ]
in the formula, v i Is a category vector of i-th class pixels, H i ,S i ,V i Respectively the hue mean, saturation mean and brightness mean, x, of all pixel points in the category i after color space conversion i ,y i Are respectively the coordinate mean value, g, of all the pixel points in the category i i The gray level mean value of all pixel points in the category i is obtained;
wherein, the overall contrast of each type of pixel point is as follows:
in the formula, | v i -v n II represents the Euclidean distance between the category vector of the ith pixel point and the category vectors of each other pixel point, C i And representing the integral contrast of the i-th pixel point, and detecting the contrast of each category in the integral image by utilizing the integral contrast.
Step three: obtaining feature points in each type of pixel points by utilizing angular point detection; obtaining the distance index of each type of pixel points according to the distance between the center of each type of pixel points and the center of the neighborhood type pixel points;
the purpose of this step is to detect the characteristic point in each type of pixel and the distance between each type of pixel and other pixels as characteristic parameters.
The feature points in each type of pixel points can be extracted through Harris corner detection and SIFT corner detection algorithms.
The method for calculating the distance index of each type of pixel point comprises the following steps:
in the formula, L i Is the distance index of the ith pixel point, M is the number of neighborhood type pixel points of the ith pixel point, and D i,m For the spatial distance from the center point of the ith pixel point to the center point of the neighborhood category m, the distance calculation methods are many: hamming distance, euclidean distance, etc.
Step four: obtaining the significance of each type of pixel points according to the overall contrast, the number of the characteristic points and the distance index of each type of pixel points, and taking the pixel point type with the significance greater than a threshold value as a defect area;
the purpose of this step is to detect the defective area according to the pixel point characteristics.
The method for calculating the saliency of each type of pixel point comprises the following steps:
in the formula, Q i Is the number of characteristic points, X, in the i-th class pixel point i The greater the significance of the i-th type pixel point is, the higher the probability that the corresponding area has defects is.
In this embodiment, a defect detection model is established according to the significance of each type of pixel: r is i ＝1-exp(-X i ) The larger the function value of the defect detection model is, the higher the defect degree is; and further setting a threshold value for the confidence coefficient of the defect, wherein the type of the defect detection model with the function value higher than 0.7 is used as a defect area, so that the surface defect of the brake disc is detected, and the significance of each defect area is obtained for analyzing the appearance defect condition of the brake disc.
Step five: acquiring edge pixel points of a brake disc, performing clockwise sliding on the edge pixel points by using a sliding window, fitting straight lines to the edge pixel points in the sliding window, and obtaining an angle index of each edge pixel point according to included angles between the fitted straight lines and the positive horizontal direction; obtaining an angle index sequence of each edge pixel point according to the angle index of each edge pixel point and the angle indexes of the adjacent pixel points in front of and behind the edge pixel point;
the purpose of this step is, detect brake disc edge profile to extract the information characteristic of marginal pixel.
According to the embodiment, the outline of the appearance edge of the brake disc is extracted through the edge detection operator to obtain the overall outline information of the brake disc, the edge detection operator comprises a plurality of canny operators, sobel operators and the like, and an implementer can select the edge detection operator by himself.
The method for acquiring the angle index of each edge pixel point comprises the following steps:
(1) sliding analysis is carried out on the edge pixel points of the brake disc outline in the reference image by using a sliding window, the window size implementer can set the window size by himself, the window size is set to be 5 x 5, the window slides by taking each edge pixel point as the center, the implementer can set the initial sliding position of the window, one edge point is selected as the center point of the initial sliding window, and the sliding direction of the window is clockwise sliding along the brake disc edge outline;
(2) and fitting a straight line segment to the edge pixel points contained in the current window every time of sliding, taking the included angle between the straight line segment and the positive horizontal direction as the angle index of the edge pixel points in the center of the current window, and acquiring the angle index of each edge pixel point according to the mode.
Wherein, the angle index sequence of each edge pixel point is as follows: the angle indexes of z edge pixels before and after the edge pixel c are obtained, 2z +1 angle indexes are used as the angle index sequence of the edge pixel c, z is set by an experience value implementer, and in the embodiment, z is 5.
Step six: obtaining the confidence coefficient to be selected of each edge pixel point according to the variance of the angle index sequence of each edge pixel point and the maximum value and the minimum value of the angle index sequence; selecting an optimal edge pixel point according to the confidence coefficient to be selected of each edge pixel point, fitting a candidate circle through the optimal edge pixel point, and obtaining a brake disc form index according to the candidate circle;
the purpose of the step is to select the optimal edge pixel point to fit the candidate circle to obtain the morphological index of the brake disc.
