CN117314925A - Metal workpiece surface defect detection method based on computer vision - Google Patents
Metal workpiece surface defect detection method based on computer vision Download PDFInfo
- Publication number
- CN117314925A CN117314925A CN202311620035.0A CN202311620035A CN117314925A CN 117314925 A CN117314925 A CN 117314925A CN 202311620035 A CN202311620035 A CN 202311620035A CN 117314925 A CN117314925 A CN 117314925A
- Authority
- CN
- China
- Prior art keywords
- edge
- image
- detected
- suspected
- air hole
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000007547 defect Effects 0.000 title claims abstract description 45
- 229910052751 metal Inorganic materials 0.000 title claims abstract description 23
- 239000002184 metal Substances 0.000 title claims abstract description 23
- 238000001514 detection method Methods 0.000 title claims abstract description 19
- 229910001297 Zn alloy Inorganic materials 0.000 claims abstract description 61
- 238000012937 correction Methods 0.000 claims abstract description 30
- 239000011148 porous material Substances 0.000 claims description 57
- 238000000034 method Methods 0.000 claims description 27
- 238000004364 calculation method Methods 0.000 claims description 14
- 238000004220 aggregation Methods 0.000 claims description 7
- 230000002776 aggregation Effects 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 7
- 230000002950 deficient Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 2
- 238000009826 distribution Methods 0.000 description 4
- 238000003708 edge detection Methods 0.000 description 4
- 238000005266 casting Methods 0.000 description 3
- 239000007788 liquid Substances 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000000712 assembly Effects 0.000 description 1
- 238000000429 assembly Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- RYZCLUQMCYZBJQ-UHFFFAOYSA-H lead(2+);dicarbonate;dihydroxide Chemical compound [OH-].[OH-].[Pb+2].[Pb+2].[Pb+2].[O-]C([O-])=O.[O-]C([O-])=O RYZCLUQMCYZBJQ-UHFFFAOYSA-H 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003647 oxidation Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 238000002310 reflectometry Methods 0.000 description 1
- 229910052709 silver Inorganic materials 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
- 238000007711 solidification Methods 0.000 description 1
- 230000008023 solidification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
-
- 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
- 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
-
- 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
- G01N2021/8858—Flaw counting
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20076—Probabilistic image processing
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
-
- 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)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computing Systems (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to a metal workpiece surface defect detection method based on computer vision, which comprises the following steps: acquiring a surface image of a zinc alloy workpiece to obtain an image to be detected and an edge image of the image to be detected; obtaining the strong edge degree of each edge according to the gradient amplitude value of the edge image; obtaining suspected air hole edges of the edge image according to the strong edge degree; obtaining the suspected degree of each suspected air hole edge according to the strong edge degree of the suspected air hole; obtaining the correction suspected degree of each suspected air hole edge according to the suspected degree of each suspected air hole edge, and further obtaining all real edges of the edge image; and performing defect detection on the zinc alloy according to all the real edges. The invention aims to solve the problem that errors exist in air hole detection caused by the influence of reflection and surface textures on the surface of a zinc alloy workpiece, and achieve the purpose of improving the air hole detection accuracy.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a metal workpiece surface defect detection method based on computer vision.
Background
Metal workpieces are commonly used for manufacturing high-demand parts and assemblies, wherein zinc alloys are extremely wide in application field, and the surface quality directly influences the appearance attractiveness, corrosion resistance and mechanical properties of products. Thus, by surface defect detection, possible voids, cracks, oxidation and other defects can be discovered and handled early, preventing further propagation of these defects and resulting product failure.
At present, when defect detection is carried out on the surface of zinc alloy, an edge detection algorithm is often used for obtaining the edge of an air hole on the surface of the zinc alloy, but the edge is interfered by the surface texture of a zinc alloy workpiece, the surface of the zinc alloy is bright silver, reflection is possible, and when the surface is detected by using a Canny edge algorithm, the surface texture, the reflection and the edge of the air hole are all segmented, so that the problem of error of a detection result is caused.
Disclosure of Invention
The invention provides a metal workpiece surface defect detection method based on computer vision, which aims to solve the problem that the existing method is influenced by the surface texture of a zinc alloy workpiece, and when a canny edge algorithm is used for detection, both the surface texture and the air hole edge are judged to be defects, so that the detection result has errors.
