CN118887217B - Test box surface weld quality detection method and system - Google Patents
Test box surface weld quality detection method and system Download PDFInfo
- Publication number
- CN118887217B CN118887217B CN202411382020.XA CN202411382020A CN118887217B CN 118887217 B CN118887217 B CN 118887217B CN 202411382020 A CN202411382020 A CN 202411382020A CN 118887217 B CN118887217 B CN 118887217B
- Authority
- CN
- China
- Prior art keywords
- welding
- image
- characteristic
- edge
- connected domain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; 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 OR CALCULATING; 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 OR CALCULATING; 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 OR CALCULATING; 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
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20076—Probabilistic image processing
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06T2207/30152—Solder
-
- 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)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to the field of image processing, in particular to a method and a system for detecting the quality of a welding line on the surface of a test box, wherein the method comprises the steps of obtaining a welding line image of the test box to obtain a welding line edge image, segmenting according to the welding line direction, and obtaining an edge communicating domain of welding points in each segment; the method comprises the steps of calculating deformation probability of each pixel point in an edge connected domain, obtaining characteristic points on the edge connected domain of each welding spot, constructing a characteristic matrix to obtain characteristic values of each welding spot, forming characteristic value sequences by the characteristic values of the welding spots in each welding seam image, clustering, calculating abnormality indexes of each category, obtaining the maximum value of the abnormality indexes in all the categories as the abnormality index of the segmented welding seam image, and displaying abnormality of the welding seam when the abnormality index is larger than or equal to an abnormality threshold value, otherwise, judging the welding seam to be qualified. According to the invention, the characteristic value of each welding spot is obtained through the texture information of the welding spot, and the accuracy of welding seam detection is improved.
Description
Technical Field
The present invention relates to the field of image processing. More particularly, the invention relates to a test box surface weld quality detection method and system.
Background
Test chambers are commonly used in engineering or scientific research, and refer to devices or systems used to conduct experiments or tests, typically to simulate specific environmental conditions or to conduct control experiments.
The prior China patent application document with the publication number of CN115294034A discloses a welding seam quality detection method and a training method, more information is acquired through a first normal angle diagram, the carried information is more accurate, the detection precision of welding seam quality detection can be improved, and the problems of high misjudgment rate and high missed judgment rate when a detected object with high surface reflectivity detects welding seam quality are solved.
In addition to the problems in the above-mentioned scheme, in actual production, the welding bead is partially askew and uneven welding spots may occur under the condition that the welding bead is kept generally straight due to the local disturbance to which the mechanical arm is subjected. Therefore, when quality detection is carried out on the welding seam on the surface of the test box, poor welding seam formation often occurs, a welding bead formed by a welding interface is in a snake shape, and fish scale lines formed by overlapping welding spots are uneven and irregular. The serpentine structure may cause non-uniformity in the metal structure of the weld area, thereby affecting the mechanical properties of the welded structure, reducing its load carrying capacity and fatigue resistance. For the welded structure of the test box, which needs to be sealed, the poor formation of the welding seam can cause leakage, and potential safety hazards exist.
In addition to the above scheme, a common method for detecting whether the running direction of the welding seam is straight is generally a hough straight line detection algorithm, but on one hand, the hough straight line detection cannot capture such local welding bead bending, is insensitive to local welding seam direction variation, and on the other hand, the line information of the welding spot cannot be analyzed, so that the welding seam quality detection result is inaccurate.
Disclosure of Invention
In order to solve the problem that the Hough straight line detection cannot capture the local weld bead bending and the texture information of the welding spot cannot be analyzed, the invention provides the following aspects.
In a first aspect, a test box surface weld quality detection method includes obtaining a weld image of a test box, preprocessing to obtain a weld edge image, segmenting the weld edge image according to a weld direction, obtaining edge connected domains of welding spots in each segment, calculating difference values of curvature values of all pixel points in the edge connected domains to obtain deformation probability of each pixel point, obtaining feature points on the edge connected domains of each welding spot, constructing a feature matrix, performing data dimension reduction on the feature matrix to obtain feature values of each welding spot, constructing a feature value sequence of the feature values of the welding spots in each segment of the weld image, clustering to obtain a plurality of categories, calculating abnormality indexes of each category,In which, in the process,Is the firstThe abnormality index of the individual category(s),Is the firstThe maximum number of consecutive welds of each category,For the maximum continuous number average for all classes,Is the firstThe mean of the characteristic values of the welding spots of the largest continuous segment in each category,For the mean of the largest continuous segment feature values in all classes,And (3) for normalizing the function, obtaining the maximum value of the abnormality indexes in all the categories as the abnormality index of the segmented weld image, and displaying abnormality on the weld when the abnormality index is greater than or equal to an abnormality threshold value, otherwise, qualifying the weld.