The method for calculating the confidence to be selected of each edge pixel point comprises the following steps:
in the formula, σ c The variance of the angle index sequence corresponding to the edge pixel c,
and the confidence coefficient to be selected of the edge pixel point c is obtained. Obtaining all the pixels to be treated of the contour edge of the brake disc according to the method of the embodimentSelecting confidence coefficient, normalizing each confidence coefficient to be selected to ensure that the function value is in [0,1 ]]And thus, the confidence to be selected of the pixel points at the edge of the contour of the brake disc can be obtained.
The method for selecting the optimal edge pixel point comprises the following steps:
the confidence degrees to be selected of the edge pixel points of the brake disc contour are arranged according to the sequence from large to small to obtain a confidence degree sequence to be selected, then the first S edge points are selected as the optimal edge points to be used for fitting candidate circles, the accuracy of detection of the brake disc contour form is improved, S is an empirical value and can be set by an implementer, and in the embodiment, S is 10.
The method for calculating the brake disc form index comprises the following steps:
in the formula, U is the number of pixel points at the contour edge of the brake disc, delta is the morphological index of the brake disc to be detected, and the function value is normalized to ensure that the function value is in [0,1 ]]For detecting the appearance of the brake disc comprehensively and realizing the comprehensive detection of the appearance defects of the brake disc, x u ，y u And (3) coordinates of edge pixel points used for fitting the candidate circle, wherein a and b are coordinates of the candidate circle, and r is the radius of the candidate circle.
It should be noted that, in consideration of the fact that the appearance contour of the automobile brake disc is circular, in this embodiment, the information of the candidate circle of the brake disc is obtained based on a circle equation and an edge pixel point set, where the circle equation is: (x-a)2+ (y-b)2 ═ r, where (a, b) is the center of a circle and r is the radius of the circle, then the embodiment will obtain each candidate circle based on the edge pixel point set, the embodiment will fit the candidate circle through the edge points, the fitting method is the least square method, the candidate circle is fitted, the circle containing the most edge points is taken as the candidate circle, and the corresponding candidate circle parameters (a, b, r) are obtained.
In the traditional fitting process, a plurality of points are randomly selected from the edge pixel point set for fitting the circle, and in order to improve the accurate detection of the contour form of the brake disc, a pixel point candidate confidence coefficient analysis model is established and used for analyzing each pixel point in the edge pixel point set, so that the pixel points capable of accurately evaluating the contour form characteristics of the brake disc are screened out, and the selection precision of the candidate circle is ensured.
Step seven: and obtaining an appearance defect index of the brake disc according to the significance of the defect area and the form index of the brake disc, and evaluating the quality of the appearance defect of the brake disc to be detected by using the defect index.
The purpose of the step is to establish a defect detection model for each characteristic index, quantitatively analyze the appearance defect condition of the brake disc and realize automatic detection of the surface defect of the brake disc.
The method for calculating the appearance defect index of the brake disc comprises the following steps:
wherein A is the number of defective regions, obtained by the step four, X a Q is an index of the apparent defect of the brake disc to be detected, the higher the index value is, the more serious the defect condition of the surface appearance of the brake disc to be detected is considered, the defect detection model is normalized, and the function value is ensured to be in the value of 0,1]So as to visually acquire the appearance defect condition of the brake disc.
