CN117593295A - Nondestructive testing method for production defects of mobile phone data line - Google Patents

Nondestructive testing method for production defects of mobile phone data line Download PDF

Info

Publication number
CN117593295A
CN117593295A CN202410069850.0A CN202410069850A CN117593295A CN 117593295 A CN117593295 A CN 117593295A CN 202410069850 A CN202410069850 A CN 202410069850A CN 117593295 A CN117593295 A CN 117593295A
Authority
CN
China
Prior art keywords
cluster
initial
class
merging
pixel
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202410069850.0A
Other languages
Chinese (zh)
Inventor
沈卫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Dongguan Immediately Technology Co ltd
Original Assignee
Dongguan Immediately Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Dongguan Immediately Technology Co ltd filed Critical Dongguan Immediately Technology Co ltd
Priority to CN202410069850.0A priority Critical patent/CN117593295A/en
Publication of CN117593295A publication Critical patent/CN117593295A/en
Pending legal-status Critical Current

Links

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
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • G01N21/952—Inspecting the exterior surface of cylindrical bodies or wires
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/761—Proximity, similarity or dissimilarity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • 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
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8854—Grading and classifying of flaws
    • G01N2021/8874—Taking dimensions of defect into account
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/06—Recognition of objects for industrial automation

Abstract

The invention relates to the technical field of image processing, in particular to a nondestructive testing method for production defects of a mobile phone data line, which comprises the following steps: obtaining the abrasion degree of each pixel point according to the gray level distribution of the pixel points in the local window, and obtaining the similarity degree between the pixel points according to the difference of the distance and the abrasion degree between the reference pixel point and the initial clustering center point to obtain a plurality of initial class clusters; obtaining the defect degree of each initial cluster according to the abrasion degree of the pixel points, the gray level distribution of the pixel points and the difference of the gradient directions of the pixel points, and obtaining the clustering priority characteristic of each initial cluster and the merging priority characteristic between the reference initial clusters according to the difference between the average values of the defect degrees of the initial clusters and the distance between the pixel points to obtain a plurality of termination clusters; and carrying out defect detection according to the abrasion degree. The invention determines more accurate class clusters and improves the accuracy of detecting the defects of the mobile phone data lines.

