CN115115612A - Surface defect detection method and system for mechanical parts - Google Patents
Surface defect detection method and system for mechanical parts Download PDFInfo
- Publication number
- CN115115612A CN115115612A CN202210880834.0A CN202210880834A CN115115612A CN 115115612 A CN115115612 A CN 115115612A CN 202210880834 A CN202210880834 A CN 202210880834A CN 115115612 A CN115115612 A CN 115115612A
- Authority
- CN
- China
- Prior art keywords
- defect
- target
- connected domain
- edge
- 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.)
- Granted
Links
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
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a surface defect detection method and system for mechanical parts, and belongs to the technical field of image processing. The method comprises the following steps: obtaining a first fitted straight line of the first edge line and a second fitted straight line of the second edge line; obtaining each target defect connected domain corresponding to the first defect type according to the slope of the first fitting straight line and the slope of the second fitting straight line; recording other target defect connected domains except the target defect connected domains corresponding to the first defect type as connected domains to be judged; obtaining a fitting curve according to the distance between each edge pixel point on the first edge line corresponding to the connected domain to be judged and the corresponding first fitting straight line and the distance between each edge pixel point on the second edge line and the corresponding second fitting straight line; and obtaining each defect connected domain corresponding to the second defect type and each defect connected domain corresponding to the third defect type according to the fitting curve. The invention can efficiently identify the end surface defects of the bearing ring.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method and a system for detecting surface defects of mechanical parts.
Background
The bearing is one of essential important basic accessories in the assembly of various types of automobiles such as cars, heavy trucks and the like, the bearing ring is an annular part forming the bearing, the end face of the bearing ring can have some common defects in the production and processing process of the bearing ring under general conditions, the three defect types of excessive grinding, scratching and bruising are mainly used, the three defect types have similar appearances and are difficult to distinguish, the three defect types have inconsistent influences on the use effect and the quality of the bearing ring, and the treatment modes of the bearing rings with the three defect types are different; the quality of the bearing ring directly influences the quality of the bearing, and the quality of the bearing directly determines the performance and stability of equipment and products, so that the defect detection or identification of the bearing ring is particularly important.
The existing method generally realizes the identification of the defect area of the end face of the bearing ring and the identification of the defect type corresponding to the defect area based on the visual inspection of workers, and has the advantages of strong subjectivity, large workload and low efficiency, so the method for detecting the end face defect of the bearing ring based on manual work has low efficiency.
Disclosure of Invention
The invention provides a surface defect detection method and system for mechanical parts, which are used for solving the problem of low efficiency of defect detection on the end face of a bearing ring by the existing method, and adopt the following technical scheme:
in a first aspect, an embodiment of the present invention provides a surface defect detection method for a machine component, including the following steps:
acquiring an end face image of a metal bearing ring to be detected;
obtaining an edge contour image corresponding to the end face image by utilizing a maximum inter-class variance method and an edge detection operator; carrying out Hough circle detection on the edge contour image to obtain a target annular area and a target circle center coordinate;
obtaining each target defect connected domain in the target annular region and each edge pixel point corresponding to each target defect connected domain by using a seed filling method;
according to the distance between the edge pixel point coordinates and the target circle center coordinates, obtaining a first edge line and a second edge line corresponding to each target defect connected domain; fitting to obtain a first fitting straight line of the first edge line and a second fitting straight line of the second edge line according to the coordinates of each edge pixel point on the first edge line and the coordinates of each edge pixel point on the second edge line;
obtaining the angle difference between the first edge line corresponding to each target defect connected domain and the corresponding second edge line according to the slope of the first fitting straight line and the slope of the second fitting straight line; obtaining each target defect connected domain corresponding to the first defect type according to the angle difference;
recording other target defect connected domains except the target defect connected domains corresponding to the first defect type as connected domains to be judged; fitting to obtain a fitting curve corresponding to each connected domain to be determined according to the distance between each edge pixel point on the first edge line corresponding to each connected domain to be determined and the corresponding first fitting straight line and the distance between each edge pixel point on the second edge line corresponding to each connected domain to be determined and the corresponding second fitting straight line;
and obtaining each defect connected domain corresponding to the second defect type and each defect connected domain corresponding to the third defect type according to the number of the wave troughs on the fitting curve.
In a second aspect, the present invention provides a surface defect detecting system for a mechanical component, including a memory and a processor, wherein the processor executes a computer program stored in the memory to implement the above-mentioned surface defect detecting method for a mechanical component.
Preferably, the method for obtaining the coordinates of the target annular area and the target circle center includes:
carrying out Hough circle detection on the edge contour image to obtain two Hough circles of which the distance between the circle centers is smaller than a preset distance, and recording the two Hough circles as a target Hough circle;
calculating to obtain the mean value of the circle center coordinates corresponding to the two target Hough circles as the target circle center coordinates; the region range corresponding to the target Hough circle with the smaller radius in the two target Hough circles is a subset of the region range corresponding to the target Hough circle with the larger radius;
recording areas except the area range corresponding to the target Hough circle corresponding to the smaller radius in the area range corresponding to the target Hough circle corresponding to the larger radius as annular areas corresponding to the edge contour image;
setting the pixel value of a pixel point in the annular region as 1, setting the pixel values of other pixel points except the annular region on the edge contour image as 0, and recording as a mask image corresponding to the edge contour image;
and multiplying the mask image and the end face image to obtain a target annular region on the end face image.
Preferably, the method for obtaining each target defect connected domain in the target annular region and each edge pixel point corresponding to each target defect connected domain includes:
reversing the color of the target annular region, recording the reversed target annular region as a characteristic annular region, wherein the pixel value of a pixel point of a defect region on the characteristic annular region is 1;
obtaining a set of adjacent pixels with the pixel value of 1 in the characteristic annular region by using a seed filling method, and recording the set as a defective pixel set, wherein the number of the pixels in the defective pixel set is more than or equal to 1; a plurality of defective pixel point sets may exist in the characteristic annular region;
obtaining a defect connected domain corresponding to each defect pixel point set according to each defect pixel point set in the characteristic annular region;
screening out defect connected domains with the number of pixel values smaller than the preset number in each defect connected domain, and recording the remaining defect connected domains as target defect connected domains; extracting edge pixel point coordinates of each target defect connected domain by using a Canny edge detection operator;
and obtaining each target defect connected domain in the target annular region and each edge pixel point corresponding to each target defect connected domain according to each target defect connected domain in the characteristic annular region and the edge pixel point coordinates of each target defect connected domain.