The method for evaluating the quality of the appearance defects of the brake disc to be detected comprises the following steps: when the appearance defect index Q of the brake disc to be detected is greater than 0.5, the system gives out early warning, the appearance defect condition of the brake disc is considered to be serious, and the surface of the brake disc needs to be processed again.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (9)
1. A method for detecting cosmetic defects of a brake disc, comprising:
acquiring a surface image of a brake disc to be detected;
clustering interested region pixel points of a surface image of a brake disc to be detected, and constructing a category vector of each type of pixel points according to a coordinate mean value and a gray mean value of each type of pixel points and a hue mean value, a saturation mean value and a brightness mean value of the type of pixel points in an HSV space;
obtaining the integral contrast of each type of pixel points according to the type vector of each type of pixel points;
obtaining characteristic points in each type of pixel points by utilizing angular point detection;
obtaining the distance index of each type of pixel points according to the distance between the center of each type of pixel points and the center of the neighborhood type pixel points;
obtaining the significance of each type of pixel points according to the overall contrast, the number of the characteristic points and the distance index of each type of pixel points, and taking the pixel point type with the significance larger than a threshold value as a defect area;
obtaining edge pixel points of a brake disc, performing clockwise sliding on the edge pixel points by using a sliding window, fitting straight lines to the edge pixel points in the sliding window, and obtaining an angle index of each edge pixel point according to an included angle between the fitted straight lines and the positive horizontal direction;
obtaining an angle index sequence of each edge pixel point according to the angle index of each edge pixel point and the angle indexes of the adjacent pixel points in front of and behind the edge pixel point;
obtaining the confidence coefficient to be selected of each edge pixel point according to the variance of the angle index sequence of each edge pixel point and the maximum value and the minimum value of the angle index sequence;
selecting an optimal edge pixel point according to the confidence coefficient to be selected of each edge pixel point, fitting a candidate circle through the optimal edge pixel point, and obtaining a brake disc form index according to the candidate circle;
and obtaining an appearance defect index of the brake disc according to the significance of the defect area and the shape index of the brake disc, and evaluating the quality of the appearance defect of the brake disc to be detected by using the defect index.
2. The method for detecting appearance defects of brake discs as claimed in claim 1, wherein the overall contrast of each type of pixel is calculated by:
calculating Euclidean distances between the category vector of each type of pixel point and the category vectors of other each type of pixel points;
and taking the sum of the Euclidean distances as the integral contrast of the pixel points.
3. The method for detecting appearance defects of brake discs as claimed in claim 2, wherein the significance of each type of pixel point is calculated by:
in the formula, X i Significance, Q, of the ith pixel i Is the number of characteristic points, L, in the i-th class pixel point i Is a distance index of the i-th type pixel point, C i The integral contrast of the ith pixel point is obtained.
4. The method for detecting appearance defects of brake discs as claimed in claim 1, wherein the angle index sequence of each edge pixel point is obtained by:
obtaining an angle index of each edge pixel point;
obtaining angle indexes of z pixel points in front of the edge pixel point and angle indexes of z pixel points behind the edge pixel point, wherein z is an empirical value;
and taking the 2z +1 angle index as an angle index sequence of each edge pixel point.
5. A method for detecting apparent defects on a brake disc according to claim 4, wherein said confidence level to be selected for each edge pixel is calculated by:
in the formula (I), the compound is shown in the specification,
is the confidence coefficient, sigma, to be selected of the edge pixel point c c Variance, gamma, of the sequence of angle indices for the edge pixel c max Is the maximum value, gamma, in the angle index sequence of the edge pixel point c min And the minimum value in the angle index sequence of the edge pixel point c is obtained.
6. A method for detecting apparent defects on a brake disc according to claim 5, wherein the optimal edge pixels are selected by:
and sequencing the confidence degrees to be selected of all the edge pixel points in a descending order, and taking the first S edge pixel points as the optimal edge pixel points, wherein S is an empirical value.
7. A method for detecting apparent defects of a brake disc according to claim 1, wherein the rotor contour index is calculated as follows:
in the formula, U is the number of pixel points at the edge of the outline of the brake disc, delta is the outline form index of the brake disc to be detected, and x u ，y u The coordinates of the edge pixel points are shown, a and b are the coordinates of the center of a fitting circle, and r is the radius of the fitting circle.
8. The method for detecting the appearance defect of the brake disc according to claim 7, wherein the method for calculating the appearance defect index of the brake disc comprises the following steps:
wherein A is the number of defective regions, X a And Q is the significance of the a-th defect area, and is the index of the appearance defect of the brake disc to be detected.
