CN116818778A - Rapid and intelligent detection method and system for automobile parts - Google Patents

Rapid and intelligent detection method and system for automobile parts Download PDF

Info

Publication number
CN116818778A
CN116818778A CN202311114198.1A CN202311114198A CN116818778A CN 116818778 A CN116818778 A CN 116818778A CN 202311114198 A CN202311114198 A CN 202311114198A CN 116818778 A CN116818778 A CN 116818778A
Authority
CN
China
Prior art keywords
detected
standard
curve
image
contour curve
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.)
Granted
Application number
CN202311114198.1A
Other languages
Chinese (zh)
Other versions
CN116818778B (en
Inventor
闻敏
徐灿彬
姜萍
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.)
Jingjiang Hengyou Auto Parts Manufacturing Co ltd
Original Assignee
Jingjiang Hengyou Auto Parts Manufacturing Co ltd
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 Jingjiang Hengyou Auto Parts Manufacturing Co ltd filed Critical Jingjiang Hengyou Auto Parts Manufacturing Co ltd
Priority to CN202311114198.1A priority Critical patent/CN116818778B/en
Publication of CN116818778A publication Critical patent/CN116818778A/en
Application granted granted Critical
Publication of CN116818778B publication Critical patent/CN116818778B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • 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 by the use of local operators
    • 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
    • G06T7/0006—Industrial image inspection using a design-rule based approach
    • 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
    • G06T7/001—Industrial image inspection using an image reference approach
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • 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
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
    • 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
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G06V10/752—Contour matching
    • 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
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/761—Proximity, similarity or dissimilarity measures
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8854—Grading and classifying of flaws
    • G01N2021/8874—Taking dimensions of defect into account
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
    • 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G06T2207/10012—Stereo images
    • 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
    • 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
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Abstract

The application provides a rapid and intelligent detection method and a system for automobile parts, which relate to the technical field of automobile detection, and the rapid and intelligent detection method for the automobile parts comprises the following steps: respectively scanning an automobile part to be detected and a standard automobile part, and preprocessing and contour extraction are carried out on the obtained image to be detected and the standard image to obtain a contour curve to be detected and a standard contour curve; performing curve similarity calculation based on a curve segmentation similarity matching algorithm; dividing the difference between the contour curve to be detected and the standard contour curve, and calculating the difference; judging whether the automobile part to be detected can be repaired or not, and feeding back a judging result. The application can rapidly detect the automobile parts, greatly improves the detection efficiency, and can judge whether the automobile parts to be detected can be repaired or not; can help the staff on the production line to judge which defects can be repaired fast, thereby improving the production efficiency and reducing the production cost.