Description

Nondestructive testing method for production defects of mobile phone data line
Technical Field
The invention relates to the technical field of image processing, in particular to a nondestructive testing method for production defects of a mobile phone data line.
Background
Among the components of the data line of the mobile phone, the quality of the USB connector connected to the charger or other external devices is particularly required to be strictly controlled. The abrasion defect is one of common defects on a metal shell of the USB connector, and the defects can cause the problems of reduced power output rate, cutoff of data transmission, difficult plugging and unplugging and the like of a data line, so that the use experience of a user is greatly reduced; therefore, the quality of the data line directly affects the efficiency and the safety of the mobile phone charging and data transmission.
In the conventional USB joint defect detection of a mobile phone data line, firstly, clustering pixel points in a USB joint image by an iterative self-organizing clustering algorithm, and then determining a defect area of the USB joint according to the gray distribution difference of the pixel points in each class cluster; however, the clustering in the iterative self-organizing clustering algorithm is only performed according to the distance between the pixels in the image, so that the pixels in the defect area and the pixels in the non-defect area can be clustered in one cluster, and the accuracy of defect detection of the mobile phone data line in the production process is reduced.
Disclosure of Invention
The invention provides a nondestructive testing method for production defects of a mobile phone data line, which aims to solve the existing problems.
The invention relates to a nondestructive testing method for production defects of a mobile phone data line, which adopts the following technical scheme:
the embodiment of the invention provides a nondestructive testing method for production defects of a mobile phone data line, which comprises the following steps:
collecting USB interface images of a mobile phone data line;
obtaining a local window of each pixel point in the USB interface image, obtaining the abrasion degree of each pixel point according to the gray distribution of the pixel points in the local window of each pixel point and the difference between the average value of the gray values of all the pixel points in the local window and the average value of the gray values of all the pixel points in the USB interface image, obtaining a plurality of initial clustering center points and reference pixel points according to the size characteristics of the USB interface image, and obtaining a plurality of initial clusters according to the differences of the distances and the abrasion degrees between the reference pixel points and the initial clustering center points and the similarity degrees between each reference pixel point and each initial clustering center point;
obtaining the defect degree of each initial cluster according to the average value of the abrasion degree of all pixel points in each initial cluster, the gray level distribution of all pixel points and the gradient direction difference of all pixel points, obtaining the cluster priority feature of each initial cluster according to the defect degree of each initial cluster, the difference between the average value of the defect degree of each initial cluster and the distance between each pixel point and the initial cluster center point, recording all initial clusters adjacent to each initial cluster as the reference initial cluster corresponding to each initial cluster, obtaining the merging priority feature between each initial cluster and each corresponding reference initial cluster according to the difference of the distance and the defect degree between each initial cluster and each corresponding reference initial cluster, and obtaining a plurality of termination clusters according to the cluster priority feature of each initial cluster and the merging priority feature between each initial cluster and each corresponding reference initial cluster;
and detecting the defects of the data lines according to the abrasion degree of the pixel points in the termination cluster.
Further, the step of obtaining the local window of each pixel point in the USB interface image includes the following specific steps:
taking each pixel point in the USB interface image as a local window center point, so as toObtaining a local window of each pixel point in the USB interface image for the size of the local window; wherein A is a preset parameter.
Further, the wear degree of each pixel point is obtained according to the gray level distribution of the pixel point in the local window of each pixel point and the difference between the average value of the gray level values of all the pixel points in the local window and the average value of the gray level values of all the pixel points in the USB interface image, and the calculation formula includes:
in the method, in the process of the invention,indicate->Gray value of each pixel, +.>Indicate->The +.>Gray value of each pixel, +.>Indicate->The +.>The number of all pixels corresponding to the gray level is equal to +.>Ratio of total number of all pixels in local window of each pixel, +.>Represents a logarithmic function with base 2, +.>Representing the total number of all pixel points in the local window of each pixel point, +.>Representing the number of all gray levels in the local window of each pixel,/for each pixel>Indicate->The mean value of gray values of all pixel points in a local window of each pixel point is +.>Representing the average value of gray values of all pixel points in the USB interface image, < >>Is absolute sign, ++>Indicate->The degree of abrasion of the individual pixel points.
Further, the acquiring a plurality of initial clustering center points and reference pixel points according to the size characteristics of the USB interface image comprises the following specific steps:
equally dividing the USB interface image into B small areas, acquiring centroid points of each small area, and marking the centroid points as initial clustering center points; so far, B initial clustering center points are obtained, wherein B is a preset parameter;
and recording all pixel points except the initial clustering center point in the USB interface image as reference pixel points.
Further, the method for obtaining the similarity between each reference pixel point and each initial clustering center point according to the difference of the distance and the wear degree between the reference pixel point and the initial clustering center point, and obtaining a plurality of initial class clusters according to the similarity between each reference pixel point and each initial clustering center point comprises the following specific steps:
the calculation formula of the similarity degree between each reference pixel point and each initial clustering center point is as follows:
in the method, in the process of the invention,indicate->Reference pixel and +.>Distance between the initial cluster center points, +.>Indicate->Wear degree of each reference pixel point, +.>Indicate->Wear degree of the center points of the initial clusters, +.>As a sign of the absolute value of the sign,indicate->Reference pixel and +.>The degree of similarity between the initial cluster center points;
and acquiring an initial cluster center point corresponding to the maximum similarity between each two reference pixel points, recording the initial cluster center point as a target initial cluster center point of each reference pixel point, dividing each reference pixel point into class clusters corresponding to the target initial cluster center point, and obtaining a plurality of initial class clusters after division.