The method has the advantages that the geometric features and the edge shape of the welding line are extracted by preprocessing the welding line image of the test box, the welding line edge image is segmented, the welding point and the surrounding local structure are positioned, the deformation degree of the welding point is evaluated according to the deformation probability of the welding point, the quality of the welding line is judged, the characteristic values of the welding line are clustered, the abnormal index of the welding line is obtained, the automatic detection and evaluation of the quality of the welding line on the surface of the test box are realized, and the method has higher efficiency and accuracy.
In one embodiment, preprocessing the weld image to obtain a weld edge image, segmenting the weld edge image according to a weld direction, and obtaining an edge connected domain of a welding spot in each segment, including:
Denoising and graying the weld joint image, and performing edge detection on a gray scale image to obtain a weld joint edge image;
Segmenting the welding seam edge image according to the welding seam direction to obtain a welding image of each segment, and performing mask operation on the welding image to obtain a welding area of the welding image;
And extracting the central line of the welding area, and obtaining the vertical line direction of the welding direction corresponding to each welding spot on the central line to perform edge detection so as to obtain the edge connected domain of each welding spot.
The method has the advantages that noise in the weld image is eliminated, the color image is converted into a gray image, various filters such as Gaussian filters or median filters can be used for denoising, the weld edge is obtained through edge detection, skeletonizing or other morphological operations are used for extracting the center line of a welding area, wherein the center line represents the main axis of the weld, edge detection is carried out along the direction perpendicular to the center, and therefore the edge connected domain of the welding point in each section is obtained. And the shape, the size and the distribution of welding spots and the quality and the characteristics of welding seams are further analyzed.
In one embodiment, obtaining the deformation probability comprises:
And calculating the distance between each pixel point in the edge connected domain and the central coordinate, taking the difference between the curvatures of the left and right adjacent pixel points of any pixel point in the calculated edge connected domain as a curvature change value, and taking the product of the distance and the curvature change value as the deformation probability of the welding spot.
The method has the advantages that the deformation condition around the welding spot can be reflected through the calculation, the pixel points with larger distance and smaller curvature change value can have lower deformation probability, the pixel points with smaller distance and larger curvature change value can have higher deformation probability, the deformation degree of the welding spot can be evaluated, and further welding quality and possible defects can be analyzed.
In one embodiment, the distance between each pixel point in the edge connected domain and the center coordinate includes:
and acquiring the average value of the pixel points in the edge connected domain as a center coordinate, and calculating the Euclidean distance from each pixel point in the edge connected domain to the center point coordinate.
In one embodiment, the obtaining the feature points on the edge connected domain of each welding spot, and constructing the feature matrix include:
taking 2 pixel points with the maximum deformation probability in the obtained edge connected domain as characteristic points, dividing the edge connected domain into two sections by the characteristic points, taking the sequential midpoints in the two sections of connected domain as other 2 characteristic points, and extracting 4 characteristic points of each welding point;
and constructing a 4 x2 feature matrix for each welding spot by taking the abscissa and the ordinate of the 4 feature points as two attribute values of the feature, wherein the feature matrix is composed of 4 feature points and 2 attribute values thereof.
The method has the advantages that two pixel points with the largest deformation probability generally represent the largest deformation or inflection point of the edge connected domain, the scene in the embodiment is combined, the edge of each welding point is approximately crescent, the two pixel points with the largest deformation probability of the welding point are generally arranged at the two crescent ends of the welding point, and the average position of the pixel points in each connected domain is obtained and used as two additional characteristic points. These feature points represent the weld center around each weld, and help to analyze the shape and position characteristics of each weld to obtain a feature value for each weld.
In one embodiment, the forming and clustering the characteristic values of the welding spots in each welding seam image into the characteristic value sequence includes:
Normalizing the characteristic value sequence, and classifying the normalized characteristic value sequence;
And constructing a histogram by taking the characteristic value sequence as an abscissa and the number as an ordinate, and taking the number of peak points in the obtained histogram as the number of clusters of the welding spot characteristic value sequence to obtain a plurality of categories.