Description

Rapid and intelligent detection method and system for automobile parts
Technical Field
The application relates to the technical field of automobile detection, in particular to a rapid intelligent detection method and system for automobile parts.
Background
In modern society, automobiles become main transportation means for home travel, and the safety performance of the automobiles is also highly concerned by vast users. With the development of the automobile industry, it is increasingly recognized that the performance and quality of parts are one of the most important factors in determining the performance and quality of the entire automobile. In the face of increased market competition and consumer expectations for product quality, manufacturers and component suppliers must try to reduce the occurrence of unacceptable components. There is a need for them to use more efficient and reliable detection means to ensure the quality of the automobile parts.
The basic feature of the machine vision system is to increase the flexibility and automation level of production. Machine vision is often used instead of human vision in some hazardous work environments or where human vision is difficult to meet. With the continuous development of technology, machine vision is increasingly widely applied to automobile part detection, and is used for detecting whether a part has defects. However, at present, when detecting automobile parts through machine vision, defects of the parts can only be effectively detected, but for some defects, it is difficult to accurately identify the sizes of the defects, and for some defects, repair can be completed, if repairable defects cannot be identified, the parts with repairable defects are erroneously judged to be unqualified, and therefore opportunities and cost for repairing the defects are wasted.
For the problems in the related art, no effective solution has been proposed at present.
Disclosure of Invention
In view of the above, the present application provides a method and a system for fast and intelligent detection of automobile parts, which solve the above-mentioned problems that failure to identify repairable defects may cause the parts with repairable defects to be erroneously determined as being unqualified, thereby wasting opportunities and costs for repairing the defects.
In order to solve the problems, the application adopts the following specific technical scheme:
according to one aspect of the present application, there is provided a rapid intelligent detection method for an automobile part, the intelligent detection method comprising the steps of:
s1, respectively scanning an automobile part to be detected and a standard automobile part by a 3D scanner to obtain an image to be detected and a standard image;
s2, preprocessing and contour extraction are respectively carried out on the obtained image to be detected and the standard image to obtain a contour curve to be detected and a standard contour curve;
s3, calculating the curve similarity of the contour curve to be detected and the standard contour curve based on a curve segmentation similarity matching algorithm, and determining the difference between the contour curve to be detected and the standard contour curve;
s4, dividing the difference between the contour curve to be detected and the standard contour curve, and calculating the difference;
s5, judging whether the automobile part to be detected can be repaired or not according to the difference, and feeding back a judging result.
Preferably, the preprocessing and contour extraction are performed on the obtained image to be detected and the standard image respectively, and the obtaining of the contour curve to be detected and the standard contour curve includes the following steps:
s21, respectively carrying out image denoising, image filtering and image enhancement processing on an image to be detected and a standard image;
s22, respectively carrying out edge detection on the processed image to be detected and the standard image, and fitting the detected edges by adopting a broken line approximation method;
s23, measuring fitting edges of the image to be detected and the standard image respectively based on the closeness in the format tower shaping rule to obtain a closed relation between the edges, and solving a preset cost function through a minimum weight optimal matching algorithm to obtain closed contours of the image to be detected and the standard image;
s24, adopting a B spline curve method to carry out smoothing treatment on the closed contour of the image to be detected and the closed contour of the standard image respectively, and obtaining a contour curve to be detected and a standard contour curve.
Preferably, the measuring the closeness in the format tower shape rule based on the fitting edges of the image to be detected and the standard image respectively to obtain a closed relation between the edges, and solving a preset cost function through a minimum weight optimal matching algorithm to obtain closed contours of the image to be detected and the standard image comprises the following steps:
s231, taking the closure in the format tower shape rule as a judging target, measuring fitting edges of the image to be detected and the standard image respectively, and judging whether the fitted edge lines belong to the same contour or not, wherein the measuring mode comprises smoothness measurement, jerkiness measurement and notch measurement caused by shielding;
s232, defining a boundary cost function by taking the maximum probability of the same contour between edges connected end to end in the contour as a target;
s233, solving the defined cost function by utilizing a minimum weight optimal matching algorithm.
Preferably, the curve similarity calculation is performed on the profile curve to be detected and the standard profile curve based on the curve segment similarity matching algorithm, and the determining of the difference between the profile curve to be detected and the standard profile curve includes the following steps:
s31, selecting a common reference point from the contour curve to be detected and the standard contour curve as a reference point, and mapping the contour curve to be detected and the standard contour curve into a two-dimensional coordinate system respectively;
s32, respectively carrying out sectional treatment on the contour curve to be detected and the standard contour curve according to the visual key points;
s33, respectively extracting features of all segments on the contour curve to be detected and the standard contour curve in a clockwise direction according to the selected reference points, wherein the features comprise lengths and angles;
s34, combining the feature values extracted by each segment to form a feature sequence;
s35, calculating correlation coefficients of the contour curve to be detected and the standard contour curve according to the extracted feature sequence;
s36, judging whether the profile curve to be detected and the standard profile curve have differences according to the calculated correlation coefficient.
Preferably, the segmenting processing of the contour curve to be detected and the standard contour curve according to the visual key points includes the following steps:
s321, respectively carrying out vertex domain calculation on each point on the contour curve to be detected and the standard contour curve;
s322, calculating the curvature value of each point according to the vertex domain of the point;
s323, selecting a maximum value point as a segmentation point according to the calculated curvature value.
Preferably, the calculation formula for calculating the correlation coefficient of the profile curve to be detected and the standard profile curve according to the extracted feature sequence is as follows:
in the method, in the process of the application,rrepresenting the correlation coefficient of the contour curve to be detected and the standard contour curve;
t f andk f respectively representing the feature sequence of the contour curve to be detected and the feature sequence of the standard contour curvefCharacteristic values of the individual segments;