9. The method for detecting the appearance defects of the brake disc according to claim 8, wherein the method for evaluating the quality of the appearance defects of the brake disc to be detected by using the defect indexes comprises the following steps:
and when the appearance defect index Q of the brake disc to be detected is greater than 0.5, the appearance defect condition of the brake disc to be detected is serious, and the surface of the brake disc to be detected needs to be processed again.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210473779.3A CN114926410A (en) | 2022-04-29 | 2022-04-29 | Method for detecting appearance defects of brake disc |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210473779.3A CN114926410A (en) | 2022-04-29 | 2022-04-29 | Method for detecting appearance defects of brake disc |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114926410A true CN114926410A (en) | 2022-08-19 |
Family
ID=82806707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210473779.3A Pending CN114926410A (en) | 2022-04-29 | 2022-04-29 | Method for detecting appearance defects of brake disc |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114926410A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115496918A (en) * | 2022-11-16 | 2022-12-20 | 山东高速股份有限公司 | Method and system for detecting abnormal highway conditions based on computer vision |
| CN115496762A (en) * | 2022-11-21 | 2022-12-20 | 深圳市富安娜家居用品股份有限公司 | Textile technology-based dyeing defect identification method |
| CN115631199A (en) * | 2022-12-21 | 2023-01-20 | 深圳新视智科技术有限公司 | Pin needle defect detection method, device, equipment and storage medium |
| CN115841489A (en) * | 2023-02-21 | 2023-03-24 | 华至云链科技(苏州)有限公司 | Intelligent point inspection method and platform |
| CN116958125A (en) * | 2023-09-18 | 2023-10-27 | 惠州市鑫晖源科技有限公司 | Electronic contest host power supply element defect visual detection method based on image processing |
-
2022
- 2022-04-29 CN CN202210473779.3A patent/CN114926410A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115496918A (en) * | 2022-11-16 | 2022-12-20 | 山东高速股份有限公司 | Method and system for detecting abnormal highway conditions based on computer vision |
| CN115496762A (en) * | 2022-11-21 | 2022-12-20 | 深圳市富安娜家居用品股份有限公司 | Textile technology-based dyeing defect identification method |
| CN115496762B (en) * | 2022-11-21 | 2023-01-24 | 深圳市富安娜家居用品股份有限公司 | Textile technology-based dyeing defect identification method |
| CN115631199A (en) * | 2022-12-21 | 2023-01-20 | 深圳新视智科技术有限公司 | Pin needle defect detection method, device, equipment and storage medium |
| CN115841489A (en) * | 2023-02-21 | 2023-03-24 | 华至云链科技(苏州)有限公司 | Intelligent point inspection method and platform |
| CN116958125A (en) * | 2023-09-18 | 2023-10-27 | 惠州市鑫晖源科技有限公司 | Electronic contest host power supply element defect visual detection method based on image processing |
| CN116958125B (en) * | 2023-09-18 | 2023-12-26 | 惠州市鑫晖源科技有限公司 | Electronic contest host power supply element defect visual detection method based on image processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115082683B (en) | Injection molding defect detection method based on image processing | |
| CN114926410A (en) | Method for detecting appearance defects of brake disc | |
| CN110334706B (en) | Image target identification method and device | |
| Oliveira et al. | Automatic road crack detection and characterization | |
| CN109377485B (en) | Machine vision detection method for instant noodle packaging defects | |
| CN109816644B (en) | Bearing defect automatic detection system based on multi-angle light source image | |
| CN109658402B (en) | Automatic detection method for geometric dimension of industrial profile based on computer vision imaging | |
| CN116205919B (en) | Hardware part production quality detection method and system based on artificial intelligence | |
| CN109444169B (en) | Bearing defect detection method and system | |
| CN110210477B (en) | Digital instrument reading identification method | |
| CN114972356B (en) | Plastic product surface defect detection and identification method and system | |
| CN115082419A (en) | Blow-molded luggage production defect detection method | |
| CN108734131B (en) | Method for detecting symmetry of traffic sign in image | |
| CN109658391B (en) | Circle radius measuring method based on contour merging and convex hull fitting | |
| CN113935666B (en) | Building decoration wall tile abnormity evaluation method based on image processing | |
| CN103295013A (en) | Pared area based single-image shadow detection method | |
| CN112907519A (en) | Metal curved surface defect analysis system and method based on deep learning | |
| CN111161237A (en) | Fruit and vegetable surface quality detection method, storage medium and sorting device thereof | |
| CN112067618A (en) | Product quality defect acquisition and analysis system based on imaging | |
| CN107516315B (en) | Tunneling machine slag tapping monitoring method based on machine vision | |
| CN114820625A (en) | Automobile top block defect detection method | |
| CN111476804A (en) | Method, device and equipment for efficiently segmenting carrier roller image and storage medium | |
| CN115100196A (en) | Method for evaluating derusting effect of stamping part based on image segmentation | |
| CN114155226A (en) | Micro defect edge calculation method | |
| CN111815575B (en) | Bearing steel ball part detection method based on machine vision |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20220819 |