Preferably, the method for obtaining the first edge line and the second edge line corresponding to each target defect connected domain includes:
for any target defect connected domain:
calculating to obtain the distance between the coordinate of each edge pixel point corresponding to the target defect connected domain and the target circle center coordinate, and recording as the characteristic distance corresponding to each edge pixel point;
obtaining the minimum characteristic distance and the maximum characteristic distance in the characteristic distances corresponding to the edge pixel points corresponding to the target defect connected domain; obtaining an empirical error amount corresponding to the target defect connected domain according to the minimum characteristic distance and the maximum characteristic distance; calculating an empirical error amount corresponding to the target defect connected domain according to the following formula:
wherein the content of the first and second substances,
the empirical error amount corresponding to the target defect connected domain,
in order to be the maximum feature distance,
is the minimum feature distance;
obtaining a target characteristic distance range corresponding to the target defect connected domain according to the empirical error amount; and eliminating each edge pixel point of which the characteristic distance is out of the range of the target characteristic distance, and obtaining a first edge line corresponding to the target defect connected domain and a second edge line corresponding to the target defect connected domain according to the residual edge pixel points after elimination.
Preferably, the method for obtaining each target defect connected domain corresponding to the first defect type according to the angle difference includes:
and judging whether the angle difference between the first edge line corresponding to each target defect connected domain and the corresponding second edge line is smaller than a preset angle threshold value, and if so, marking the corresponding target defect connected domain as a first defect type.
Preferably, the method for obtaining the fitting curve corresponding to each connected domain to be determined by fitting includes:
calculating to obtain the distance between each edge pixel point on the first edge line corresponding to each connected domain to be judged and the corresponding first fitting straight line, and recording as the first distance corresponding to each edge pixel point on the first edge line corresponding to each connected domain to be judged;
calculating to obtain the distance between each edge pixel point on the second edge line corresponding to each connected domain to be determined and the corresponding second fitting straight line, and recording as the second distance corresponding to each edge pixel point on the second edge line corresponding to each connected domain to be determined;
for any connected domain to be judged:
fitting to obtain a first fitting curve of the first edge line corresponding to the connected domain to be determined by taking the first distance corresponding to each edge pixel point on the first edge line corresponding to the connected domain to be determined as a vertical coordinate and taking the characteristic distance corresponding to each edge pixel point on the first edge line corresponding to the connected domain to be determined as a horizontal coordinate; and fitting to obtain a second fitting curve of the second edge line corresponding to the connected domain to be determined by taking the second distance corresponding to each edge pixel point on the second edge line corresponding to the connected domain to be determined as a vertical coordinate and taking the characteristic distance corresponding to each edge pixel point on the second edge line corresponding to the connected domain to be determined as a horizontal coordinate.
Preferably, the method for obtaining each defect connected domain corresponding to the second defect type and each defect connected domain corresponding to the third defect type according to the number of troughs on the fitting curve includes:
counting to obtain the number of troughs on a first fitting curve of a first edge line corresponding to each connected domain to be judged and the number of troughs on a second fitting curve of a corresponding second edge line;
judging whether the number of wave troughs on a first fitting curve of a first edge line corresponding to each connected domain to be judged or the number of wave troughs on a second fitting curve of a second edge line corresponding to each connected domain to be judged is greater than or equal to a preset wave trough number threshold value or not, and if so, marking the corresponding connected domain to be judged as a second defect type; otherwise, marking the corresponding connected domain to be judged as a third defect type.
Preferably, the first defect type is an excessive grinding defect, the second defect type is a scratch defect, and the third defect type is a gouge defect.
Has the advantages that: firstly, extracting an end face image of a metal bearing ring to be detected based on computer vision, and then obtaining a target annular area to be specifically researched and analyzed based on image processing; then, based on a seed filling method, obtaining each target defect connected domain in the target annular region and each edge pixel point corresponding to each target defect connected domain; the method for obtaining the target annular region and the target defect connected domain is obtained according to image analysis, and compared with a manual visual inspection mode, the method is higher in accuracy and reliability; then, obtaining each target defect connected domain corresponding to the first defect type based on the characteristic that the angle difference between the first edge line corresponding to the target defect connected domain of the first defect type and the second edge line corresponding to the target defect connected domain is small; analyzing other target defect connected domains except the target defect connected domains corresponding to the first defect type, and recording the other target defect connected domains except the target defect connected domains corresponding to the first defect type as connected domains to be judged; then fitting according to the distance between each edge pixel point on the first edge line corresponding to each connected domain to be determined and the corresponding first fitting straight line and the distance between each edge pixel point on the second edge line corresponding to each connected domain to be determined and the corresponding second fitting straight line to obtain a fitting curve corresponding to each connected domain to be determined; then obtaining each target defect connected domain corresponding to the second defect type and each target defect connected domain corresponding to the third defect type according to the number of wave troughs on the fitting curve and the edge characteristics of the second defect type and the third defect type, wherein the edge characteristics of the second defect type and the third defect type refer to a smooth state of the edge of the second defect type connected domain and a non-smooth state of the edge of the third defect type connected domain; therefore, the method for obtaining the target defect connected domains corresponding to the first defect type, the defect connected domains corresponding to the second defect type and the defect connected domains corresponding to the third defect type according to the connected domain edge characteristics of the first defect type, the second defect type and the third defect type is more reliable, accurate and efficient than the method according to the manual mode. Therefore, the invention provides a defect detection method with higher automation; the method can avoid subjectivity and inefficiency of manual visual inspection, and can improve accuracy and reliability of end surface defect identification of the bearing ring.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or 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 other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a method for surface defect detection of a machine component according to the present invention;
FIG. 2 is a schematic diagram of edge lines corresponding to a target defect connected domain according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying 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, rather than all embodiments, and all other embodiments obtained by those skilled in the art based on the embodiments of the present invention belong to the protection scope of the embodiments of the present invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The embodiment provides a surface defect detection method for a mechanical part, which is described in detail as follows:
as shown in fig. 1, the method for detecting surface defects of a machine component includes the following steps:
and S001, acquiring coordinates of the target annular area and the target circle center.