and->And respectively representing the expected sectional values of the profile characteristic sequence to be detected and the standard profile characteristic sequence.
Preferably, the process of dividing the difference between the contour curve to be detected and the standard contour curve and calculating the size of the difference includes the following steps:
s41, determining segments with differences in the profile curve to be detected and the standard profile curve according to the calculated correlation coefficient;
s42, segmenting the difference between the contour curve to be detected and the standard contour curve to obtain a difference region;
s43, determining the length and the width of the difference region according to the position of the difference region in the two-dimensional coordinate system;
s44, determining a rectangular boundary according to the length and the width of the difference region;
s45, calculating the area of the difference region by using a Monte Carlo method to obtain the area of the difference region.
Preferably, the calculating the area of the difference region by using the monte carlo method to obtain the area size of the difference region includes the following steps:
s451, generating random points in the determined rectangular boundary by using a uniform random number generator;
s452, counting the number of the generated random points and the number falling in the difference area;
s453, calculating the area of the difference region according to the difference region area calculation formula.
Preferably, the step of judging whether the automobile part to be detected can be repaired according to the difference and feeding back the judgment result comprises the following steps:
s51, comparing and judging the difference between the automobile part to be detected and the standard automobile part with a preset threshold value;
s52, if the difference value is smaller than or equal to a preset threshold value, the automobile part to be detected is repairable, and if the difference value is larger than the preset threshold value, the automobile part to be detected is not repairable;
and S53, feeding back the judging result to the automobile part detecting personnel.
According to another aspect of the present application, there is provided a rapid intelligent detection system for an automotive part, the intelligent detection system comprising: the device comprises a component scanning module, a feature extraction module, a similarity calculation module, a difference calculation module and a judgment feedback module;
the component scanning module is used for scanning the automobile component to be detected and the standard automobile component through the 3D scanner respectively to obtain an image to be detected and a standard image;
the feature extraction module is used for respectively preprocessing and extracting the profile of the obtained image to be detected and the standard image to obtain a profile curve to be detected and a standard profile curve;
the similarity calculation module is used for calculating the similarity of the curve to be detected and the standard contour curve based on a curve segmentation similarity matching algorithm and determining the difference between the curve to be detected and the standard contour curve;
the difference calculation module is used for dividing the difference between the contour curve to be detected and the standard contour curve, and calculating the difference;
and the judging feedback module is used for judging whether the automobile part to be detected can be repaired or not according to the difference and feeding back the judging result.
Compared with the prior art, the application provides the rapid intelligent detection method and system for the automobile parts, which have the following beneficial effects:
(1) According to the application, the 3D scanner and the curve segment similarity matching algorithm are used, so that the automobile part can be rapidly detected, the detection efficiency is greatly improved, the similarity between the contour curve to be detected and the standard contour curve can be accurately calculated by the detection method based on the curve segment similarity matching algorithm, the detection accuracy is improved, and whether the automobile part to be detected can be repaired or not can be judged by calculating the difference; can help the staff on the production line to judge which defects can be repaired fast, thereby improving the production efficiency and reducing the production cost.
(2) The application carries out denoising, filtering and enhancing treatment on the image to be detected and the standard image, can improve the quality and definition of the image, carries out edge detection on the processed image by adopting an edge detection algorithm, carries out fitting on edges by adopting a broken line approximation method, thereby better extracting the edge information of the image, measures the fitted edges based on the closeness in a format tower shaping rule, can judge whether the fitted edge lines belong to the same contour, accurately extracts the closed contour of the image to be detected and the standard image, solves the cost function by defining a boundary cost function and utilizing a minimum weight optimal matching algorithm, can accurately determine the corresponding relation between the closed contour of the image to be detected and the closed contour of the standard image, carries out smoothing treatment on the closed contour by adopting a B spline curve method, and can further improve the fineness and smoothness of the contour.
(3) According to the method, the vertex domain and the curvature value of each point on the profile curve to be detected and the standard profile curve are calculated, the maximum point can be selected as a segmentation point, so that key points are extracted, subsequent similarity calculation and difference analysis are facilitated, the profile curve to be detected and the standard profile curve are divided into a plurality of sections, the curves in each section have good continuity and similarity, the similarity calculation and the difference analysis of the curves can be decomposed into a plurality of small sections for calculation, the complexity of the problem is simplified, and the correlation coefficient of the profile curve to be detected and the standard profile curve is calculated; by calculating the correlation coefficient, the similarity between the curves can be quantified, and whether the curves have differences or not can be further judged.
(4) According to the method, the length and the width of the difference region are determined according to the position of the difference region in the two-dimensional coordinate system, the area of the difference region is calculated by using the Monte Carlo method, the size of the difference region can be quantized to obtain a specific numerical value, the specific numerical value is used for measuring the difference degree between the contour curve to be detected and the standard contour curve, the area of the difference region is calculated according to the area calculation formula of the difference region, the area of the difference region can be estimated by using the Monte Carlo method in a random sampling mode, the calculation efficiency and the accuracy are high, and the method is beneficial to accurately calculating the difference between the contour curve to be detected and the standard contour curve.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
FIG. 1 is a flow chart of a method for rapid intelligent detection of an automotive component according to an embodiment of the application;
fig. 2 is a functional block diagram of a rapid intelligent detection system for automotive components according to an embodiment of the present application.
In the figure:
1. a component scanning module; 2. a feature extraction module; 3. a similarity calculation module; 4. a difference calculation module; 5. and a judgment feedback module.
Detailed Description