Further, the method for obtaining the defect degree of each initial cluster according to the average value of the abrasion degree of all the pixel points in each initial cluster, the gray level distribution of all the pixel points and the gradient direction difference of all the pixel points comprises the following specific steps:
obtaining gradient directions of all pixel points in the USB interface image through canny edge detection, and recording an included angle between the gradient direction of each pixel point and the horizontal right direction as a gradient included angle of each pixel point;
the calculation formula of the defect degree of each initial cluster is as follows:
in the method, in the process of the invention,indicate->The first group of groups->Degree of abrasion of individual pixels, +.>Representing the total number of all pixels in each initial cluster class, +.>Indicate->The first group of groups->The number of all pixels corresponding to the gray level is equal to +.>Ratio of total number of all pixel points in initial class cluster, +.>Represents a logarithmic function with base 2, +.>Representing the number of all gray levels in each initial cluster of classes,/-, for example>Indicate->Standard deviation of gradient included angles of all pixel points in the initial clusters,indicate->Defect level of the initial cluster.
Further, the clustering priority feature of each initial cluster is obtained according to the defect degree of each initial cluster, the difference between the average value of the defect degree of the initial cluster and the distance between each pixel point and the initial cluster center point, and the calculation formula comprises the following steps:
in the method, in the process of the invention,indicate->Defect level of the initial cluster of class, +.>Indicate->Defect level of the initial cluster of class, +.>Representing the number of all initial class clusters, +.>Indicate->The first group of groups->Distance between individual pixel points and initial cluster center point,/->Representing the total number of all pixels in each initial cluster class, +.>Indicate->First->Initial class clusters.
Further, the merging priority feature between each initial cluster and each corresponding reference initial cluster is obtained according to the difference of the distance and the defect degree between each initial cluster and each corresponding reference initial cluster, and the calculation formula is as follows:
in the method, in the process of the invention,indicate->Defect level of the initial cluster of class, +.>Indicate->The first group of clusters>Reference numberDefect level of initial cluster class,/>Indicate->Initial cluster center point of initial cluster and corresponding +.>Distance between initial cluster center points of each reference initial cluster, +.>Is absolute sign, ++>Indicate->The initial cluster and the corresponding +.>The merging priority features between the initial class clusters are referenced.
Further, the obtaining a plurality of termination class clusters according to the clustering priority feature of each initial class cluster and the merging priority feature between each initial class cluster and each corresponding reference initial class cluster comprises the following specific steps:
according to the cluster priority characteristics of each initial class cluster and the merging priority characteristics between each class cluster and each corresponding reference initial class cluster, carrying out first merging iteration of the initial class clusters, wherein the specific process is as follows:
selecting an initial cluster with the largest cluster priority characteristic, marking the initial cluster as a first target cluster, selecting a reference initial cluster with the largest merging priority characteristic with the first target cluster, and marking the reference initial cluster as a first reference initial cluster; merging the first target class cluster and the first reference initial class cluster, and marking the merged class cluster as a first merged class cluster;
all initial class clusters except the first merging class cluster are marked as first residual initial class clusters, the initial class cluster with the largest clustering priority characteristic in the first residual initial class clusters is marked as a second target class cluster, one reference initial class cluster with the largest merging priority characteristic with the second target class cluster is marked as a second reference initial class cluster in the first residual initial class clusters, the second target class cluster and the second reference initial class cluster are merged, and the merged class cluster is marked as a second merging class cluster;
all initial class clusters except the second merging class cluster are marked as second residual initial class clusters, the initial class cluster with the largest clustering priority characteristic in the second residual initial class cluster is marked as a third target class cluster, one reference initial class cluster with the largest merging priority characteristic with the third target class cluster is marked as a third reference initial class cluster in the second residual initial class cluster, the third target class cluster and the third reference initial class cluster are merged, and the merged class cluster is marked as a third merging class cluster;
and the like, obtaining all the merging clusters; at this time, the first merging iteration is completed;
marking all adjacent merging clusters of each merging cluster as reference merging clusters corresponding to each merging cluster, obtaining the clustering priority characteristics of each merging cluster according to the acquisition process of the clustering priority characteristics of each initial cluster, obtaining the merging priority characteristics between each merging cluster and each corresponding reference merging cluster according to the acquisition process of the merging priority characteristics between each initial cluster and each corresponding reference merging cluster, carrying out second merging iteration according to the process of first merging iteration, merging on the result of the previous merging in the subsequent merging process, and merging the whole merging process for G times only;
splitting the class clusters after the combination according to a splitting process in an iterative self-organizing clustering algorithm after the completion of the G times of combination, so as to obtain a plurality of termination class clusters after the splitting;
wherein G is a preset parameter.
Further, the detecting the defect of the data line according to the abrasion degree of the pixel points in the termination cluster includes the following specific steps:
acquiring the average value of the defect degrees of all the pixel points in each termination cluster, and judging the defect area when the average value of the defect degrees of all the pixel points in each termination cluster is larger than a preset threshold T; and when the average value of the defect degrees of all the pixel points in each termination cluster is smaller than or equal to a preset threshold value T, judging the normal area.
The technical scheme of the invention has the beneficial effects that: according to the invention, the wear degree of each pixel point is obtained according to the gray level distribution of the pixel points in the local window of each pixel point and the difference between the average value of the gray level values of all the pixel points in the local window and the average value of the gray level values of all the pixel points in the USB interface image, and the similarity degree between each reference pixel point and each initial clustering center point is obtained according to the difference of the distance and the wear degree between the reference pixel point and the initial clustering center point, so that the fault tolerance rate of merging clusters only according to the distance is improved;