The method mainly comprises the steps of extracting an end face image of a metal bearing ring to be detected based on a computer vision technology, and then realizing intelligent detection of the defect types of the end face of the bearing ring through image processing and mathematical modeling, wherein the detected defect types comprise excessive grinding, scratch and bump damage and are respectively recorded as a first defect type, a second defect type and a third defect type; the embodiment provides a bearing ring end surface defect detection method with high automation; the method can not only avoid subjectivity and low efficiency of manual visual inspection, but also improve accuracy and reliability of bearing ring end surface defect identification.
In the embodiment, an image acquisition system is used for acquiring an end face image of a metal bearing ring to be detected, wherein the image acquisition system mainly comprises a camera, a bracket, a ball integral light source, the metal bearing ring to be detected and a diffuse reflection sampling plate, wherein the bearing ring is horizontally placed on the diffuse reflection sampling plate, the camera is arranged above the horizontally placed bearing ring, the acquisition visual angle of the camera is downward, and the ball integral light source is arranged between the camera and the bearing ring; and when the image is collected, firstly, the ball integral light source is controlled to irradiate the bearing ring and is reflected to the camera to shoot the upper end face area and the lower end face area of the bearing ring, namely, one bearing ring corresponds to two end face images, and only one end face image is analyzed in the following process.
Then, carrying out graying processing on the end face image to obtain a grayscale image corresponding to the end face image; noise is inevitably introduced in the image acquisition process, so that the subsequent analysis effect is influenced; therefore, the gray level image needs to be preprocessed to obtain a target gray level image corresponding to the end face image; the pretreatment process comprises the following steps: the method includes the steps that a gray level image is denoised by median filtering, noise can be eliminated, image details can be well kept, the calculation time is short, then the denoised image is subjected to image enhancement processing, the contrast of characteristics of a suspicious defect area is improved, defect characteristic enhancement is carried out by exponential transformation, and the gray level image after denoising and characteristic enhancement is recorded as a target gray level image.
Next, in this embodiment, the maximum inter-class variance method (OSTU) is used to count the histogram characteristics of the entire target grayscale image to realize automatic selection of the global threshold T, and then the grayscale value of the pixel point on the target grayscale image that is greater than the global threshold T is set to 1, and the grayscale value of the pixel point that is less than the global threshold T is set to 0, so as to obtain a binary image corresponding to the target grayscale image; the binarization of the image by using the maximum inter-class variance method is a well-known technique, and therefore, will not be described in detail.
Since the present embodiment mainly studies and analyzes the defect of the end face of the bearing ring, in order to reduce the time for subsequent analysis and processing and the accuracy of detection, the end face area of the bearing ring on the end face image of the bearing ring needs to be obtained; the specific process is as follows:
firstly, detecting a binary image corresponding to the obtained target gray image by using a Canny edge operator to obtain an edge contour image corresponding to the binary image and each edge pixel point on the edge contour image; the edge contour image comprises contour pixel information of the end face of the bearing ring; because the target gray level image corresponds to the end face image, the edge image also corresponds to the end face image; then the lower left corner on the edge contour image is taken as the origin of coordinates
Establishing a coordinate system by taking the horizontal direction as the positive direction of a transverse axis and the vertical direction as the positive direction of a longitudinal axis, and further obtaining coordinate information of each edge pixel point on the edge contour image; then, Hough circle detection is carried out on the edge outline image to obtain two Hough circles of which the distance between the circle centers is smaller than a preset distance, and the two Hough circles are recorded as eyesMarking a Hough circle; the preset distance needs to be set according to actual conditions. The basic idea of Hough circle detection is to map a detected image to a parameter space, and the boundary point of an edge curve in the detected image needs to meet a certain parameter, and the image problem is converted into a simple local peak value detection problem by applying a method of setting an accumulator and voting; the Hough circle detection is a well-known technology, and a specific process is not described in detail.
Because the inner edge and the outer edge of the normal bearing ring are concentric circles, the distance difference between the centers of two target Hoff circles is small, and the value of the preset distance is small; calculating to obtain the mean value of the circle center coordinates corresponding to the two target Hough circles, and taking the mean value as the target circle center coordinates; the region range corresponding to the target Hough circle with the smaller radius in the two target Hough circles is a subset of the region range corresponding to the target Hough circle with the larger radius; recording an area except for an area corresponding to the target Hough circle with a smaller radius in an area corresponding to the target Hough circle with a larger radius as an annular area corresponding to the edge contour image, wherein the annular area is also an end surface area of the bearing ring; setting the pixel value of a pixel point of the annular region to be 1, setting the pixel values of other pixel points except the annular region on the edge contour image to be 0, and recording as a mask image corresponding to the edge contour image; the bearing ring end face image, the gray level image, the target gray level image, the edge profile image and each pixel point on the mask image are in one-to-one correspondence; and multiplying the mask image and the end face image to extract an ROI (region of interest) which is recorded as a target annular region on the end face image.
And step S002, obtaining each target defect connected domain in the target annular region and each edge pixel corresponding to each target defect connected domain by using a seed filling method.
In step S002, a target annular region to be specifically researched and analyzed is obtained, and then each target defect connected domain on the target annular region and each edge pixel point coordinate corresponding to each target defect connected domain are obtained by analyzing the target annular region; the specific process is as follows:
firstly, reversing a target annular region on an end face image, recording the reversed target annular region as a characteristic annular region, and setting the pixel value of a pixel point of a defect region on the characteristic annular region to be 1, namely the defect region is white; then, a seed filling method is utilized to obtain a set of adjacent pixel points with the pixel value of 1 in the characteristic annular region, the set is recorded as a defective pixel point set, and the number of the pixel points in the defective pixel point set is more than or equal to 1; but a defect pixel point set may not exist in the characteristic annular region, if the defect pixel point set does not exist, the defect does not exist in the characteristic annular region, namely the end face of the metal bearing ring to be detected does not have a defect; if the characteristic annular region has the defective pixel point set, the defect of the characteristic annular region is indicated, the defective pixel point set region needs to be further analyzed, and a plurality of defective pixel point sets may exist in the characteristic annular region.