In order to make the technical solution of the present application better understood by those skilled in the art, the technical solution of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments of the present application. All other embodiments, based on the embodiments of the application, which would be apparent to one of ordinary skill in the art without undue burden are intended to be within the scope of the application.
According to the embodiment of the application, a rapid intelligent detection method and a rapid intelligent detection system for automobile parts are provided.
The application will be further described with reference to the accompanying drawings and detailed description, as shown in fig. 1, according to an embodiment of the application, there is provided a method and a system for rapid and intelligent detection of an automobile part, the intelligent detection method comprising the following steps:
s1, respectively scanning an automobile part to be detected and a standard automobile part by a 3D scanner to obtain an image to be detected and a standard image;
it should be noted that, the automobile parts to be detected and the standard automobile parts are placed in the scanning range of the 3D scanner, then the scanner is started to scan, and the 3D scanner scans the automobile parts to be detected and the standard automobile parts at multiple angles.
S2, preprocessing and contour extraction are respectively carried out on the obtained image to be detected and the standard image, and a contour curve to be detected and a standard contour curve are obtained.
As a preferred embodiment, the preprocessing and contour extraction are performed on the obtained image to be detected and the standard image respectively, and the obtaining of the contour curve to be detected and the standard contour curve includes the following steps:
s21, respectively carrying out image denoising, image filtering and image enhancement processing on the image to be detected and the standard image.
It should be noted that image denoising refers to removing noise in an image by using various methods to improve the quality of the image. Common image denoising methods include median filtering, gaussian filtering, and the like.
Image filtering refers to processing an image using filters to remove or emphasize certain characteristics. Common image filtering methods include low pass filtering (removing high frequency noise), high pass filtering (emphasizing edges and details), and the like.
Image enhancement refers to the use of various methods to enhance the visual effect of an image or to make certain characteristics of an image more apparent. Common image enhancement methods include histogram equalization, contrast enhancement, and the like.
S22, respectively carrying out edge detection on the processed image to be detected and the standard image, and fitting the detected edges by adopting a broken line approximation method.
Specifically, the edge detection of the processed image to be detected and the standard image generally uses Canny edge detection, and the main steps include: the image is subjected to graying processing, and the color image is converted into a gray image. The gray scale image is gaussian filtered to reduce noise in the image. The gradient and direction of the image are calculated, and the edge intensity and direction in the image are found out. And carrying out non-maximum suppression on the edge, and only reserving local maximum points to obtain a thinned edge. And setting a high threshold and a low threshold, and performing edge connection and edge screening to obtain a final edge image.
Fitting the detected edges using a polyline approximation can be achieved by:
and acquiring an edge point set obtained by edge detection. These edge points represent edge locations in the image.
And sequencing the edge point sets, and connecting the points according to a certain sequence to form a continuous edge curve.
And setting the number of the fitted line segments or a maximum error threshold value, and then performing broken line approximation fitting on the edge curve.
S23, measuring fitting edges of the image to be detected and the standard image based on the closeness in the format tower shaping rule, obtaining a closed relation between the edges, and solving a preset cost function through a minimum weight optimal matching algorithm to obtain closed contours of the image to be detected and the standard image.
As a preferred embodiment, the measuring the closeness in the format tower shape rule based on the fitting edges of the image to be detected and the standard image respectively to obtain a closed relation between the edges, and solving a preset cost function through a minimum weight optimal matching algorithm to obtain closed contours of the image to be detected and the standard image comprises the following steps:
s231, taking the closure in the format tower shape rule as a judging target, measuring fitting edges of the image to be detected and the standard image respectively, and judging whether the fitted edge lines belong to the same contour or not, wherein the measuring mode comprises smoothness measurement, jerkiness measurement and notch measurement caused by shielding.
The smoothness measurement means that the smoothness of the edge line is measured using an index such as curvature or curvature change rate. Smooth edge lines generally have less curvature variation. The curvature or curvature change rate between adjacent points on the edge line can be calculated, then the numerical values are statistically analyzed, and the smoothness metric value of the edge line of the image to be detected and the standard image is compared. If the metrics of the two images are similar, the fitted edge lines are considered to have similar smoothness.
The jerkiness measure is to measure the jerkiness of the edge line using an index such as an angular change or a curvature abrupt change. The sharp edge line is typically characterized by a large angular variation. The angular change or curvature jump between adjacent points on the edge line can be calculated, and then these values are statistically analyzed to compare the sharp rotation metric values of the edge line of the image to be detected and the standard image. If the metrics of the two images are similar, the fitted edge lines are considered to have similar jerkiness.
Occlusion-induced notch metrics refer to detecting whether a notch exists in an edge line, which may be due to occlusion of a target. Occlusion induced gaps can be measured by counting the number of breaks or gaps in the edge line. And counting the number of breaks or gaps of the edge lines of the image to be detected and the standard image, and comparing the number of breaks or gaps. If the measurement values of the two images are similar, the fitted edge line is considered to have a similar shielding notch.
S232, defining a boundary cost function by taking the maximum probability of the same contour between edges connected end to end in the contour as a target, wherein the calculation formula of the boundary cost function is as follows:
in the method, in the process of the application,β(A) Representing a defined boundary cost function;
Acandidate boundaries representing a closed contour;
Urepresenting pixel points in the image;
P(A) Representing the sum of probabilities that the end-to-end fitting edges do not belong to the same contour;
P(e i e j ) Representing edgese i Ande j belonging to the same closed contour probability;
R(B) Candidate boundaries representing closed contoursAAn included region;
S(U) Representing pixel pointsUIs a significant value of (2);
s233, solving the defined cost function by utilizing a minimum weight optimal matching algorithm.
It should be noted that the minimum weight optimal matching algorithm, also called as hungarian algorithm or Kuhn-Munkres algorithm, is a classical algorithm for solving the optimal matching problem. The goal is to find a set of optimal matches in the cost matrix so that the total matching cost is minimized.
S24, adopting a B spline curve method to carry out smoothing treatment on the closed contour of the image to be detected and the closed contour of the standard image respectively, and obtaining a contour curve to be detected and a standard contour curve.