The method has the advantages that the characteristic value sequences are normalized to ensure that the calculation scales are the same, and the characteristic value sequences are clustered to obtain welding spots with similar characteristic values to be divided into a group, so that the analysis of the distribution condition of the welding spots is facilitated.
In one embodiment, the calculating the abnormality index of each category further includes:
Extracting the lengths of the continuous fragments in each category, calculating the average continuous fragment length of the continuous fragments in each category, calculating the square of the difference between the lengths of the continuous fragments and the average continuous fragment length, and calculating the ratio of the sum of square differences of all the categories to the number of the continuous fragments to obtain the continuous length variance;
And taking the continuous length variance as a parameter for calculating the abnormality index of each category.
The effect is that the continuous length variance represents the degree of variation of the continuous segment length in each category by taking the continuous length variance as the adjustment parameter of the abnormality index of the category. A large variance in continuous length may indicate uneven or unstable solder joint distribution in this category.
In a second aspect, a test box surface weld quality detection system includes a processor and a memory storing computer program instructions that when executed by the processor implement the test box surface weld quality detection method described above.
The invention has the following effects:
1. According to the invention, the welding lines are segmented by utilizing the line information of the welding points, and the characteristic value of each welding point is obtained, so that the abnormal region in the welding line can be accurately identified, and the accuracy of welding line quality detection is improved.
2. Compared with the traditional Hough straight line detection algorithm, the method is more sensitive to the small change of the local part of the welding seam, and can capture the local bending of the welding bead and the uneven distribution of welding spots, so that the quality of the welding seam is better estimated.
3. According to the invention, the welding point characteristic values can be rapidly classified by a minimum and maximum value normalization method and a k-means algorithm, and the abnormal indexes of the welding lines are calculated, so that the abnormal welding lines can be rapidly positioned.
Drawings
The above, as well as additional purposes, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description when read in conjunction with the accompanying drawings. In the drawings, embodiments of the invention are illustrated by way of example and not by way of limitation, and like reference numerals refer to similar or corresponding parts and in which:
FIG. 1 is a flow chart of a method for detecting the quality of a weld on the surface of a test chamber in accordance with an embodiment of the present invention, from step S1 to step S4.
Fig. 2 is a flowchart of a method of step S10-step S12 in a test chamber surface weld quality detection method according to an embodiment of the present invention.
Fig. 3 is a flowchart of a method of step S30-step S32 in a test chamber surface weld quality detection method according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of feature points of an extracted welding spot in a method for detecting quality of a welding seam on a surface of a test box according to an embodiment of the invention.
FIG. 5 is a flowchart of a method for detecting the quality of a weld on a surface of a test chamber in accordance with an embodiment of the present invention, from step S40 to step S42.
FIG. 6 is a block diagram of a test chamber surface weld quality inspection system in accordance with an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.
Referring to fig. 1, a method for detecting quality of a weld on a surface of a test chamber includes steps S1 to S4, specifically as follows:
The invention aims at the specific scene that the surface of the welding line in the test box is uniform in light, the camera is perpendicular to the welding line, and the shot welding line image has uniform scale patterns and welding point edges.
S1, acquiring a welding line image of a test box, preprocessing to obtain a welding line edge image, segmenting the welding line edge image according to the welding line direction, and acquiring an edge connected domain of a welding point in each segment.
Further, the segmentation of the weld seam image is performed by those skilled in the art because the welding direction of the welding position of the test box may be different, which results in poor edge detection results, affects the subsequent analysis of the abnormal condition of each welding spot, and determines the welding direction of each segment.
Referring to fig. 2, the method includes steps S10 to S12:
S10, denoising and graying treatment are carried out on the weld joint image, and edge detection is carried out on the gray scale image to obtain a weld joint edge image;
Further, the common denoising method includes median filtering, gaussian filtering and the like to eliminate noise in an image, the denoised weld joint image is converted into a single-channel gray image, an edge detection algorithm is used for obtaining an edge image, and the edge detection algorithm can use a Sobel operator, canny edge detection and the like.