The method for detecting the surface defects of the metal workpiece based on computer vision adopts the following technical scheme:
one embodiment of the invention provides a method for detecting surface defects of a metal workpiece based on computer vision, which comprises the following steps:
collecting a surface image of a zinc alloy workpiece to obtain an image to be detected;
obtaining an edge image of each image to be detected according to the image to be detected; obtaining the strong edge degree of each edge according to the gradient amplitude value of each edge pixel point of each edge in the edge image; obtaining all suspected air hole edges of the edge image of the image to be detected according to the strong edge degree of each edge in the image to be detected; obtaining the suspected degree of each suspected air hole edge in the edge image of the image to be detected according to the number of all edge pixel points of each suspected air hole edge in the edge image of the image to be detected and the strong edge degree of each suspected air hole edge; clustering centroids of all suspected air hole edges in an edge image of an image to be detected to obtain a plurality of clusters, and obtaining the correction suspected degree of each suspected air hole edge in the edge image of the image to be detected according to the aggregation relation between each suspected air hole edge in the edge image of the image to be detected and each suspected air hole edge in the cluster to which the suspected air hole edge belongs; the correction suspected degree of each suspected air hole edge in the edge image of the image to be detected is judged by presetting a real edge threshold value, and all the real edges of the air holes in the edge image of the image to be detected are obtained;
and performing defect detection on the zinc alloy workpiece by using all the real edges of the air holes in the edge image of the image to be detected.
Further, the obtaining the strong edge degree of each edge according to the gradient magnitude of each edge pixel point of each edge in the edge image includes:
will be the firstThe edge image of the individual images to be detected is denoted +.>First->Edge image of the individual images to be detected +.>Middle->No. H of the edge>The calculation mode of the strong edge probability of each edge pixel point is as follows:
wherein,is->Edge image of the individual images to be detected +.>Middle->No. H of the edge>Gradient magnitude of each edge pixel, +.>Is->Edge image of the individual images to be detected +.>Middle->The number of edge pixel points contained in each edge;
is->Edge image of the individual images to be detected +.>Middle->In the edges, except->The +.>Gradient magnitude of each edge pixel point;
representing a linear normalization function, ++>Represents an exponential function based on natural constants, < ->Taking absolute value symbols;
computing edge imagesIs>Strong edge probability mean value of all edge pixels in the edge, said mean value being denoted +.>Edge image of the individual images to be detected +.>Middle->Strong edge degree of individual edge->。
Further, the obtaining all suspected air hole edges of the edge image of the image to be detected includes:
presetting a strong edge thresholdWhen->Edge image of the individual images to be detected +.>Middle->Degree of strong edge of individual edgesSatisfy->At the time, and edge image->Middle->When the edge is a closed edge, the first ∈>Edge image of the individual images to be detected +.>Middle->The edge is marked as->Edge image of the individual images to be detected +.>Is a suspected air hole edge; get->Edge image of the individual images to be detected +.>The strong edge degree of the satisfying edge is equal to or higher than the strong edge threshold, and all edges of the closed edge are taken as the +.>Edge image of the individual images to be detected +.>Is included.
Further, the obtaining the suspected degree of each suspected air hole edge in the edge image of the image to be detected includes:
will be the firstEdge image of the individual images to be detected +.>Middle->The strong edge degree of the edge of the suspected air hole is marked +.>Will->Edge image of the individual images to be detected +.>Middle->The degree of rounding of the edges of the suspected pores is denoted +.>First->Edge image of the individual images to be detected +.>Middle->Degree of plausibility of the edge of each plausible poreThe calculation mode of (a) is as follows:
wherein,indicate->Edge image of the individual images to be detected +.>Middle->The degree of plausibility of the edges of the plausible pores.
Further, the obtaining manner of the near-circle degree includes:
will be the firstEdge image of the individual images to be detected +.>Middle->The number of edge pixel points of the suspected air hole edges is recorded asWill->The individual images to be detected and the edge image +.>Middle->The number of the pixel points in the connected domain surrounded by the edges of the suspected air holes is +.>First->Edge image of the individual images to be detected +.>Middle->The degree of rounding of the edges of the suspected pores +.>The calculation mode of (a) is as follows:
wherein,indicate->Edge image of the individual images to be detected +.>Middle->Theoretical radius of the edge of each suspected air hole, +.>Representing the circumference ratio>Represents an exponential function based on natural constants, < ->To take absolute value symbols.
Further, the method for obtaining the theoretical radius of the edge of the suspected air hole includes:
will be the firstEdge image of the individual images to be detected +.>Middle->The number of edge pixel points of the suspected air hole edges is recorded asThe number of the edge pixel points is +.>As->Edge image of the individual images to be detected +.>Middle->The perimeter of the edge of each suspected air hole is obtained according to the perimeter formula>Edge image of the individual images to be detected +.>Middle->Theoretical radius of the edge of the suspected air holes +.>。
Further, the obtaining the corrected suspected degree of each suspected air hole edge in the edge image of the image to be detected includes:
first, theEdge image of the individual images to be detected +.>Middle->The calculation method for correcting the suspected degree of each suspected air hole edge is as follows:
wherein,is->Edge image of the individual images to be detected +.>Middle->Correction of the suspected air hole edges, +.>Indicate->Edge image of the individual images to be detected +.>Middle->The degree of plausibility of the edges of the individual plausible pores;
is->Edge image of the individual images to be detected +.>Middle->In the cluster to which the suspected pore edges belong, the average value of the suspected degrees of all the suspected pore edges;
is->Edge image of the individual images to be detected +.>Middle->In the cluster to which each suspected pore edge belongs, the centroid of each suspected pore edge reaches +.>Euclidean distance average value of the center of mass of each suspected air hole edge;
an exponential function that is based on a natural constant;
obtaining correction suspected coefficients of all suspected air hole edges in an edge image of an image to be detected, and carrying out linear normalization on the correction suspected coefficients of each suspected air hole edge in the edge image of the image to be detected, wherein the obtained result is recorded as the correction suspected degree of each suspected air hole edge in the edge image of the image to be detected.