according to the average value of the abrasion degree of all pixel points in each initial class cluster, the gray level distribution of all pixel points and the gradient direction difference of all pixel points, the defect degree of each initial class cluster is obtained, the clustering priority characteristic of each initial class cluster and the merging priority characteristic between each initial class cluster and each corresponding reference initial class cluster are obtained, merging iteration of class clusters is completed, finally, defect detection of a data line is carried out, more accurate class clusters are determined, and the defect detection accuracy of the mobile phone data line in the production process is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart showing the steps of a method for nondestructive testing of defects in a mobile phone data line according to the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a non-destructive testing method for the production defects of the mobile phone data line according to the invention, which is provided by combining the accompanying drawings and the preferred embodiment, and the specific implementation, structure, characteristics and effects thereof are described in detail below. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
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 following specifically describes a specific scheme of the nondestructive testing method for the production defects of the mobile phone data line.
Referring to fig. 1, a flowchart of a method for nondestructive testing of production defects of a mobile phone data line according to an embodiment of the invention is shown, the method includes the following steps:
step S001: and collecting USB interface images of the mobile phone data line.
In order to perform nondestructive defect detection on the mobile phone data line, it is necessary to collect an external image of the USB connector of the mobile phone data line, screen out a metal area, and perform abnormality detection on the USB connector of the mobile phone data line according to the metal area.
Specifically, an appearance image of a USB connector of a mobile phone data line is collected, gray pretreatment is carried out on the appearance image of the USB connector to obtain a USB connector gray image, a metal area in the USB connector gray image is screened out through a semantic segmentation algorithm and is recorded as a USB interface gray image, and denoising is carried out on the USB interface gray image through Gaussian filtering to obtain a denoised USB interface image. The semantic segmentation algorithm and the gaussian filtering are known techniques, and detailed descriptions thereof are omitted here.
So far, a USB interface image is obtained.
Step S002: obtaining a local window of each pixel point in the USB interface image, obtaining the abrasion degree of each pixel point according to the gray distribution of the pixel points in the local window of each pixel point and the difference between the average value of the gray values of all the pixel points in the local window and the average value of the gray values of all the pixel points in the USB interface image, obtaining a plurality of initial clustering center points and reference pixel points according to the size characteristics of the USB interface image, and obtaining a plurality of initial clusters according to the difference of the distances and the abrasion degrees between the reference pixel points and the initial clustering center points and the similarity degree between each reference pixel point and each initial clustering center point.
It should be noted that, when the metal interface of the USB has a wear defect, that is, the defective area is not smooth, the difference of gray values between the pixels in the defective area is large, and the gray distribution of the pixels in the defective area is relatively wide; the normal area is smoother, namely the gray value difference between the pixels in the normal area is smaller, the gray distribution range of the pixels in the normal area is smaller, and the gray values of the pixels are concentrated.
It should be further noted that, at the metal interface of the USB, most of the metal interface is a normal area, and the defective area is only a few partial areas, so that the difference between the average value of the gray values of all the pixels in the defective area and the average value of the gray values of all the pixels in the USB interface image is larger, and the difference between the average value of the gray values of all the pixels in the normal area and the average value of the gray values of all the pixels in the USB interface image is smaller, so that the wear degree of each pixel can be determined according to the gray distribution of the pixels in the neighborhood of each pixel and the difference between the average value of the gray values of all the pixels in the neighborhood and the average value of the gray values of all the pixels in the USB interface image.
Specifically, a parameter a is preset, where the embodiment is described by taking a=5 as an example, and the embodiment is not specifically limited, where a may be determined according to the specific implementation situation. Taking each pixel point in the USB interface image as a local window center point, so as toAnd obtaining a local window of each pixel point in the USB interface image for the local window size.
According to the gray distribution of the pixel points in the local window of each pixel point in the USB interface image and the difference between the average value of the gray values of all the pixel points in the local window and the average value of the gray values of all the pixel points in the USB interface image, the abrasion degree of each pixel point is obtained, and the abrasion degree is expressed as the formula:
in the method, in the process of the invention,indicate->Gray value of each pixel, +.>Indicate->The +.>Gray value of each pixel, +.>Indicate->The +.>The number of all pixels corresponding to the gray level is equal to +.>Ratio of total number of all pixels in local window of each pixel, +.>Represents a logarithmic function with base 2, +.>Representing the total number of all pixel points in the local window of each pixel point, +.>Representing the number of all gray levels in the local window of each pixel,/for each pixel>Indicate->The mean value of gray values of all pixel points in a local window of each pixel point is +.>Representing the average value of gray values of all pixel points in the USB interface image, < >>Is absolute sign, ++>Indicate->The degree of abrasion of the individual pixel points. One gray level is a gray value, and the division of gray levels is the same as conventional, and a specific description is not given in this embodiment.
Wherein,the average value of the difference between the gray value of each pixel point and the gray values of all the pixel points in the local window of each pixel point is represented, and when the average value is larger, the gray difference of the pixel points in the local window of the pixel point is represented to be larger, namely the possibility that the pixel point is a pixel point of a wearing area is larger; when the average value is smaller, the gray scale difference of the pixel point in the local window representing the pixel point is smaller, i.e. the pixel point is a wearing areaThe less likely the pixel points of the domain are.The gray information entropy of all the pixel points in the local window of each pixel point is represented, when the gray information entropy is larger, the gray values of the pixel points in the local window are more disordered and the distribution is wider, and the possibility that the pixel point is a pixel point of a wearing area is higher; when the smaller the gray information entropy is, the more concentrated the gray values of the pixel points in the local window are represented, the less likely the pixel points are the pixel points of the abrasion region. />Representing the difference between the average value of the gray values of all the pixel points in the local window of each pixel point and the average value of the gray values of all the pixel points in the USB interface image, and when the difference is larger, the probability that the pixel point is the pixel point of the abrasion area is larger; the smaller the difference, the less likely the pixel point is to be a worn area. Wherein the difference represents the absolute value of the difference.