S11, segmenting a welding seam edge image according to the welding seam direction to obtain a welding image of each segment, and performing mask operation on the welding image to obtain a welding area of the welding image;
Further, the welding image is subjected to a mask operation, wherein the mask is formed by setting the pixel value of the non-welding area to 0 (black), setting the pixel area of the welding area to 255 (white), and obtaining a binary image so as to reserve the welding area.
And S12, extracting the central line of the welding area, and obtaining the vertical line direction of the welding direction corresponding to each welding point on the central line to perform edge detection so as to obtain the edge connected area of each welding point.
S2, calculating the difference value of the curvature values of the pixel points in the edge connected domain to obtain the deformation probability of each pixel point.
And obtaining the average value of the pixel points in the edge connected domain as a center coordinate, calculating the Euclidean distance from each pixel point in the edge connected domain to the center point coordinate, taking the difference between the curvatures of the left and right adjacent pixel points of any pixel point in the calculated edge connected domain as a curvature change value, and taking the product of the distance and the curvature change value as the deformation probability of the welding point.
Specifically, the first edge connected domain is acquiredCenter point coordinates of each welding pointCalculate the firstOn the edge of each welding spotThe coordinates of each pixel point areCoordinates of pixel point to center point of (c)Is the Euclidean distance of (2)。
Acquisition of the firstOn the edge of each welding spotLeft and right sides of each pixel pointAnd (b)Curvature values of the pixel points are calculated, and curvature difference values are calculated to obtain curvature change values。
Specifically, the curvature change value satisfies the following relation:
,
In the formula, Represent the firstOn the edge of each welding spotThe curvature change value of each pixel point,Respectively the current firstLeft side of each pixel pointCurvature value and right side of each pixel pointThe curvature value of each pixel point, namely the curvature change value of the current pixel point is the difference between the curvatures of the adjacent left and right pixel points.
The curvature values of the pixel points are calculated using a Hessian Matrix (Hessian Matrix), which is a square Matrix containing information of second partial derivatives, and is generally used to describe information of second partial derivatives of a multivariate function. This technique is well known to those skilled in the art and will not be described in detail.
And then according to the Euclidean distanceAnd curvature change valueThe product of (2) yields the deformation probability。
It should be noted that, the edge lengths of each welding spot are different, so as to unify the dimensions of the welding spot edge and reduce subsequent calculation, further extracting feature points on the welding spot edge is considered.
And S3, obtaining characteristic points on the edge connected domain of each welding point, constructing a characteristic matrix, and performing data dimension reduction on the characteristic matrix to obtain the characteristic value of each welding point.
Referring to fig. 3, the method includes steps S30 to S32:
s30, taking 2 pixel points with the maximum deformation probability in the obtained edge connected domain as characteristic points, wherein the characteristic points divide the edge connected domain into two sections, taking the sequential midpoints in the two sections of connected domain as other 2 characteristic points, and extracting 4 characteristic points of each welding point;
Referring to FIG. 4, an edge connected region of a part of welding spots in the segment is obtained, wherein a first welding spot is generally circular, and edge profiles of other welding spots except the first welding spot are generally crescent-shaped to form a third welding spot Taking a welding spot as an example, obtaining the maximum deformation probability in the edge connected domain of the welding spotTwo pixel points, wherein the maximum deformation probability generally refers to the maximum deformation of the curvature of the pixel points in the edge contour or expressed as inflection points, and the maximum deformation probability is obtainedMidpoint of both sides, getAnd further obtaining 4 characteristic points of the edge connected domain of each welding spot, wherein the first welding spot obtains round quarter points, namely 4 characteristic points of the welding spot.
S31, taking the abscissa and the ordinate of the obtained 4 feature points as two attribute values of the feature, and constructing a4 multiplied by 2 feature matrix for each welding spot, wherein the feature matrix is composed of 4 feature points and 2 attribute values thereof.
S32, traversing each welding spot, reducing the dimension of the characteristic matrix of the welding spot, wherein one welding spot corresponds to one characteristic value.
Further, in this embodiment, PAC (Patch-based Adaptive Compression) algorithm is adopted, which is an adaptive block compression algorithm for image compression. The PAC algorithm maps the feature matrix of each welding spot to a space with a lower dimension, and retains important feature information, and in addition, the dimension reduction may be implemented by using methods such as principal component analysis (PRINCIPAL COMPONENT ANALYSIS, PCA) and linear discriminant analysis (LINEAR DISCRIMINANT ANALYSIS, LDA).