Then, in the embodiment, the region formed by each pixel point in each defective pixel point set in the characteristic annular region is recorded as each defect connected domain in the characteristic annular region, and a label value is set for each defect connected domain, so that subsequent processing is facilitated; since the defect connected domains with small areas may be caused by external noise, in this embodiment, in order to reduce the calculation amount of subsequent analysis, the defect connected domains with the pixel value number smaller than the preset number in the defect connected domains are screened out, the remaining defect connected domains are marked as target defect connected domains, and edge pixel point coordinates of each target defect connected domain are extracted by using a Canny edge detection operator; the preset number needs to be set according to an actual situation or according to the set defect region detection precision, where the defect region detection precision refers to a defect region with the smallest area that can be detected and analyzed, for example, if the defect region with the smallest area is detected and analyzed by 5mm × 1mm, the number of pixels in a connected domain with the preset number of 5mm × 1mm is preset. Therefore, in this embodiment, each target defect connected domain in the feature annular region is obtained through the above process, and each pixel point on the target annular region corresponds to each pixel point on the feature annular region one to one, so according to each target defect connected domain in the feature annular region, each target defect connected domain in the target annular region and each edge pixel point corresponding to each target defect connected domain can be obtained.
S003, obtaining a first edge line and a second edge line corresponding to each target defect connected domain according to the distance between the edge pixel point coordinate and the target circle center coordinate; and fitting according to the coordinates of each edge pixel point on the first edge line and the coordinates of each edge pixel point on the second edge line to obtain a first fitting straight line of the first edge line and a second fitting straight line of the second edge line.
In this embodiment, analysis is performed only on three defect types, i.e., excessive grinding, scratching, and gouging, where the excessive grinding is compared with the other two defect types, and the upper and lower edge lines of the excessive grinding defect connected domain have obvious directional consistency, i.e., the difference in slope between the upper and lower edge lines of the excessive grinding defect connected domain is small, and the upper and lower edge lines of the connected domain of the other two defect types are not fixed, which causes the above phenomenon; in addition, the edges of the communication area with excessive grinding and bruising are in a smooth state, the scratches generally extend from one boundary of the target annular area to the other boundary, and are more discrete from the head to the tail, namely are more discrete along the extending direction, and the distribution state is not smooth; therefore, the defect type corresponding to each defect connected domain is obtained in the embodiment based on the characteristics of the different defect types.
Firstly, obtaining upper and lower edge lines corresponding to each target defect connected domain, marking the upper edge line as a first edge line, and marking the lower edge line as a second edge line; the specific process of obtaining the first edge line and the corresponding second edge line corresponding to each target defect connected domain is as follows:
for any target defect connected domain:
calculating to obtain the distance between the coordinate of each edge pixel point corresponding to the target defect connected domain and the target circle center coordinate, and recording as the characteristic distance corresponding to each edge pixel point; for any edge pixel point corresponding to the target defect connected domain, calculating the characteristic distance corresponding to the edge pixel point according to the following formula:
wherein, the first and the second end of the pipe are connected with each other,
the feature distance corresponding to the edge pixel point is,
is the abscissa corresponding to the edge pixel point,
is the ordinate corresponding to the edge pixel point,
is the abscissa of the center of the target circle,
is the ordinate of the center of the target circle.
Therefore, the characteristic distance corresponding to each edge pixel point corresponding to the target defect connected domain can be obtained through the process, and the minimum characteristic distance and the maximum characteristic distance are obtained; obtaining an empirical error amount corresponding to the target defect connected domain according to the minimum characteristic distance and the maximum characteristic distance; calculating an empirical error amount corresponding to the target defect connected domain according to the following formula:
wherein the content of the first and second substances,
the empirical error amount corresponding to the target defect connected domain,
in order to be the maximum feature distance,
is the minimum feature distance.
Obtaining a target characteristic distance range corresponding to the target defect connected domain according to the empirical error quantity
Then eliminating each edge pixel point with the characteristic distance outside the target characteristic distance range, wherein the remaining edge pixel points form two edge lines, namely a first edge line corresponding to the target defect connected domain and a second edge line corresponding to the target defect connected domain, and obtaining each edge pixel point on the first edge line and each edge pixel point on the second edge line; as shown in fig. 2, the dotted line
And the dotted line
The two edge lines in the area range between the two edge lines are a first edge line corresponding to the target defect connected domain and a second edge line corresponding to the target defect connected domain, and the first edge line is above the second edge line.
Therefore, the first edge line and the second edge line corresponding to each target defect connected domain, and each edge pixel point on the first edge line and each edge pixel point on the second edge line can be obtained through the process; in the process, the edge line pixel points which are close to the edge of the target annular region are eliminated, because the interference with the part close to the edge of the target annular region is large, the subsequent analysis is influenced; and the first edge line and the second edge line corresponding to each target defect connected domain are the basis for obtaining the defect type corresponding to each target defect connected domain through subsequent analysis.
Then, according to each edge pixel point on the first edge line corresponding to each target defect connected domain, fitting by using a least square method to obtain a fitted straight line of the first edge line corresponding to each target defect connected domain, marking the fitted straight line as a first fitted straight line, and obtaining the slope and intercept of the first fitted straight line; then according to each edge pixel point on the second edge line corresponding to each target defect connected domain, fitting by using a least square method to obtain a fitting straight line of the second edge line corresponding to each target defect connected domain, marking the fitting straight line as a second fitting straight line, and obtaining the slope and intercept of the second fitting straight line; the fitting of a straight line by the least square method is a well-known technique and therefore will not be described in detail; and the slope and the intercept of the fitting straight line are the basis for identifying the defect type corresponding to each target defect connected domain through subsequent analysis.
Step S004, obtaining the angle difference between the first edge line corresponding to each target defect connected domain and the corresponding second edge line according to the slope of the first fitting straight line and the slope of the second fitting straight line; and obtaining each target defect connected domain corresponding to the first defect type according to the angle difference.