It should be noted that, the smoothing processing of the closed contour of the image to be detected and the standard image by using the B-spline curve method specifically includes the following steps:
and acquiring a closed contour point set of the image to be detected and the standard image. These points represent the contour positions in the image.
B spline curve fitting is carried out on the closed contour point set. B-spline curves are a commonly used curve fitting method, where the shape of the curve can be defined by control points and node vectors.
Control points and node vectors of the B-spline curve are determined. Control points are points for controlling the shape of the curve, and node vectors are points for defining the distribution of the curve in the parameter space.
And performing curve fitting on the closed contour point sets of the image to be detected and the standard image by using a fitting algorithm of the B spline curve. In the fitting process, a fitted curve is calculated according to the control points and the node vectors.
And adjusting the smoothness of the B spline curve according to the requirement. The degree of smoothness of the B-spline curve can be adjusted by controlling the number of points and the distribution of node vectors. Increasing the number of control points or adjusting the distribution of node vectors can increase the smoothness of the curve to obtain a contour curve to be detected and a standard contour curve.
And S3, calculating the curve similarity of the contour curve to be detected and the standard contour curve based on a curve segmentation similarity matching algorithm, and determining the difference between the contour curve to be detected and the standard contour curve.
As a preferred embodiment, the curve similarity calculation is performed on the profile curve to be detected and the standard profile curve based on the curve segment similarity matching algorithm, and the determining of the difference between the profile curve to be detected and the standard profile curve includes the following steps:
s31, selecting a common reference point from the contour curve to be detected and the standard contour curve as a reference point, and mapping the contour curve to be detected and the standard contour curve into a two-dimensional coordinate system respectively.
It should be noted that the contour curve to be detected and the standard contour curve are mapped into a two-dimensional coordinate system. A rectangular coordinate system may be used, and if a rectangular coordinate system is used, the x-coordinate and the y-coordinate of each point on the curve may be used as mapped coordinate values. This allows the shape of the curve to be accurately represented on a two-dimensional plane.
S32, respectively carrying out sectional treatment on the contour curve to be detected and the standard contour curve according to the visual key points;
as a preferred embodiment, the segmentation processing of the contour curve to be detected and the standard contour curve according to the visual key points includes the following steps:
s321, respectively carrying out vertex domain calculation on each point on the contour curve to be detected and the standard contour curve.
It should be noted that each point on the contour curve to be detected and the standard contour curve is traversed, and for each point, its vertex domain is determined. A vertex field refers to a set of points on a curve centered around a current point and may be used to describe the curve segment around the current point. The size of the vertex field may be specifically defined according to a preset radius.
S322, calculating the curvature value of each point according to the vertex domain of the point.
It should be noted that, for each point on the contour curve to be detected and the standard contour curve, a set of points on a section of curve around the point is obtained according to the vertex domain, and the curvature value of the current point can be calculated by using the curvature calculation formula through the set of points. Curvature refers to the degree of curvature of a curve at a point. By calculating the curvature value of each point, the curvature distribution of the curve at each point can be obtained. This allows a more complete description of the degree of curvature and shape characteristics of the curve.
S323, selecting a maximum value point as a segmentation point according to the calculated curvature value.
For the calculated curvature value, the curvature value of each point is traversed. The curvature value of the current point is checked and compared with the curvature values of the adjacent points. If the curvature value of the current point is greater than the curvature value of its neighboring points, then that point is considered a maximum point. All points identified as maximum points are taken as segmentation points. These segmentation points will be used to divide the curve into segments, each segment having different curvature characteristics.
S33, respectively extracting features of all segments on the contour curve to be detected and the standard contour curve in a clockwise direction according to the selected reference points, wherein the features comprise lengths and angles.
It should be noted that the length characteristic may be obtained by calculating the sum of euclidean distances between adjacent points on the segment. The angular characteristic may be obtained by calculating the angle between the start and end points of the segment. The angle may be calculated using an inverse cosine function or a tangent function.
S34, combining the feature values extracted by each segment to form a feature sequence.
S35, calculating correlation coefficients of the contour curve to be detected and the standard contour curve according to the extracted feature sequence.
As a preferred embodiment, the calculation formula for calculating the correlation coefficient of the profile curve to be detected and the standard profile curve according to the extracted feature sequence is as follows:
in the method, in the process of the application,rrepresenting the correlation coefficient of the contour curve to be detected and the standard contour curve;
t f andk f respectively represent the characteristic sequence of the contour curve to be detected and the characteristic sequence of the standard contour curveColumn 1fCharacteristic values of the individual segments;
and->And respectively representing the expected sectional values of the profile characteristic sequence to be detected and the standard profile characteristic sequence.
S36, judging whether the profile curve to be detected and the standard profile curve have differences according to the calculated correlation coefficient.
It should be noted that a correlation coefficient (such as pearson correlation coefficient) can measure the linear correlation between two feature vectors. The correlation coefficient has a value ranging from-1 to 1, wherein-1 represents a complete negative correlation, 1 represents a complete positive correlation, and 0 represents no correlation. If the correlation coefficient is 1 or-1, the correlation coefficient indicates that no difference exists between the contour curve to be detected and the standard contour curve, the automobile part with detection is a qualified part, and if the correlation coefficient is close to 1 or-1, the correlation coefficient indicates that high linear correlation exists between the contour curve to be detected and the standard contour curve, and the difference between the correlation coefficient and the standard contour curve is small, so that the repair can be performed. If the correlation coefficient is close to 0, the linear correlation between the contour curve to be detected and the standard contour curve does not exist, and the difference between the contour curve to be detected and the standard contour curve is large, so that the restoration is difficult.
S4, dividing the difference between the contour curve to be detected and the standard contour curve, and calculating the difference.