S4, forming a characteristic value sequence by characteristic values of welding spots in each section of welding seam image, clustering to obtain a plurality of categories, calculating abnormality indexes of each category, obtaining the maximum value of the abnormality indexes in all the categories as the abnormality index of the segmented welding seam image, and displaying abnormality of the welding seam when the abnormality index is larger than or equal to an abnormality threshold value, otherwise, qualifying the welding seam.
Further to the description, in the present embodiment, the abnormality threshold is set to 20.
Referring to fig. 5, steps S40 to S42 are included:
s40, carrying out normalization processing on the characteristic value sequence, and classifying the normalized characteristic value sequence;
In this embodiment, a maximum and minimum normalization method is adopted for processing, if welding spots with similar characteristic values are adjacent on the sequence numbers before normalization, the welding spots with similar characteristic values after normalization are not adjacent on the sequence numbers, and should be uniformly dispersed in the sequence. Therefore, if a local continuous characteristic value approximate sequence appears after normalization and the sequence segment has a larger difference from the overall characteristic value, the segment is considered to be an abnormal part.
S41, constructing a histogram by taking the characteristic value sequence as an abscissa and the number as an ordinate, and taking the number of peak points in the acquired histogram as the number of clusters of the welding spot characteristic value sequence to obtain a plurality of categories;
further, the number of clusters is determined in a self-adaptive mode according to the distribution condition of the characteristic value sequences, and the accuracy and stability of the clusters are improved.
S42, calculating abnormality indexes of all categories;
specifically, the abnormality indexes of the respective categories satisfy the following relational expression:
,
In the formula, Is the firstThe abnormality index of the individual category(s),Is the firstThe maximum number of consecutive welds of each category,For the maximum continuous number average for all classes,Is the firstThe mean of the characteristic values of the welding spots of the largest continuous segment in each category,For the mean of the largest continuous segment feature values in all classes,Is a normalization function.
As a further explanation of the present invention,The larger the value of the sequence of the welding spots is, the more the welding spot sequence fragments belong to the same class and the continuous number is, the larger the abnormal probability of the region is; The larger the difference between the continuous number of the fragments and the average continuous number is larger; the larger the difference between the morphological feature value and the overall feature of the segment, and therefore, AndThe larger the value of (c), the more likely an anomaly is to occur for the largest consecutive segment in the class.
In addition, in another embodiment, the lengths of the continuous segments in each category are extracted, the average continuous segment length of each category continuous segment is calculated, the square of the difference between the lengths of the continuous segments and the average continuous segment length is calculated, the ratio of the sum of the square variances of all the categories to the number of the continuous segments is calculated, the continuous length variance is obtained, and the continuous length variance is used as a parameter for calculating the abnormality index of each category.
Specifically, the abnormality indexes of the respective categories satisfy the following relational expression:
,
In the formula, Is the firstThe abnormality index of the individual category(s),Is the firstThe maximum number of consecutive welds of each category,For the maximum continuous number average for all classes,Is the firstThe mean of the characteristic values of the welding spots of the largest continuous segment in each category,For the mean of the largest continuous segment feature values in all classes,As a function of the normalization,Is the firstVariance of maximum continuous number of welds for each category.
The degree of variation in the length of the continuous segments in the category may be represented by a continuous length variance, a larger continuous length variance may represent uneven or unstable distribution of welds in the category, and a larger continuous length variance may result in an increased anomaly index for the category, thereby identifying irregular or anomalous regions in the weld.
The invention is completed so as to obtain the normal or abnormal condition of the welding line.
The invention also provides a test box surface weld quality detection system. As shown in fig. 6, the system comprises a processor and a memory storing computer program instructions which, when executed by the processor, implement a method of testing the quality of a surface weld of a test chamber according to the first aspect of the invention.
The system further comprises other components known to those skilled in the art, such as communication buses and communication interfaces, the arrangement and function of which are known in the art and therefore will not be described in detail herein.