Further, the cluster acquisition mode includes:
acquisition of the firstEdge image of the individual images to be detected +.>Centroid position of each suspected pore edge +.>Edge image of the individual images to be detected +.>The centroid positions of all suspected pore edges in the wafer form a clustering space, the sample radius and the minimum sample number of a DBSCAN clustering algorithm are preset, and the DBSCAN clustering algorithm is used for obtaining the +.>Edge image of the individual images to be detected +.>A number of clusters in the matrix.
Further, the obtaining all the real edges of the air holes in the edge image of the image to be detected includes:
will be the firstEdge image of the individual images to be detected +.>Middle->The corrected suspected degree of the edge of each suspected air hole is marked +.>Presetting a real edge threshold +.>When->Edge image of the individual images to be detected +.>Middle->Correction of the suspected pore edges>Satisfy->When it is, will be->Edge image of the individual images to be detected +.>Middle->The edge of the suspected air holes is marked as->A real edge of the edge image of the image to be detected; and judging the correction suspected degree of each suspected air hole edge in the edge image of the image to be detected by using the real edge threshold value, and obtaining all real edges with the correction suspected degree of all the suspected air hole edges in the edge image of the image to be detected being greater than or equal to the real edge threshold value.
Further, the defect detection of the zinc alloy workpiece by using all the real edges of the air holes in the edge image of the image to be detected comprises the following steps:
counting the total number of real edges in all to-be-detected images of the zinc alloy workpiece, and presetting a defect evaluation valueWhen the total number of the real edges in all the images to be detected is greater than or equal to the defect determination value +.>When the zinc alloy workpiece is marked as a defective product; when the total number of the real edges in all the images to be detected is smaller than the defect determination value +.>And when the zinc alloy workpiece is used, marking the zinc alloy workpiece as a qualified product.
The technical scheme of the invention has the beneficial effects that: according to the method, the image to be detected and the edge image are obtained by collecting the surface image of the zinc alloy workpiece, the strong edge degree of each edge is obtained according to the gradient amplitude value of each edge pixel point of each edge in the edge image, and further the suspected air hole edge in the edge image is obtained, so that the aim of screening out the weak edge caused by the surface texture and the light spot of the zinc alloy workpiece is achieved, and the influence of the edge on the detection result of the air hole defect is avoided; according to the number of all edge pixel points in the suspected air hole edges and the strong edge degree of each suspected air hole edge, the suspected degree of each suspected air hole edge is obtained, and the purpose that the round edge is optimized to be used as the air hole edge is achieved; according to the distribution of the centroids of each suspected pore edge in the edge image, a plurality of clusters are obtained by using a clustering algorithm, the correction suspected degree of each suspected pore edge in the edge image of the image to be detected is obtained according to the aggregation relation of each suspected pore edge in the clusters, the influence of structural components on the surface of the zinc alloy workpiece on pore edge extraction is removed through the aggregation relation of each suspected pore edge, and the influence of the circular structural components on pore extraction accuracy is avoided; and extracting all real edges in the edge image through the correction suspected degree of each suspected air hole edge, further detecting the surface air hole defects of the zinc alloy workpiece, and finally reaching the requirement of screening out unqualified zinc alloy workpieces.
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 of the steps of the method for detecting surface defects of a metal workpiece based on computer vision.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of the metal workpiece surface defect detection method based on computer vision according to the invention, which is provided by the invention, with reference to the accompanying drawings and the preferred embodiment. 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 an embodiment 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 method for detecting the surface defects of the metal workpiece based on computer vision provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for detecting a surface defect of a metal workpiece based on computer vision according to an embodiment of the invention is shown, the method includes the following steps:
and S001, acquiring a surface image of the zinc alloy workpiece to obtain an image to be detected.