So far, the abrasion degree of each pixel point in the USB interface image is obtained.
It should be noted that, when the difference of the abrasion degree between any two pixel points in the USB interface image is smaller, the probability that the two pixel points belong to the same area is greater; therefore, the similarity degree between the two pixel points can be analyzed and obtained according to the difference and the distance of the abrasion degree between the two pixel points, and the areas are a normal area and a defect area.
Specifically, a parameter B is preset, where the embodiment is described by taking b=48 as an example, and the embodiment is not limited specifically, where B may be determined according to the specific implementation. Equally dividing the USB interface image into B small areas, acquiring centroid points of each small area, and marking the centroid points as initial clustering center points; so far, B initial clustering center points are obtained.
The lower left corner pixel point of the USB interface image is taken as the origin of coordinates, the horizontal right is taken as the horizontal axis, the vertical upward is taken as the vertical axis, and a coordinate system is established, and the distance between any two pixel points can be obtained in the coordinate system, wherein in the embodiment, the distance is Euclidean distance. All pixel points except the initial clustering center point in the USB interface image are marked as reference pixel points; according to the difference of the distance and the wear degree between the reference pixel points and the initial clustering center points, the similarity degree between each reference pixel point and each initial clustering center point is obtained, and the similarity degree is expressed as follows by a formula:
in the method, in the process of the invention,indicate->Reference pixel and +.>Distance between the initial cluster center points, +.>Indicate->Wear degree of each reference pixel point, +.>Indicate->Wear degree of the center points of the initial clusters, +.>As a sign of the absolute value of the sign,indicate->Reference pixel and +.>The degree of similarity between the initial cluster center points.
Wherein,representing the difference in the degree of wear between the reference pixel point and the initial cluster center point, and when the difference is larger, representing the degree of similarity between the reference pixel point and the initial cluster center point to be smaller; the smaller the difference, the greater the degree of similarity between the reference pixel point and the initial cluster center point. When the distance between the reference pixel point and the initial clustering center point is smaller, the similarity degree between the reference pixel point and the initial clustering center point is larger; the greater the distance between the reference pixel point and the initial cluster center point, the less the degree of similarity between the reference pixel point and the initial cluster center point.
And acquiring an initial cluster center point corresponding to the maximum similarity between each reference pixel point, recording the initial cluster center point as a target initial cluster center point of each reference pixel point, dividing each reference pixel point into class clusters corresponding to the target initial cluster center point, and obtaining a plurality of initial class clusters after division.
So far, a plurality of initial class clusters are obtained.
Step S003: obtaining the defect degree of each initial cluster according to the average value of the abrasion degree of all pixel points in each initial cluster, the gray level distribution of all pixel points and the gradient direction difference of all pixel points, obtaining the cluster priority feature of each initial cluster and the merging priority feature between each initial cluster and each corresponding reference initial cluster according to the defect degree of each initial cluster, the difference between the average value of the defect degree of the initial cluster and the distance between each pixel point and the initial cluster center point, and obtaining a plurality of termination clusters according to the cluster priority feature of each initial cluster and the merging priority feature between each initial cluster and each corresponding reference initial cluster.
It should be noted that, on the surface of the USB metal connector of the data line, there are electrode holes, bonding wires and some shadow areas with poor illumination, where the areas belong to normal areas, but the pixel features in the normal areas are closer to the wear areas, if there are wear areas around the special areas, then some pixel points with insignificant features in the wear areas may be classified into the same cluster as the special areas, at this time, the feature performance of the pixels is weakened, and continuously accumulated in the subsequent iteration process, and finally larger errors are generated; the gradient directions of the pixel points in the normal area are unified, so that the gradient directions of all the pixel points in each initial cluster can be further analyzed according to the difference of the gradient directions of all the pixel points in each initial cluster, and the defect degree of each initial cluster can be obtained.
Specifically, gradient directions of all pixel points in the USB interface image are obtained through canny edge detection, and an included angle between the gradient direction of each pixel point and the horizontal right direction is recorded as a gradient included angle of each pixel point; the canny edge detection is a known technique, and detailed description thereof is omitted here. Obtaining the defect degree of each initial cluster according to the average value of the abrasion degree of all the pixel points in each initial cluster, the gray information entropy of all the pixel points and the gradient direction difference of all the pixel points, and expressing the defect degree as follows by a formula:
in the method, in the process of the invention,indicate->The first group of groups->Degree of abrasion of individual pixels, +.>Representing the total number of all pixels in each initial cluster class, +.>Indicate->The first group of groups->The number of all pixels corresponding to the gray level is equal to +.>Ratio of total number of all pixel points in initial class cluster, +.>Represents a logarithmic function with base 2, +.>Representing the number of all gray levels in each initial cluster of classes,/-, for example>Indicate->Standard deviation of gradient included angles of all pixel points in the initial clusters,indicate->Defect level of the initial cluster.