In the context of this patent, the foregoing memory may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, the computer-readable storage medium may be any suitable magnetic or magneto-optical storage medium, such as, for example, resistance change Memory RRAM (Resistive Random Access Memory), dynamic Random Access Memory DRAM (Dynamic Random Access Memory), static Random Access Memory SRAM (Static Random-Access Memory), enhanced dynamic Random Access Memory EDRAM (ENHANCED DYNAMIC Random Access Memory), high-Bandwidth Memory HBM (High-Bandwidth Memory), hybrid storage cube HMC (Hybrid Memory Cube), or the like, or any other medium that may be used to store the desired information and that may be accessed by an application, a module, or both. Any such computer storage media may be part of, or accessible by, or connectable to, the device. Any of the applications or modules described herein may be implemented using computer-readable/executable instructions that may be stored or otherwise maintained by such computer-readable media.
In the description of the present specification, the meaning of "a plurality", "a number" or "a plurality" is at least two, for example, two, three or more, etc., unless explicitly defined otherwise.
While various embodiments of the present invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Many modifications, changes, and substitutions will now occur to those skilled in the art without departing from the spirit and scope of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention.
Claims (5)
1. The method for detecting the quality of the welding line on the surface of the test box is characterized by comprising the following steps:
acquiring a welding seam image of a test box, preprocessing to obtain a welding seam edge image, segmenting the welding seam edge image according to the welding seam direction, and acquiring an edge communication domain of a welding spot in each segment;
Calculating the distance between each pixel point in the edge connected domain and the central coordinate, taking the difference between the curvatures of the left and right adjacent pixel points of any pixel point in the calculated edge connected domain as a curvature change value, and taking the product of the distance and the curvature change value as the deformation probability of a welding spot;
the method comprises the steps of taking 2 pixel points with maximum deformation probability in an edge connected domain as characteristic points, dividing the edge connected domain into two sections by the characteristic points, taking sequential midpoints in the two sections of connected domain as other 2 characteristic points, extracting 4 characteristic points of each welding point, taking the abscissa and the ordinate of the 4 characteristic points as two attribute values of the characteristics, constructing a4 multiplied by 2 characteristic matrix for each welding point, wherein the characteristic matrix is 4 characteristic points and 2 attribute values thereof, and performing data dimension reduction on the characteristic matrix to obtain the characteristic value of each welding point;
Forming characteristic value sequences by characteristic values of welding spots in each section of welding seam image, clustering to obtain a plurality of categories, calculating abnormality indexes of each category, In which, in the process,Is the firstThe abnormality index of the individual category(s),Is the firstThe maximum number of consecutive welds of each category,For the maximum continuous number average for all classes,Is the firstThe mean of the characteristic values of the welding spots of the largest continuous segment in each category,For the mean of the largest continuous segment feature values in all classes,And (3) for normalizing the function, obtaining the maximum value of the abnormality indexes in all the categories as the abnormality index of the segmented weld image, and displaying abnormality on the weld when the abnormality index is greater than or equal to an abnormality threshold value, otherwise, qualifying the weld.
2. The method for detecting the quality of a welding seam on the surface of a test box according to claim 1, wherein the method for detecting the quality of the welding seam on the surface of the test box is characterized in that the welding seam image is preprocessed to obtain a welding seam edge image, the welding seam edge image is segmented according to the welding seam direction, and an edge connected domain of welding spots in each segment is obtained, and the method comprises the following steps:
Denoising and graying the weld joint image, and performing edge detection on a gray scale image to obtain a weld joint edge image;
Segmenting the welding seam edge image according to the welding seam direction to obtain a welding image of each segment, and performing mask operation on the welding image to obtain a welding area of the welding image;
And extracting the central line of the welding area, and obtaining the vertical line direction of the welding direction corresponding to each welding spot on the central line to perform edge detection so as to obtain the edge connected domain of each welding spot.
3. The method for detecting the quality of a weld on a surface of a test chamber according to claim 1, wherein obtaining the distance between each pixel point in the edge connected domain and the center coordinates comprises:
and acquiring the average value of the pixel points in the edge connected domain as a center coordinate, and calculating the Euclidean distance from each pixel point in the edge connected domain to the center point coordinate.
4. The method for detecting the quality of welding seams on the surface of a test box according to claim 1, wherein the step of forming and clustering the characteristic values of welding spots in each welding seam image into a characteristic value sequence comprises the following steps:
Normalizing the characteristic value sequence, and classifying the normalized characteristic value sequence;
And constructing a histogram by taking the characteristic value sequence as an abscissa and the number as an ordinate, and taking the number of peak points in the obtained histogram as the number of clusters of the welding spot characteristic value sequence to obtain a plurality of categories.