Because the slope difference between the first edge line and the second edge line of the grinding excess defect connected domain is small, and the slope difference between the first edge line and the second edge line of the connected domain of the other two defect types is large, the angle difference between the first edge line and the corresponding second edge line corresponding to each target defect connected domain is obtained based on the slope of the first fitting straight line and the slope of the second fitting straight line of the first edge line corresponding to each target defect connected domain; for any target defect connected domain, calculating the angle difference between the first edge line corresponding to the target defect connected domain and the corresponding second edge line according to the following formula:
wherein the content of the first and second substances,
the angle difference between the corresponding first edge line and the corresponding second edge line of the target defect connected domain,
the slope of the first fitted straight line of the first edge line corresponding to the target defect connected domain,
the slope of the second fitted straight line of the second edge line corresponding to the target defect connected domain,
is an arctangent function;
the smaller the probability that the target defect connected domain is the first defect type is indicated to be.
Therefore, the angle difference between the first edge line and the second edge line corresponding to each target defect connected domain can be obtained through the process; then judging whether the angle difference between the first edge line corresponding to each target defect connected domain and the corresponding second edge line is smaller than a preset angle threshold value or not, if so, marking the corresponding target defect connected domain as a first defect type, namely, an excessive grinding defect, and polishing the excessive grinding region; otherwise, further analysis is carried out on the target defect connected domain to obtain the defect type; the present embodiment sets the preset angle threshold to be
As another embodiment, the value of the preset angle threshold may be set according to actual conditions.
Step S005, recording other target defect connected domains except each target defect connected domain corresponding to the first defect type as connected domains to be judged; and fitting to obtain a fitting curve corresponding to each connected domain to be determined according to the distance between each edge pixel point on the first edge line corresponding to each connected domain to be determined and the corresponding first fitting straight line and the distance between each edge pixel point on the second edge line corresponding to each connected domain to be determined and the corresponding second fitting straight line.
In this embodiment, each target defect connected domain corresponding to the first defect type is obtained through step S005, and then other target defect connected domains except each target defect connected domain corresponding to the first defect type are marked as connected domains to be determined; then calculating to obtain the distance between each edge pixel point on the first edge line corresponding to each connected domain to be judged and the corresponding first fitting straight line, and recording as the first distance corresponding to each edge pixel point on the first edge line corresponding to each connected domain to be judged; for any edge pixel point on a first edge line corresponding to any connected domain to be judged, calculating a first distance corresponding to the edge pixel point according to the following formula:
wherein the content of the first and second substances,
is the first distance corresponding to the edge pixel point,
for the abscissa corresponding to the edge pixel point,
is the ordinate corresponding to the edge pixel point,
the slope of the first fitted straight line of the first edge line corresponding to the connected domain to be determined,
and the intercept of a first fitted straight line of the first edge line corresponding to the connected domain to be judged.
Calculating to obtain the distance between each edge pixel point on the second edge line corresponding to each connected domain to be determined and the corresponding second fitting straight line, and recording as the second distance corresponding to each edge pixel point on the second edge line corresponding to each connected domain to be determined; for any edge pixel point on a second edge line corresponding to any connected domain to be determined, calculating a second distance corresponding to the edge pixel point according to the following formula:
wherein the content of the first and second substances,
a second distance corresponding to the edge pixel point,
for the abscissa corresponding to the edge pixel point,
is the ordinate corresponding to the edge pixel point,
the slope of the second fitted straight line of the second edge line corresponding to the connected domain to be determined,
and the intercept of a second fitted straight line of a second edge line corresponding to the connected domain to be determined.
For any connected domain to be judged: fitting to obtain a first fitting curve of the first edge line corresponding to the connected domain to be determined by taking the first distance corresponding to each edge pixel point on the first edge line corresponding to the connected domain to be determined as a vertical coordinate and taking the characteristic distance corresponding to each edge pixel point on the first edge line corresponding to the connected domain to be determined as a horizontal coordinate; and fitting to obtain a second fitting curve of the second edge line corresponding to the connected domain to be determined by taking the second distance corresponding to each edge pixel point on the second edge line corresponding to the connected domain to be determined as a vertical coordinate and taking the characteristic distance corresponding to each edge pixel point on the second edge line corresponding to the connected domain to be determined as a horizontal coordinate.
Therefore, in this embodiment, a first fitted curve of the first edge line corresponding to each connected component to be determined and a second fitted curve of the corresponding second edge line can be obtained through the above process.
And S006, obtaining each defect connected domain corresponding to the second defect type and each defect connected domain corresponding to the third defect type according to the number of the wave troughs on the fitting curve.
In this embodiment, a first fitted curve of a first edge line corresponding to each connected domain to be determined and a second fitted curve of a corresponding second edge line are obtained through step S006, and then the number of troughs on the first fitted curve of the first edge line corresponding to each connected domain to be determined and the number of troughs on the second fitted curve of the corresponding second edge line are obtained through statistics; then judging whether the number of wave troughs on a first fitting curve of a first edge line corresponding to each connected domain to be judged or the number of wave troughs on a second fitting curve of a corresponding second edge line is more than or equal to a preset wave trough number threshold value or not, if so, judging that unsmooth edge line segments exist on the edge line corresponding to the corresponding connected domain to be judged, marking the corresponding connected domain to be judged as a second defect type, namely a scratch defect, and polishing the scratch defect again; otherwise, judging that an uneven edge line segment does not exist on the edge line corresponding to the corresponding connected domain to be judged, marking the corresponding connected domain to be judged as a third defect type, namely a bump defect, and scrapping the bearing ring with the bump defect without using; as another embodiment, the preset valley number threshold may be set to be 3, and other values may also be set for the preset valley number threshold according to different requirements.