As a preferred embodiment, the process of dividing the difference existing between the contour curve to be detected and the standard contour curve and calculating the magnitude of the difference includes the steps of:
s41, determining segments with differences in the profile curve to be detected and the standard profile curve according to the calculated correlation coefficient.
S42, segmenting the difference between the contour curve to be detected and the standard contour curve to obtain a difference region.
S43, determining the length and the width of the difference region according to the position of the difference region in the two-dimensional coordinate system.
S44, determining a rectangular boundary according to the length and the width of the difference region.
S45, calculating the area of the difference region by using a Monte Carlo method to obtain the area of the difference region.
As a preferred embodiment, the calculating the area of the difference region by using the monte carlo method, to obtain the area size of the difference region includes the following steps:
s451, random points are generated in the determined rectangular boundary by using the uniform random number generator.
S452, counting the number of the generated random points and the number falling in the difference area.
S453, calculating the area of the difference region according to the difference region area calculation formula.
It should be noted that, the difference area calculation formula is:
in the method, in the process of the application,srepresenting the area of the difference region;
s full an area representing a rectangular boundary;
countrepresenting the number of random points generated;
in_countrepresenting the number of random points within the region of difference.
S5, judging whether the automobile part to be detected can be repaired or not according to the difference, and feeding back a judging result.
As a preferred embodiment, the step of judging whether the automobile part to be detected can be repaired according to the magnitude of the difference, and feeding back the judgment result includes the following steps:
s51, comparing and judging the difference between the automobile part to be detected and the standard automobile part with a preset threshold value.
And S52, if the difference value is smaller than or equal to a preset threshold value, the automobile part to be detected is repairable, and if the difference value is larger than the preset threshold value, the automobile part to be detected is not repairable.
And S53, feeding back the judging result to the automobile part detecting personnel.
As shown in fig. 2, according to another embodiment of the present application, there is provided a rapid smart detection system for an automobile part, the smart detection system including: the device comprises a component scanning module 1, a feature extraction module 2, a similarity calculation module 3, a difference calculation module 4 and a judgment feedback module 5;
the component scanning module 1 is used for scanning the automobile component to be detected and the standard automobile component through a 3D scanner respectively to obtain an image to be detected and a standard image;
the feature extraction module 2 is used for respectively preprocessing and extracting the profile of the obtained image to be detected and the standard image to obtain a profile curve to be detected and a standard profile curve;
the similarity calculation module 3 is configured to perform curve similarity calculation on a to-be-detected contour curve and a standard contour curve based on a curve segment similarity matching algorithm, and determine a difference between the to-be-detected contour curve and the standard contour curve;
the difference calculation module 4 is used for dividing the difference between the contour curve to be detected and the standard contour curve, and calculating the size of the difference;
and the judging feedback module 5 is used for judging whether the automobile part to be detected can be repaired or not according to the difference and feeding back the judging result.
In summary, by means of the technical scheme, the automobile part can be rapidly detected by using the 3D scanner and the curve segmentation similarity matching algorithm, so that the detection efficiency is greatly improved, the similarity between the contour curve to be detected and the standard contour curve can be accurately calculated by the detection method based on the curve segmentation similarity matching algorithm, the detection accuracy is improved, and whether the automobile part to be detected can be repaired or not can be judged by calculating the difference; the method can help workers on the production line to quickly judge which defects can be repaired, thereby improving the production efficiency and reducing the production cost; the application carries out denoising, filtering and enhancing treatment on the image to be detected and the standard image, can improve the quality and definition of the image, carries out edge detection on the processed image by adopting an edge detection algorithm, carries out fitting on edges by adopting a broken line approximation method, thereby better extracting the edge information of the image, measures the fitted edges based on the closeness in a format tower shape rule, can judge whether the fitted edge lines belong to the same contour, accurately extracts the closed contour of the image to be detected and the standard image, solves the cost function by defining a boundary cost function and utilizing a minimum weight optimal matching algorithm, can accurately determine the corresponding relation between the closed contour of the image to be detected and the closed contour of the standard image, carries out smoothing treatment on the closed contour by adopting a B spline curve method, and can further improve the fineness and smoothness of the contour; according to the method, the vertex domain and the curvature value of each point on the profile curve to be detected and the standard profile curve are calculated, the maximum point can be selected as a segmentation point, so that key points are extracted, subsequent similarity calculation and difference analysis are facilitated, the profile curve to be detected and the standard profile curve are divided into a plurality of sections, the curves in each section have good continuity and similarity, the similarity calculation and the difference analysis of the curves can be decomposed into a plurality of small sections for calculation, the complexity of the problem is simplified, and the correlation coefficient of the profile curve to be detected and the standard profile curve is calculated; the similarity between the curves can be quantified through the calculation of the correlation coefficient, and whether the curves have differences or not is further judged; according to the method, the length and the width of the difference region are determined according to the position of the difference region in the two-dimensional coordinate system, the area of the difference region is calculated by using the Monte Carlo method, the size of the difference region can be quantized to obtain a specific numerical value, the specific numerical value is used for measuring the difference degree between the contour curve to be detected and the standard contour curve, the area of the difference region is calculated according to the area calculation formula of the difference region, the area of the difference region can be estimated by using the Monte Carlo method in a random sampling mode, the calculation efficiency and the accuracy are high, and the method is beneficial to accurately calculating the difference between the contour curve to be detected and the standard contour curve.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the application, and is not meant to limit the scope of the application, but to limit the application to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the application are intended to be included within the scope of the application.