5. A test box surface weld quality detection system comprising a processor and a memory, the memory storing computer program instructions that when executed by the processor implement the test box surface weld quality detection method of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411382020.XA CN118887217B (en) | 2024-09-30 | 2024-09-30 | Test box surface weld quality detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411382020.XA CN118887217B (en) | 2024-09-30 | 2024-09-30 | Test box surface weld quality detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118887217A CN118887217A (en) | 2024-11-01 |
| CN118887217B true CN118887217B (en) | 2024-12-13 |
Family
ID=93234700
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411382020.XA Active CN118887217B (en) | 2024-09-30 | 2024-09-30 | Test box surface weld quality detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118887217B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119625245B (en) * | 2024-12-09 | 2025-06-20 | 四川福骏汽车制造有限公司 | A truck steel structure quality prediction method and system based on beam welding results |
| CN119741298B (en) * | 2025-03-04 | 2025-05-16 | 陕西金鑫电器有限公司 | Weld joint polishing quality detection method for welding shell |
| CN120807510B (en) * | 2025-09-12 | 2025-12-05 | 陕西金鑫电器有限公司 | An image localization method for defects in welded joints of aluminum alloy shells |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115423813A (en) * | 2022-11-05 | 2022-12-02 | 江苏惠汕新能源集团有限公司 | Method for detecting welding defects on surface of welded pipe |
| WO2023045237A1 (en) * | 2021-09-27 | 2023-03-30 | 深圳市联赢激光股份有限公司 | Intelligent welding method, intelligent welding system, and computer storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7628932B2 (en) * | 2021-10-20 | 2025-02-12 | 株式会社神戸製鋼所 | Welding monitoring method and welding monitoring device, and additive manufacturing method and additive manufacturing device |
-
2024
- 2024-09-30 CN CN202411382020.XA patent/CN118887217B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023045237A1 (en) * | 2021-09-27 | 2023-03-30 | 深圳市联赢激光股份有限公司 | Intelligent welding method, intelligent welding system, and computer storage medium |
| CN115423813A (en) * | 2022-11-05 | 2022-12-02 | 江苏惠汕新能源集团有限公司 | Method for detecting welding defects on surface of welded pipe |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118887217A (en) | 2024-11-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118887217B (en) | Test box surface weld quality detection method and system | |
| CN116168026B (en) | Water quality detection method and system based on computer vision | |
| CN118379286B (en) | Surface defect detection method and system for titanium alloy forging | |
| CN115690108A (en) | Aluminum alloy rod production quality evaluation method based on image processing | |
| CN119991544B (en) | Intelligent classification method and system for strain clamp defects of power transmission line based on two-dimensional image | |
| CN118334029B (en) | Titanium ingot production quality detection method and system based on image processing | |
| CN117893553B (en) | Image processing titanium metal segmentation method and system | |
| CN119417817B (en) | Paint defect detection method and system based on machine vision | |
| CN120894328B (en) | A wafer edge defect detection method, system, and storage medium based on image recognition. | |
| CN118644456B (en) | Workpiece surface defect detection and early warning method based on 3D vision | |
| CN119741303B (en) | Intelligent detection method and system for blade castings | |
| CN119741298B (en) | Weld joint polishing quality detection method for welding shell | |
| CN118570206B (en) | Capillary production data monitoring system and method | |
| CN118967678A (en) | A method and system for automobile parts production detection based on machine learning | |
| CN119313697A (en) | A method for evaluating laser welding quality | |
| CN116703883A (en) | Method, device and electronic equipment for determining pinhole defects after power battery welding | |
| CN118365867B (en) | Titanium alloy bar identification method and system based on image analysis | |
| CN119147429B (en) | Intelligent dust concentration detection method and system | |
| CN120107297A (en) | A method and device for detecting the contour of a plunger ball head | |
| CN119417839A (en) | A system and method for detecting foreign matter content in metal powder | |
| CN119444731A (en) | Plate processing drilling defect recognition method and system based on machine vision | |
| CN119169035A (en) | A tire pressure calibration method based on image analysis | |
| CN117876468A (en) | A method and system for detecting characteristic points of low beam light pattern of automobile lights based on convex hull | |
| CN120598944B (en) | Crankshaft surface defect detection method and system based on image processing | |
| Zhou et al. | Research on traditional image segmentation method based on oil drilling pipe defects |
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 |