According to the embodiment, firstly, an end face image of a metal bearing ring to be detected is extracted based on computer vision, and then a target annular region to be specifically researched and analyzed is obtained based on image processing; then, based on a seed filling method, obtaining each target defect connected domain in the target annular region and each edge pixel point corresponding to each target defect connected domain; the method for obtaining the target annular region and the target defect connected domain is obtained according to image analysis, and compared with a manual visual inspection mode, the method is higher in accuracy and reliability; then, the embodiment obtains each target defect connected domain corresponding to the first defect type based on the characteristic that the angle difference between the first edge line corresponding to the target defect connected domain of the first defect type and the second edge line corresponding to the target defect connected domain is small; analyzing other target defect connected domains except the target defect connected domains corresponding to the first defect type, and recording the other target defect connected domains except the target defect connected domains corresponding to the first defect type as connected domains to be judged; then fitting according to the distance between each edge pixel point on the first edge line corresponding to each connected domain to be determined and the corresponding first fitting straight line and the distance between each edge pixel point on the second edge line corresponding to each connected domain to be determined and the corresponding second fitting straight line to obtain a fitting curve corresponding to each connected domain to be determined; then obtaining each target defect connected domain corresponding to the second defect type and each target defect connected domain corresponding to the third defect type according to the number of wave troughs on the fitting curve and the edge characteristics of the second defect type and the third defect type, wherein the edge characteristics of the second defect type and the third defect type refer to a smooth state of the edge of the second defect type connected domain and a non-smooth state of the edge of the third defect type connected domain; therefore, the method for obtaining the target defect connected domains corresponding to the first defect type, the defect connected domains corresponding to the second defect type and the defect connected domains corresponding to the third defect type according to the connected domain edge characteristics of the first defect type, the connected domain edge characteristics of the second defect type and the connected domain edge characteristics of the third defect type in the embodiment is more reliable, accurate and efficient than the method according to the manual method. Therefore, the embodiment provides a defect detection method with high automation; the method can not only avoid subjectivity and low efficiency of manual visual inspection, but also improve accuracy and reliability of bearing ring end surface defect identification.
A surface defect detection system for a machine component of the present embodiment includes a memory and a processor, and the processor executes a computer program stored in the memory to implement the above-described surface defect detection method for a machine component.
The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (9)
1. A method for detecting surface defects of a machine component, the method comprising the steps of:
acquiring an end face image of a metal bearing ring to be detected;
obtaining an edge contour image corresponding to the end face image by utilizing a maximum inter-class variance method and an edge detection operator; carrying out Hough circle detection on the edge contour image to obtain a target annular area and a target circle center coordinate;
obtaining each target defect connected domain in the target annular region and each edge pixel point corresponding to each target defect connected domain by using a seed filling method;
according to the distance between the edge pixel point coordinates and the target circle center coordinates, obtaining first edge lines and second edge lines corresponding to the target defect connected domains; fitting to obtain a first fitting straight line of the first edge line and a second fitting straight line of the second edge line according to the coordinates of each edge pixel point on the first edge line and the coordinates of each edge pixel point on the second edge line;
obtaining the angle difference between the first edge line corresponding to each target defect connected domain and the corresponding second edge line according to the slope of the first fitting straight line and the slope of the second fitting straight line; obtaining each target defect connected domain corresponding to the first defect type according to the angle difference;
recording other target defect connected domains except the target defect connected domains corresponding to the first defect type as connected domains to be judged; fitting to obtain a fitting curve corresponding to each connected domain to be determined according to the distance between each edge pixel point on the first edge line corresponding to each connected domain to be determined and the corresponding first fitting straight line and the distance between each edge pixel point on the second edge line corresponding to each connected domain to be determined and the corresponding second fitting straight line;
and obtaining each defect connected domain corresponding to the second defect type and the defect connected domain corresponding to the third defect type according to the number of the wave troughs on the fitting curve.
2. The method of claim 1, wherein the step of obtaining coordinates of the target annular region and the target center of the circle comprises:
carrying out Hough circle detection on the edge contour image to obtain two Hough circles of which the distance between the circle centers is smaller than a preset distance, and recording the two Hough circles as a target Hough circle;
calculating to obtain the mean value of the circle center coordinates corresponding to the two target Hough circles as the target circle center coordinates; the region range corresponding to the target Hough circle with the smaller radius in the two target Hough circles is a subset of the region range corresponding to the target Hough circle with the larger radius;
recording areas except the area range corresponding to the target Hough circle corresponding to the smaller radius in the area range corresponding to the target Hough circle corresponding to the larger radius as annular areas corresponding to the edge contour image;
setting the pixel value of a pixel point of the annular region to be 1, setting the pixel values of other pixel points except the annular region on the edge contour image to be 0, and recording as a mask image corresponding to the edge contour image;
and multiplying the mask image and the end face image to obtain a target annular region on the end face image.
3. The method for detecting surface defects of mechanical parts according to claim 1, wherein the method for obtaining each target defect connected domain in the target annular region and each edge pixel point corresponding to each target defect connected domain comprises the following steps:
reversing the color of the target annular region, recording the reversed target annular region as a characteristic annular region, wherein the pixel value of a pixel point of a defect region on the characteristic annular region is 1;
obtaining a set of adjacent pixels with the pixel value of 1 in the characteristic annular region by using a seed filling method, and recording the set as a defective pixel set, wherein the number of the pixels in the defective pixel set is more than or equal to 1; a plurality of defective pixel point sets may exist in the characteristic annular region;
obtaining a defect connected domain corresponding to each defect pixel point set according to each defect pixel point set in the characteristic annular region;
screening out defect connected domains with the number of pixel values smaller than the preset number in each defect connected domain, and recording the remaining defect connected domains as target defect connected domains; extracting edge pixel point coordinates of each target defect connected domain by using a Canny edge detection operator;
and obtaining each target defect connected domain in the target annular region and each edge pixel point corresponding to each target defect connected domain according to each target defect connected domain in the characteristic annular region and the edge pixel point coordinates of each target defect connected domain.
4. The method for detecting surface defects of a mechanical part according to claim 1, wherein the method for obtaining the first edge line and the second edge line corresponding to each target defect connected domain comprises:
for any target defect connected domain:
calculating to obtain the distance between the coordinate of each edge pixel point corresponding to the target defect connected domain and the target circle center coordinate, and recording as the characteristic distance corresponding to each edge pixel point;
obtaining the minimum characteristic distance and the maximum characteristic distance in the characteristic distances corresponding to the edge pixel points corresponding to the target defect connected domain; obtaining an empirical error amount corresponding to the target defect connected domain according to the minimum characteristic distance and the maximum characteristic distance; calculating an empirical error amount corresponding to the target defect connected domain according to the following formula:
wherein the content of the first and second substances,
the empirical error amount corresponding to the target defect connected domain,
in order to be the maximum feature distance,
is the minimum feature distance;
obtaining a target characteristic distance range corresponding to the target defect connected domain according to the empirical error amount; and eliminating each edge pixel point with the characteristic distance outside the target characteristic distance range, and obtaining a first edge line corresponding to the target defect connected domain and a second edge line corresponding to the target defect connected domain according to the residual edge pixel points after elimination.