Claims (10)

1. The rapid intelligent detection method for the automobile parts is characterized by comprising the following steps of:
s1, respectively scanning an automobile part to be detected and a standard automobile part by a 3D scanner to obtain an image to be detected and a standard image;
s2, preprocessing and contour extraction are respectively carried out on the obtained image to be detected and the standard image to obtain a contour curve to be detected and a standard contour curve;
s3, calculating the curve similarity of the contour curve to be detected and the standard contour curve based on a curve segmentation similarity matching algorithm, and determining the difference between the contour curve to be detected and the standard contour curve;
s4, dividing the difference between the contour curve to be detected and the standard contour curve, and calculating the difference;
s5, judging whether the automobile part to be detected can be repaired or not according to the difference, and feeding back a judging result.
2. The method for rapid and intelligent detection of automobile parts according to claim 1, wherein the steps of preprocessing and contour extraction are performed on the obtained image to be detected and the standard image respectively to obtain a contour curve to be detected and a standard contour curve, and the steps of:
s21, respectively carrying out image denoising, image filtering and image enhancement processing on an image to be detected and a standard image;
s22, respectively carrying out edge detection on the processed image to be detected and the standard image, and fitting the detected edges by adopting a broken line approximation method;
s23, measuring fitting edges of the image to be detected and the standard image respectively based on the closeness in the format tower shaping rule to obtain a closed relation between the edges, and solving a preset cost function through a minimum weight optimal matching algorithm to obtain closed contours of the image to be detected and the standard image;
s24, adopting a B spline curve method to carry out smoothing treatment on the closed contour of the image to be detected and the closed contour of the standard image respectively, and obtaining a contour curve to be detected and a standard contour curve.
3. The method for rapidly and intelligently detecting the automobile parts according to claim 2, wherein the measuring of the closeness in the format tower-based finishing rule is performed on fitting edges of the image to be detected and the standard image respectively to obtain a closed relation between the edges, and the solving of a preset cost function is performed through a minimum weight optimal matching algorithm to obtain closed outlines of the image to be detected and the standard image comprises the following steps:
s231, taking the closure in the format tower shape rule as a judging target, measuring fitting edges of the image to be detected and the standard image respectively, and judging whether the fitted edge lines belong to the same contour or not, wherein the measuring mode comprises smoothness measurement, jerkiness measurement and notch measurement caused by shielding;
s232, defining a boundary cost function by taking the maximum probability of the same contour between edges connected end to end in the contour as a target;
s233, solving the defined cost function by utilizing a minimum weight optimal matching algorithm.
4. The method for rapidly and intelligently detecting the automobile parts according to claim 1, wherein the curve similarity calculation is performed on the profile curve to be detected and the standard profile curve based on the curve segmentation similarity matching algorithm, and the difference between the profile curve to be detected and the standard profile curve is determined, comprising the following steps:
s31, selecting a common reference point from the contour curve to be detected and the standard contour curve as a reference point, and mapping the contour curve to be detected and the standard contour curve into a two-dimensional coordinate system respectively;
s32, respectively carrying out sectional treatment on the contour curve to be detected and the standard contour curve according to the visual key points;
s33, respectively extracting features of all segments on the contour curve to be detected and the standard contour curve in a clockwise direction according to the selected reference points, wherein the features comprise lengths and angles;
s34, combining the feature values extracted by each segment to form a feature sequence;
s35, calculating correlation coefficients of the contour curve to be detected and the standard contour curve according to the extracted feature sequence;
s36, judging whether the profile curve to be detected and the standard profile curve have differences according to the calculated correlation coefficient.
5. The method for rapid and intelligent detection of automobile parts according to claim 4, wherein the step of respectively segmenting the contour curve to be detected and the standard contour curve according to the visual key points comprises the following steps:
s321, respectively carrying out vertex domain calculation on each point on the contour curve to be detected and the standard contour curve;
s322, calculating the curvature value of each point according to the vertex domain of the point;
s323, selecting a maximum value point as a segmentation point according to the calculated curvature value.
6. The method for rapid and intelligent detection of automobile parts according to claim 5, wherein the calculation formula for calculating the correlation coefficient between the profile to be detected and the standard profile according to the extracted feature sequence is as follows:
；
in the method, in the process of the application,rrepresenting the correlation coefficient of the contour curve to be detected and the standard contour curve;
t f andk f respectively representing the feature sequence of the contour curve to be detected and the feature sequence of the standard contour curvefCharacteristic values of the individual segments;
and->And respectively representing the expected sectional values of the profile characteristic sequence to be detected and the standard profile characteristic sequence.
7. The method for rapidly and intelligently detecting automobile parts according to claim 1, wherein the steps of dividing the difference between the contour curve to be detected and the standard contour curve and calculating the magnitude of the difference comprise the following steps:
s41, determining segments with differences in the profile curve to be detected and the standard profile curve according to the calculated correlation coefficient;
s42, segmenting the difference between the contour curve to be detected and the standard contour curve to obtain a difference region;
s43, determining the length and the width of the difference region according to the position of the difference region in the two-dimensional coordinate system;
s44, determining a rectangular boundary according to the length and the width of the difference region;
s45, calculating the area of the difference region by using a Monte Carlo method to obtain the area of the difference region.
8. The method for rapid and intelligent detection of automobile parts according to claim 7, wherein the calculation of the area of the difference region by using the monte carlo method to obtain the area size of the difference region comprises the following steps:
s451, generating random points in the determined rectangular boundary by using a uniform random number generator;
s452, counting the number of the generated random points and the number falling in the difference area;
s453, calculating the area of the difference region according to the difference region area calculation formula.
9. The method for rapidly and intelligently detecting the automobile parts according to claim 1, wherein the steps of judging whether the automobile parts to be detected can be repaired according to the difference, and feeding back the judgment result comprise the following steps:
s51, comparing and judging the difference between the automobile part to be detected and the standard automobile part with a preset threshold value;
s52, if the difference value is smaller than or equal to a preset threshold value, the automobile part to be detected is repairable, and if the difference value is larger than the preset threshold value, the automobile part to be detected is not repairable;
and S53, feeding back the judging result to the automobile part detecting personnel.
10. A rapid intelligent detection system for an automotive part, for implementing the rapid intelligent detection method for an automotive part according to any one of claims 1 to 9, the intelligent detection system comprising: the device comprises a component scanning module, a feature extraction module, a similarity calculation module, a difference calculation module and a judgment feedback module;
the component scanning module is used for scanning the automobile component to be detected and the standard automobile component through the 3D scanner respectively to obtain an image to be detected and a standard image;
the feature extraction module is used for respectively preprocessing and extracting the profile of the obtained image to be detected and the standard image to obtain a profile curve to be detected and a standard profile curve;
the similarity calculation module is used for calculating the similarity of the curve to be detected and the standard contour curve based on a curve segmentation similarity matching algorithm and determining the difference between the curve to be detected and the standard contour curve;
the difference calculation module is used for dividing the difference between the contour curve to be detected and the standard contour curve, and calculating the difference;
and the judging feedback module is used for judging whether the automobile part to be detected can be repaired or not according to the difference and feeding back the judging result.
CN202311114198.1A 2023-08-31 2023-08-31 Rapid and intelligent detection method and system for automobile parts Active CN116818778B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311114198.1A CN116818778B (en) 2023-08-31 2023-08-31 Rapid and intelligent detection method and system for automobile parts