Wherein,the average value of the abrasion degree of all pixel points in each initial cluster is represented, and when the average value of the abrasion degree is larger, the defect degree of the initial cluster is represented to be larger; the smaller the average value of the wear level, the smaller the defect level representing the initial cluster-like. />The gray information entropy representing all pixels in each initial class cluster,when the gray information entropy is larger, the gray values of the pixel points in the initial class cluster are more chaotic and are distributed more widely, and the defect degree of the initial class cluster is larger; when the gray information entropy is smaller, the gray values of the pixel points in the initial class cluster are more concentrated, and the defect degree of the initial class cluster is smaller. When the gradient included angles of all the pixel points in each initial cluster are smaller in difference, namely the standard deviation of the gradient included angles of all the pixel points is smaller, the defect degree of the initial cluster is smaller, and conversely, the defect degree of the initial cluster is larger.
So far, the defect degree of each initial cluster is obtained.
It should be noted that, in order to divide the defects in the USB interface image together, the strong and weak feature conditions of each initial cluster are obtained according to the difference of the defect degrees of all the pixel points in each initial cluster, and the differences between each initial cluster and the adjacent initial clusters are combined according to the strong and weak feature conditions of each initial cluster.
Specifically, according to the defect degree of each initial cluster, the difference between the average values of the defect degrees of the initial clusters and the distance between each pixel point and the initial cluster center point, the cluster priority characteristic of each initial cluster is obtained, and is expressed as follows:
in the method, in the process of the invention,indicate->Defect level of the initial cluster of class, +.>Indicate->Defect level of the initial cluster of class, +.>Representing the number of all initial class clusters, +.>Indicate->The first group of groups->Distance between individual pixel points and initial cluster center point,/->Representing the total number of all pixels in each initial cluster class, +.>Indicate->First->Initial class clusters.
Wherein, when the cluster priority feature of each initial cluster is larger, the defect degree of the initial cluster is larger, that is, the initial cluster should be merged.
All the initial class clusters adjacent to each initial class cluster are marked as reference initial class clusters corresponding to each initial class cluster, and the merging priority characteristics between each initial class cluster and each corresponding reference initial class cluster are obtained according to the difference of the distance and the defect degree between each initial class cluster and each corresponding reference initial class cluster, and are expressed as follows by a formula:
in the method, in the process of the invention,indicate->Defect level of the initial cluster of class, +.>Indicate->The first group of clusters>Reference to the extent of defect of the original cluster of class, +.>Indicate->Initial cluster center point of initial cluster and corresponding +.>Distance between initial cluster center points of each reference initial cluster, +.>Is absolute sign, ++>Indicate->The initial cluster and the corresponding +.>The merging priority features between the initial class clusters are referenced.
Wherein, when the merging priority characteristic between each class cluster and each corresponding reference initial class cluster is larger, the two initial class clusters should be merged preferentially.
According to the cluster priority characteristics of each initial class cluster and the merging priority characteristics between each class cluster and each corresponding reference initial class cluster, carrying out first merging iteration of the initial class clusters, wherein the specific process is as follows:
selecting an initial cluster with the largest cluster priority characteristic, marking the initial cluster as a first target cluster, selecting a reference initial cluster with the largest merging priority characteristic with the first target cluster, and marking the reference initial cluster as a first reference initial cluster; merging the first target class cluster and the first reference initial class cluster, and marking the merged class cluster as a first merged class cluster;
all initial class clusters except the first merging class cluster are marked as first residual initial class clusters, the initial class cluster with the largest clustering priority characteristic in the first residual initial class clusters is marked as a second target class cluster, one reference initial class cluster with the largest merging priority characteristic with the second target class cluster is marked as a second reference initial class cluster in the first residual initial class clusters, the second target class cluster and the second reference initial class cluster are merged, and the merged class cluster is marked as a second merging class cluster;
all initial class clusters except the second merging class cluster are marked as second residual initial class clusters, the initial class cluster with the largest clustering priority characteristic in the second residual initial class cluster is marked as a third target class cluster, one reference initial class cluster with the largest merging priority characteristic with the third target class cluster is marked as a third reference initial class cluster in the second residual initial class cluster, the third target class cluster and the third reference initial class cluster are merged, and the merged class cluster is marked as a third merging class cluster;
and the like, obtaining all the merging clusters;
thus, the first merge iteration is complete.
A parameter G is preset, where the embodiment is described by taking g=10 as an example, and the embodiment is not specifically limited, where G may be determined according to the specific implementation situation. All the merging clusters adjacent to each merging cluster are marked as reference merging clusters corresponding to each merging cluster, the clustering priority characteristics of each merging cluster are obtained according to the acquisition process of the clustering priority characteristics of each initial cluster, the merging priority characteristics between each merging cluster and each corresponding reference merging cluster are obtained according to the acquisition process of the merging priority characteristics between each initial cluster and each corresponding reference merging cluster, the second merging iteration is carried out according to the process of the first merging iteration, the merging is carried out on the result of the previous merging in the subsequent merging process, and the whole merging process is only merged for G times.
Splitting the class clusters after the merging according to a splitting process in an iterative self-organizing clustering algorithm after the merging is completed, and obtaining a plurality of termination class clusters after the splitting. The splitting process in the iterative self-organizing clustering algorithm is a known technology, and detailed description thereof is omitted here.
So far, a plurality of termination class clusters are obtained.
Step S004: and detecting the defects of the data lines according to the abrasion degree of the pixel points in the termination cluster.
A threshold T is preset, where the embodiment is described by taking t=0.65 as an example, and the embodiment is not specifically limited, where T may be determined according to the specific implementation situation. Acquiring the average value of the defect degrees of all the pixel points in each termination cluster, and judging the defect area when the average value of the defect degrees of all the pixel points in each termination cluster is larger than a preset threshold T; and when the average value of the defect degrees of all the pixel points in each termination cluster is smaller than or equal to a preset threshold value T, judging the normal area.
This embodiment is completed.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.