5. The method for detecting surface defects of a mechanical part according to claim 1, wherein the method for obtaining the target defect connected domain corresponding to the first defect type according to the angle difference comprises:
and judging whether the angle difference between the first edge line corresponding to each target defect connected domain and the corresponding second edge line is smaller than a preset angle threshold value, and if so, marking the corresponding target defect connected domain as a first defect type.
6. The method for detecting surface defects of a mechanical part according to claim 4, wherein the fitting method for obtaining the fitting curve corresponding to each connected domain to be determined comprises the following steps:
calculating to obtain the distance between each edge pixel point on the first edge line corresponding to each connected domain to be judged and the corresponding first fitting straight line, and recording as the first distance corresponding to each edge pixel point on the first edge line corresponding to each connected domain to be judged;
calculating to obtain the distance between each edge pixel point on the second edge line corresponding to each connected domain to be determined and the corresponding second fitting straight line, and recording as the second distance corresponding to each edge pixel point on the second edge line corresponding to each connected domain to be determined;
for any connected domain to be judged:
fitting to obtain a first fitting curve of the first edge line corresponding to the connected domain to be determined by taking the first distance corresponding to each edge pixel point on the first edge line corresponding to the connected domain to be determined as a vertical coordinate and taking the characteristic distance corresponding to each edge pixel point on the first edge line corresponding to the connected domain to be determined as a horizontal coordinate; and fitting to obtain a second fitting curve of the second edge line corresponding to the connected domain to be determined by taking the second distance corresponding to each edge pixel point on the second edge line corresponding to the connected domain to be determined as a vertical coordinate and taking the characteristic distance corresponding to each edge pixel point on the second edge line corresponding to the connected domain to be determined as a horizontal coordinate.
7. The method for detecting surface defects of a mechanical part according to claim 6, wherein the method for obtaining defect connected domains corresponding to a second defect type and defect connected domains corresponding to a third defect type according to the number of troughs on the fitted curve comprises:
counting to obtain the number of troughs on a first fitting curve of a first edge line corresponding to each connected domain to be judged and the number of troughs on a second fitting curve of a corresponding second edge line;
judging whether the number of wave troughs on a first fitting curve of a first edge line corresponding to each connected domain to be judged or the number of wave troughs on a second fitting curve of a second edge line corresponding to each connected domain to be judged is greater than or equal to a preset wave trough number threshold value or not, and if so, marking the corresponding connected domain to be judged as a second defect type; otherwise, marking the corresponding connected domain to be judged as a third defect type.
8. The surface defect detection method for a mechanical part according to claim 1, wherein the first defect type is an excessive grinding defect, the second defect type is a scratch defect, and the third defect type is a gouge defect.
9. A surface defect detection system for a machine part, comprising a memory and a processor, wherein the processor executes a computer program stored in the memory to implement a surface defect detection method for a machine part as claimed in any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210880834.0A CN115115612B (en) | 2022-07-26 | 2022-07-26 | Surface defect detection method and system for mechanical parts |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210880834.0A CN115115612B (en) | 2022-07-26 | 2022-07-26 | Surface defect detection method and system for mechanical parts |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115115612A true CN115115612A (en) | 2022-09-27 |
| CN115115612B CN115115612B (en) | 2022-11-04 |
Family
ID=83334469
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210880834.0A Active CN115115612B (en) | 2022-07-26 | 2022-07-26 | Surface defect detection method and system for mechanical parts |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115115612B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115375685A (en) * | 2022-10-25 | 2022-11-22 | 临沂天元混凝土工程有限公司 | Method for detecting sand particle size abnormity in concrete raw material |
| CN116152234A (en) * | 2023-04-18 | 2023-05-23 | 山东阁林板建材科技有限公司 | Template end face defect identification method based on image processing |
| CN116188468A (en) * | 2023-04-27 | 2023-05-30 | 济宁市长荣电器科技有限公司 | HDMI cable transmission letter sorting intelligent control system |
| CN116703903A (en) * | 2023-08-04 | 2023-09-05 | 山东济宁运河煤矿有限责任公司 | Anchor rod repairing quality detection method based on machine vision |
| CN116958144A (en) * | 2023-09-20 | 2023-10-27 | 东莞市南谷第电子有限公司 | Rapid positioning method and system for surface defect area of new energy connecting line |
| CN117197138A (en) * | 2023-11-07 | 2023-12-08 | 深圳市极摩客科技有限公司 | Computer host shell scratch detection method based on image characteristics |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335963A (en) * | 2015-09-24 | 2016-02-17 | 凌云光技术集团有限责任公司 | Edge defect detection method and apparatus |
| CN106778734A (en) * | 2016-11-10 | 2017-05-31 | 华北电力大学（保定） | A kind of insulator based on rarefaction representation falls to go here and there defect inspection method |
| WO2017164338A1 (en) * | 2016-03-25 | 2017-09-28 | マークテック株式会社 | Flaw detection apparatus and flaw detection method using flaw detection apparatus |
| CN107256549A (en) * | 2017-06-06 | 2017-10-17 | 滁州市天达汽车部件有限公司 | A kind of bamboo strip defect detection method based on machine vision |
| CN107796826A (en) * | 2017-09-27 | 2018-03-13 | 浙江大学 | Miniature duplicate gear broken teeth defect inspection method based on the analysis of tooth heart distance Curve |