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311114198.1A CN116818778B (en) 2023-08-31 2023-08-31 Rapid and intelligent detection method and system for automobile parts

Publications (2)

Publication Number Publication Date
CN116818778A true CN116818778A (en) 2023-09-29
CN116818778B CN116818778B (en) 2023-11-10

Family

ID=88139620

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311114198.1A Active CN116818778B (en) 2023-08-31 2023-08-31 Rapid and intelligent detection method and system for automobile parts

Country Status (1)

Country Link
CN (1) CN116818778B (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170147990A1 (en) * 2015-11-23 2017-05-25 CSI Holdings I LLC Vehicle transactions using objective vehicle data
CN108496124A (en) * 2015-11-09 2018-09-04 艾天诚工程技术系统股份有限公司 The automatic detection and robot assisted processing of surface defect
CN111415376A (en) * 2020-02-27 2020-07-14 湖南大学 Automobile glass sub-pixel contour extraction method and automobile glass detection method
CN111415378A (en) * 2020-02-27 2020-07-14 湖南大学 Image registration method for automobile glass detection and automobile glass detection method
US20200344449A1 (en) * 2017-05-11 2020-10-29 Inovision Software Solutions, Inc. Object inspection system and method for inspecting an object
CN113554649A (en) * 2021-09-22 2021-10-26 中科慧远视觉技术(北京)有限公司 Defect detection method and device, computer equipment and storage medium
CN115063429A (en) * 2022-08-18 2022-09-16 山东安德机械科技有限公司 Quality detection method for mechanical parts
CN116075855A (en) * 2020-08-19 2023-05-05 3M创新有限公司 Robot repair control system and method

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108496124A (en) * 2015-11-09 2018-09-04 艾天诚工程技术系统股份有限公司 The automatic detection and robot assisted processing of surface defect
US20170147990A1 (en) * 2015-11-23 2017-05-25 CSI Holdings I LLC Vehicle transactions using objective vehicle data
US20200344449A1 (en) * 2017-05-11 2020-10-29 Inovision Software Solutions, Inc. Object inspection system and method for inspecting an object
CN111415376A (en) * 2020-02-27 2020-07-14 湖南大学 Automobile glass sub-pixel contour extraction method and automobile glass detection method
CN111415378A (en) * 2020-02-27 2020-07-14 湖南大学 Image registration method for automobile glass detection and automobile glass detection method
CN116075855A (en) * 2020-08-19 2023-05-05 3M创新有限公司 Robot repair control system and method
CN113554649A (en) * 2021-09-22 2021-10-26 中科慧远视觉技术(北京)有限公司 Defect detection method and device, computer equipment and storage medium
CN115063429A (en) * 2022-08-18 2022-09-16 山东安德机械科技有限公司 Quality detection method for mechanical parts

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘克平;乔宇;李岩;张振国;杨宏韬;: "基于HALCON的汽车涂胶质量检测方法研究", 组合机床与自动化加工技术, no. 06 *

Also Published As

Publication number Publication date
CN116818778B (en) 2023-11-10

Similar Documents

Publication Publication Date Title
CN108921176B (en) Pointer instrument positioning and identifying method based on machine vision
CN110866924B (en) Line structured light center line extraction method and storage medium
CN106780486B (en) Steel plate surface defect image extraction method
Fujita et al. A robust automatic crack detection method from noisy concrete surfaces
CN116168026B (en) Water quality detection method and system based on computer vision
CN109872303B (en) Surface defect visual detection method and device and electronic equipment
JP5852919B2 (en) Crack detection method
Halder et al. Edge detection: A statistical approach
CN116758491B (en) Printing monitoring image analysis method and system applied to 3D printing
CN111724376B (en) Paper disease detection method based on texture feature analysis
CN112508913A (en) Cable section edge detection method based on image detection
CN112950594B (en) Method, device and storage medium for detecting surface defects of product
CN114581805A (en) Coating roller surface defect detection method adopting 3D line laser profile technology
CN116883408B (en) Integrating instrument shell defect detection method based on artificial intelligence
CN116818778B (en) Rapid and intelligent detection method and system for automobile parts
CN112669295A (en) Lithium battery pole piece defect detection method based on secondary threshold segmentation theory
CN109426770B (en) Iris identification method
KR20210082624A (en) Fingerprint Enhancement method
Zhu Moving Objects Detection and Segmentation Based on Background Subtraction and Image Over-Segmentation.
CN114994072A (en) Magnetic bar end surface defect detection method based on machine vision
CN114529515A (en) Method for automatically identifying internal defects of solar cell
Chong et al. Fabric Defect Detection Method Based on Projection Location and Superpixel Segmentation
CN110298816B (en) Bridge crack detection method based on image regeneration
Muravyov et al. Weld Defects Automatic Visual Recognition by Combined Application of Canny Edge Detector and Interval Fusion with Preference Aggregation
Adu-Gyamfi et al. Performance assessment of flexible pavements using active contour models

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