Claims (10)

1. A nondestructive testing method for production defects of a mobile phone data line is characterized by comprising the following steps:
collecting USB interface images of a mobile phone data line;
obtaining a local window of each pixel point in the USB interface image, obtaining the abrasion degree of each pixel point according to the gray distribution of the pixel points in the local window of each pixel point and the difference between the average value of the gray values of all the pixel points in the local window and the average value of the gray values of all the pixel points in the USB interface image, obtaining a plurality of initial clustering center points and reference pixel points according to the size characteristics of the USB interface image, and obtaining a plurality of initial clusters according to the differences of the distances and the abrasion degrees between the reference pixel points and the initial clustering center points and the similarity degrees between each reference pixel point and each initial clustering center point;
obtaining the defect degree of each initial cluster according to the average value of the abrasion degree of all pixel points in each initial cluster, the gray level distribution of all pixel points and the gradient direction difference of all pixel points, obtaining the cluster priority feature of each initial cluster according to the defect degree of each initial cluster, the difference between the average value of the defect degree of each initial cluster and the distance between each pixel point and the initial cluster center point, recording all initial clusters adjacent to each initial cluster as the reference initial cluster corresponding to each initial cluster, obtaining the merging priority feature between each initial cluster and each corresponding reference initial cluster according to the difference of the distance and the defect degree between each initial cluster and each corresponding reference initial cluster, and obtaining a plurality of termination clusters according to the cluster priority feature of each initial cluster and the merging priority feature between each initial cluster and each corresponding reference initial cluster;
and detecting the defects of the data lines according to the abrasion degree of the pixel points in the termination cluster.
2. The method for non-destructive testing of production defects of a mobile phone data line according to claim 1, wherein the step of obtaining a local window of each pixel point in a USB interface image comprises the following specific steps:
taking each pixel point in the USB interface image as a local window center point, so as toObtaining a local window of each pixel point in the USB interface image for the size of the local window; wherein A is a preset parameter.
3. The method for non-destructive testing of mobile phone data line production defects according to claim 1, wherein the obtaining the abrasion degree of each pixel point according to the gray scale distribution of the pixel point in the local window of each pixel point and the difference between the average value of the gray scale values of all the pixel points in the local window and the average value of the gray scale values of all the pixel points in the USB interface image comprises the following calculation formulas:
in the method, in the process of the invention,indicate->Gray value of each pixel, +.>Indicate->The +.>Gray value of each pixel, +.>Indicate->The +.>The number of all pixels corresponding to the gray level is equal to +.>All within a local window of individual pixelsRatio of total number of pixels, +.>Represents a logarithmic function with base 2, +.>Representing the total number of all pixel points in the local window of each pixel point, +.>Representing the number of all gray levels in the local window of each pixel,/for each pixel>Indicate->The mean value of gray values of all pixel points in a local window of each pixel point is +.>Representing the average value of gray values of all pixel points in the USB interface image, < >>Is absolute sign, ++>Indicate->The degree of abrasion of the individual pixel points.
4. The method for non-destructive testing of production defects of a mobile phone data line according to claim 1, wherein the steps of obtaining a plurality of initial clustering center points and reference pixel points according to the size characteristics of the USB interface image comprise the following specific steps:
equally dividing the USB interface image into B small areas, acquiring centroid points of each small area, and marking the centroid points as initial clustering center points; so far, B initial clustering center points are obtained, wherein B is a preset parameter;
and recording all pixel points except the initial clustering center point in the USB interface image as reference pixel points.
5. The method for non-destructive testing of mobile phone data line production defects according to claim 1, wherein the steps of obtaining the similarity between each reference pixel point and each initial clustering center point according to the difference of the distance and the wear degree between the reference pixel point and the initial clustering center point and obtaining a plurality of initial clusters according to the similarity between each reference pixel point and each initial clustering center point comprise the following specific steps:
the calculation formula of the similarity degree between each reference pixel point and each initial clustering center point is as follows:
in the method, in the process of the invention,indicate->Reference pixel and +.>Distance between the initial cluster center points, +.>Indicate->Wear degree of each reference pixel point, +.>Indicate->Wear degree of the center points of the initial clusters, +.>Is absolute sign, ++>Indicate->Reference pixel and +.>The degree of similarity between the initial cluster center points;
and acquiring an initial cluster center point corresponding to the maximum similarity between each two reference pixel points, recording the initial cluster center point as a target initial cluster center point of each reference pixel point, dividing each reference pixel point into class clusters corresponding to the target initial cluster center point, and obtaining a plurality of initial class clusters after division.
6. The method for non-destructive testing of defects in mobile phone data line production according to claim 1, wherein the obtaining the defect degree of each initial cluster according to the average value of the abrasion degree of all the pixels in each initial cluster, the gray distribution of all the pixels and the gradient direction difference of all the pixels comprises the following specific steps:
obtaining gradient directions of all pixel points in the USB interface image through canny edge detection, and recording an included angle between the gradient direction of each pixel point and the horizontal right direction as a gradient included angle of each pixel point;
the calculation formula of the defect degree of each initial cluster is as follows:
in the method, in the process of the invention,indicate->The first group of groups->Degree of abrasion of individual pixels, +.>Representing the total number of all pixels in each initial cluster class, +.>Indicate->The first group of groups->The number of all pixels corresponding to the gray level is equal to +.>Ratio of total number of all pixel points in initial class cluster, +.>Represents a logarithmic function with base 2, +.>Representing the number of all gray levels in each initial cluster of classes,/-, for example>Indicate->Standard deviation of gradient included angles of all pixel points in each initial cluster>Indicate->Defect level of the initial cluster.
7. The method for non-destructive testing of defects in mobile phone data line production according to claim 1, wherein the obtaining the clustering priority feature of each initial cluster according to the defect degree of each initial cluster, the difference between the average value of the defect degree of the initial cluster and the distance between each pixel point and the initial cluster center point comprises the following calculation formula:
in the method, in the process of the invention,indicate->Defect level of the initial cluster of class, +.>Indicate->Defect level of the initial cluster of class, +.>Representing the number of all initial class clusters, +.>Indicate->Initial classesThe +.>The distance between the individual pixel points and the initial cluster center point,representing the total number of all pixels in each initial cluster class, +.>Indicate->First->Initial class clusters.
8. The method for non-destructive testing of defects in mobile phone data line production according to claim 1, wherein the obtaining the merging priority feature between each initial cluster and each corresponding reference initial cluster according to the difference of the distance and the defect degree between each initial cluster and each corresponding reference initial cluster comprises the following calculation formula:
in the method, in the process of the invention,indicate->Defect level of the initial cluster of class, +.>Indicate->The first group of clusters>Reference to the extent of defect of the original cluster of class, +.>Indicate->Initial cluster center point of initial cluster and corresponding +.>Distance between initial cluster center points of each reference initial cluster, +.>Is absolute sign, ++>Indicate->The initial cluster and the corresponding +.>The merging priority features between the initial class clusters are referenced.
9. The method for non-destructive testing of production defects of a mobile phone data line according to claim 1, wherein the obtaining a plurality of termination clusters according to the clustering priority characteristics of each initial cluster and the merging priority characteristics between each initial cluster and each corresponding reference initial cluster comprises the following specific steps:
according to the cluster priority characteristics of each initial class cluster and the merging priority characteristics between each class cluster and each corresponding reference initial class cluster, carrying out first merging iteration of the initial class clusters, wherein the specific process is as follows:
selecting an initial cluster with the largest cluster priority characteristic, marking the initial cluster as a first target cluster, selecting a reference initial cluster with the largest merging priority characteristic with the first target cluster, and marking the reference initial cluster as a first reference initial cluster; merging the first target class cluster and the first reference initial class cluster, and marking the merged class cluster as a first merged class cluster;
all initial class clusters except the first merging class cluster are marked as first residual initial class clusters, the initial class cluster with the largest clustering priority characteristic in the first residual initial class clusters is marked as a second target class cluster, one reference initial class cluster with the largest merging priority characteristic with the second target class cluster is marked as a second reference initial class cluster in the first residual initial class clusters, the second target class cluster and the second reference initial class cluster are merged, and the merged class cluster is marked as a second merging class cluster;
all initial class clusters except the second merging class cluster are marked as second residual initial class clusters, the initial class cluster with the largest clustering priority characteristic in the second residual initial class cluster is marked as a third target class cluster, one reference initial class cluster with the largest merging priority characteristic with the third target class cluster is marked as a third reference initial class cluster in the second residual initial class cluster, the third target class cluster and the third reference initial class cluster are merged, and the merged class cluster is marked as a third merging class cluster;
and the like, obtaining all the merging clusters; at this time, the first merging iteration is completed;
marking all adjacent merging clusters of each merging cluster as reference merging clusters corresponding to each merging cluster, obtaining the clustering priority characteristics of each merging cluster according to the acquisition process of the clustering priority characteristics of each initial cluster, obtaining the merging priority characteristics between each merging cluster and each corresponding reference merging cluster according to the acquisition process of the merging priority characteristics between each initial cluster and each corresponding reference merging cluster, carrying out second merging iteration according to the process of first merging iteration, merging on the result of the previous merging in the subsequent merging process, and merging the whole merging process for G times only;
splitting the class clusters after the combination according to a splitting process in an iterative self-organizing clustering algorithm after the completion of the G times of combination, so as to obtain a plurality of termination class clusters after the splitting;
wherein G is a preset parameter.
10. The method for non-destructive testing of production defects of a mobile phone data line according to claim 1, wherein the defect testing of the data line is performed according to the abrasion degree of the pixel points in the termination cluster, comprising the following specific steps:
acquiring the average value of the defect degrees of all the pixel points in each termination cluster, and judging the defect area when the average value of the defect degrees of all the pixel points in each termination cluster is larger than a preset threshold T; and when the average value of the defect degrees of all the pixel points in each termination cluster is smaller than or equal to a preset threshold value T, judging the normal area.
CN202410069850.0A 2024-01-18 2024-01-18 Nondestructive testing method for production defects of mobile phone data line Pending CN117593295A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410069850.0A CN117593295A (en) 2024-01-18 2024-01-18 Nondestructive testing method for production defects of mobile phone data line