| CN110163853A (en) * | 2019-05-14 | 2019-08-23 | 广东奥普特科技股份有限公司 | A kind of detection method of edge defect |
| CN113658133A (en) * | 2021-08-16 | 2021-11-16 | 江苏鑫丰源机电有限公司 | Gear surface defect detection method and system based on image processing |
| CN113884299A (en) * | 2021-12-02 | 2022-01-04 | 武汉市书豪塑胶有限公司 | Rotational molding machine fault detection method based on artificial intelligence |
| CN114757949A (en) * | 2022-06-15 | 2022-07-15 | 济宁市海富电子科技有限公司 | Wire and cable defect detection method and system based on computer vision |
-
2022
- 2022-07-26 CN CN202210880834.0A patent/CN115115612B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335963A (en) * | 2015-09-24 | 2016-02-17 | 凌云光技术集团有限责任公司 | Edge defect detection method and apparatus |
| WO2017164338A1 (en) * | 2016-03-25 | 2017-09-28 | マークテック株式会社 | Flaw detection apparatus and flaw detection method using flaw detection apparatus |
| CN106778734A (en) * | 2016-11-10 | 2017-05-31 | 华北电力大学（保定） | A kind of insulator based on rarefaction representation falls to go here and there defect inspection method |
| CN107256549A (en) * | 2017-06-06 | 2017-10-17 | 滁州市天达汽车部件有限公司 | A kind of bamboo strip defect detection method based on machine vision |
| CN107796826A (en) * | 2017-09-27 | 2018-03-13 | 浙江大学 | Miniature duplicate gear broken teeth defect inspection method based on the analysis of tooth heart distance Curve |
| CN110163853A (en) * | 2019-05-14 | 2019-08-23 | 广东奥普特科技股份有限公司 | A kind of detection method of edge defect |
| CN113658133A (en) * | 2021-08-16 | 2021-11-16 | 江苏鑫丰源机电有限公司 | Gear surface defect detection method and system based on image processing |
| CN113884299A (en) * | 2021-12-02 | 2022-01-04 | 武汉市书豪塑胶有限公司 | Rotational molding machine fault detection method based on artificial intelligence |
| CN114757949A (en) * | 2022-06-15 | 2022-07-15 | 济宁市海富电子科技有限公司 | Wire and cable defect detection method and system based on computer vision |
Non-Patent Citations (2)
| Title |
|---|
| QINGFENG WANG 等: "Online Incipient Fault Detection Method Based on Improved ℓ1 Trend Filtering and Support Vector Data Description", 《IEEE ACCESS》 * |
| 胡佳成 等: "ABS齿圈环形表面缺陷检测方法", 《电子测量与仪器学报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115375685A (en) * | 2022-10-25 | 2022-11-22 | 临沂天元混凝土工程有限公司 | Method for detecting sand particle size abnormity in concrete raw material |
| CN115375685B (en) * | 2022-10-25 | 2023-03-24 | 临沂天元混凝土工程有限公司 | Method for detecting sand and stone particle size abnormity in concrete raw material |
| CN116152234A (en) * | 2023-04-18 | 2023-05-23 | 山东阁林板建材科技有限公司 | Template end face defect identification method based on image processing |
| CN116152234B (en) * | 2023-04-18 | 2023-08-01 | 山东阁林板建材科技有限公司 | Template end face defect identification method based on image processing |
| CN116188468A (en) * | 2023-04-27 | 2023-05-30 | 济宁市长荣电器科技有限公司 | HDMI cable transmission letter sorting intelligent control system |
| CN116703903A (en) * | 2023-08-04 | 2023-09-05 | 山东济宁运河煤矿有限责任公司 | Anchor rod repairing quality detection method based on machine vision |
| CN116703903B (en) * | 2023-08-04 | 2023-10-20 | 山东济宁运河煤矿有限责任公司 | Anchor rod repairing quality detection method based on machine vision |
| CN116958144A (en) * | 2023-09-20 | 2023-10-27 | 东莞市南谷第电子有限公司 | Rapid positioning method and system for surface defect area of new energy connecting line |
| CN116958144B (en) * | 2023-09-20 | 2024-01-12 | 东莞市南谷第电子有限公司 | Rapid positioning method and system for surface defect area of new energy connecting line |
| CN117197138A (en) * | 2023-11-07 | 2023-12-08 | 深圳市极摩客科技有限公司 | Computer host shell scratch detection method based on image characteristics |
| CN117197138B (en) * | 2023-11-07 | 2024-02-23 | 深圳市极摩客科技有限公司 | Computer host shell scratch detection method based on image characteristics |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115115612B (en) | 2022-11-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115115612B (en) | Surface defect detection method and system for mechanical parts | |
| CN108921176B (en) | Pointer instrument positioning and identifying method based on machine vision | |
| CN116721106B (en) | Profile flaw visual detection method based on image processing | |
| CN109816652B (en) | Complex casting defect identification method based on gray level significance | |
| CN106404793B (en) | Bearing sealing element defect detection method based on vision | |
| CN109682839B (en) | Online detection method for surface defects of metal arc-shaped workpiece | |
| CN114842017A (en) | HDMI cable surface quality detection method and system | |
| CN108256521B (en) | Effective area positioning method for vehicle body color identification | |
| CN114913177B (en) | Automobile part defect detection method based on Hough circle | |
| CN108764345B (en) | Underwater dam crack detection method based on local and global clustering | |
| CN110070523B (en) | Foreign matter detection method for bottle bottom | |
| CN113781486B (en) | IC card image defect detection method | |
| CN112651968A (en) | Wood board deformation and pit detection method based on depth information | |
| CN113298776B (en) | Method for detecting appearance defects of metal closed water pump impeller | |
| CN115331119B (en) | Solid waste identification method | |
| CN114972892B (en) | Automobile brake pad defect classification method based on computer vision | |
| CN113177924A (en) | Industrial production line product flaw detection method | |
| CN116109637B (en) | System and method for detecting appearance defects of turbocharger impeller based on vision | |
| CN114332026A (en) | Visual detection method and device for scratch defects on surface of nameplate | |
| CN115100191A (en) | Metal casting defect identification method based on industrial detection | |
| CN107516315B (en) | Tunneling machine slag tapping monitoring method based on machine vision | |
| CN110648330A (en) | Defect detection method for camera glass | |
| CN111476804A (en) | Method, device and equipment for efficiently segmenting carrier roller image and storage medium | |
| CN110728286B (en) | Abrasive belt grinding material removal rate identification method based on spark image | |
| CN110060239B (en) | Defect detection method for bottle opening of bottle |
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 |