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410069850.0A CN117593295A (en) 2024-01-18 2024-01-18 Nondestructive testing method for production defects of mobile phone data line

Publications (1)

Publication Number Publication Date
CN117593295A true CN117593295A (en) 2024-02-23

Family

ID=89911885

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410069850.0A Pending CN117593295A (en) 2024-01-18 2024-01-18 Nondestructive testing method for production defects of mobile phone data line

Country Status (1)

Country Link
CN (1) CN117593295A (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106525859A (en) * 2016-09-23 2017-03-22 广东工业大学 Method for detecting strand breakage defect of high-voltage electric transmission wire by unmanned aerial vehicle in real time
US20210041375A1 (en) * 2018-04-13 2021-02-11 Maillefer S.A. A surface scanner, an arrangement and a method for surface defect detection of a cable
CN113777030A (en) * 2021-07-08 2021-12-10 杭州信畅信息科技有限公司 Cloth surface defect detection device and method based on machine vision
CN115035113A (en) * 2022-08-11 2022-09-09 深圳市通泰详特种电线电缆有限公司 Nondestructive testing method for production defects of mobile phone data line based on X-ray
CN117076960A (en) * 2023-08-16 2023-11-17 成都数之联科技股份有限公司 Panel defect point aggregation detection method, system, equipment and storage medium
CN117351214A (en) * 2023-12-06 2024-01-05 上海电气集团腾恩驰科技(苏州)有限公司 Heat tracing cable detection method and system based on image processing

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106525859A (en) * 2016-09-23 2017-03-22 广东工业大学 Method for detecting strand breakage defect of high-voltage electric transmission wire by unmanned aerial vehicle in real time
US20210041375A1 (en) * 2018-04-13 2021-02-11 Maillefer S.A. A surface scanner, an arrangement and a method for surface defect detection of a cable
CN113777030A (en) * 2021-07-08 2021-12-10 杭州信畅信息科技有限公司 Cloth surface defect detection device and method based on machine vision
CN115035113A (en) * 2022-08-11 2022-09-09 深圳市通泰详特种电线电缆有限公司 Nondestructive testing method for production defects of mobile phone data line based on X-ray
CN117076960A (en) * 2023-08-16 2023-11-17 成都数之联科技股份有限公司 Panel defect point aggregation detection method, system, equipment and storage medium
CN117351214A (en) * 2023-12-06 2024-01-05 上海电气集团腾恩驰科技(苏州)有限公司 Heat tracing cable detection method and system based on image processing

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
谢娟英;高红超;谢维信;: "K近邻优化的密度峰值快速搜索聚类算法", 中国科学:信息科学, no. 02, 20 February 2016 (2016-02-20), pages 258 - 280 *

Similar Documents

Publication Publication Date Title
CN111681197B (en) Remote sensing image unsupervised change detection method based on Siamese network structure
CN114170228B (en) Computer image edge detection method
CN103914683A (en) Gender identification method and system based on face image
CN116721107B (en) Intelligent monitoring system for cable production quality
CN114820625B (en) Automobile top block defect detection method
CN111667470B (en) Industrial pipeline flaw detection inner wall detection method based on digital image
CN115272312B (en) Plastic mobile phone shell defect detection method based on machine vision
CN108665464A (en) A kind of foreign matter detecting method based on morphologic high tension electric tower and high-tension bus-bar
CN115330767A (en) Method for identifying production abnormity of corrosion foil
CN112232356A (en) Event camera denoising method based on cluster degree and boundary characteristics
CN115311262A (en) Printed circuit board defect identification method
CN107578011A (en) The decision method and device of key frame of video
CN112052730B (en) 3D dynamic portrait identification monitoring equipment and method
CN114863492B (en) Method and device for repairing low-quality fingerprint image
CN113920080A (en) Power grid fault positioning method based on generation countermeasure network
CN112508913A (en) Cable section edge detection method based on image detection
CN115471476A (en) Method, device, equipment and medium for detecting component defects
CN116703925B (en) Bearing defect detection method and device, electronic equipment and storage medium
CN112288780B (en) Multi-feature dynamically weighted target tracking algorithm
CN116432052B (en) Quality detection method for clamp for new energy automobile die
CN117593295A (en) Nondestructive testing method for production defects of mobile phone data line
CN115311456A (en) Tuyere coke segmentation method based on improved DeepLabv3+
CN110349129B (en) Appearance defect detection method for high-density flexible IC substrate
CN113554036A (en) Characteristic point extraction and matching method for improving ORB algorithm
CN115187604B (en) Contour line-based composite wall plate array hole detection method, device and system

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