It should be noted that, the purpose of this embodiment is to detect the air hole defect existing on the surface of the zinc alloy workpiece produced by casting, and ensure that the quality of the produced zinc alloy workpiece meets the production requirement, so that the image of the zinc alloy workpiece needs to be collected first; according to the embodiment, the workpiece surface acquisition module is arranged at the outlet of the casting production line of the zinc alloy workpiece, the acquisition module comprises an industrial camera rotating shaft, an industrial camera and a soft light, the soft light rotates along with the industrial camera rotating shaft in the acquisition process, when the industrial camera and the surface of the zinc alloy workpiece form a vertical angle, an image acquired by the industrial camera at the moment is recorded to obtain a surface image of the zinc alloy workpiece, the surface image acquired by each acquisition is preprocessed by the mean value graying algorithm to obtain a to-be-detected image of the zinc alloy workpiece, the mean value graying algorithm is a known technology, and excessive redundant description is omitted in the embodiment; all surface images of a zinc alloy workpiece are preprocessed through a mean value graying algorithm, and all images to be detected of the zinc alloy workpiece are obtained.
So far, all images to be detected of the zinc alloy workpiece are acquired.
Step S002, obtaining an edge image of each image to be detected of the zinc alloy workpiece, and obtaining strong edge probability of each edge pixel point of each edge according to gradient amplitude values of each edge pixel point of each edge in the edge image, so as to obtain all suspected air hole edges in the edge image.
It should be noted that, the air holes present recessed dots distributed in clusters on the surface of the zinc alloy workpiece, while the surface of the zinc alloy workpiece produced by casting is flat and smooth, the edges of the recessed dots divide the smooth surface of the zinc alloy workpiece and the concave air holes to generate strong edges, so the embodiment uses the first zinc alloy workpieceThe number of images to be detected +.>For example, the image to be detectedInputting into Canny edge detection algorithm to obtain +.>Edge image of the individual images to be detected +.>And acquires an edge image +.>The position of each edge pixel point of each edge is preset with the neighborhood side length as +.>Is used in the present embodiment +.>To describe, in the edge image +.>Is +.>Obtaining an image to be detected by calculating Sobel operator in a neighborhood range>Gradient amplitude of each edge pixel point; it should be noted that, the Canny edge detection algorithm and the Sobel operator are known techniques, and the embodiment is not repeated.
It should be further noted that, after the air holes on the surface of the zinc alloy workpiece are processed by the Canny edge detection algorithm, a closed circular area is presented in the edge image, and because the reflectivity of the concave dots of the air holes is poorer than that of the silver-white zinc alloy, the gray value of the air holes in the image to be detected is lower, so that the edge gradient of each air hole is larger, therefore, the embodiment obtains the strong edge probability of each edge pixel point in the edge image according to the gradient amplitude of each edge pixel point in the edge image.
Specifically, the firstEdge image of the individual images to be detected +.>Middle->No. H of the edge>The calculation mode of the strong edge probability of each edge pixel point is as follows:
wherein the method comprises the steps ofIs->Edge image of the individual images to be detected +.>Middle->No. H of the edge>Strong edge probability of individual edge pixels, < +.>Is->Edge image of the individual images to be detected +.>Middle->No. H of the edge>The gradient magnitude of the individual edge pixels,is->Edge image of the individual images to be detected +.>Middle->The number of edge pixel points contained in each edge; />Is->Edge image of the individual images to be detected +.>Middle->In the edges, except->The +.>Gradient magnitude of each edge pixel point; />Represents a linear normalization function, the normalization range is +.>，/>Represents an exponential function based on natural constants, < ->To take absolute value symbols. />The larger the value of (2) is, the more (18) the edge image is indicated>Middle->No. H of the edge>The pixel points of each edge belong to the pixel points of the strong edge in the image to be detected, and the pixel points are more consistent with the characteristic that the difference of the surface gray scale of the zinc alloy workpiece is large due to bubbles, so that the probability of the strong edge is +.>The larger the value of (2); />For edge image +.>Middle->In the edges, except->Gradient amplitude mean value and +.>Gradient amplitude difference values of the edge pixel points are smaller to indicate edge image +.>Middle->No. H of the edge>Edge pixel and +.>The other edge pixel points in the edge all belong to the same edge intensity, and the +.>The more uniform the edges are, when +.>When the gradient amplitude of each edge pixel point is larger, the first part is>Gradient amplitude values of all edge pixel points in each edge are larger, and strong edge probability is +.>The larger the value of (2).
Further, calculating the strong edge probability average value of all edge pixel points of each edge in the edge image, and taking the strong edge probability average value as the strong edge degree of each edge in the edge image; then for the firstEdge image of the individual images to be detected +.>Middle->Strong edge degree of individual edge->Presetting strong edge threshold->In this embodiment->To describe, when->Edge image of the individual images to be detected +.>Middle->Strong edge degree of individual edge->Satisfy->At the time, the edge image +.>Middle->The strong edge probability of all edge pixel points of each edge is larger, and all the edge pixel points belong to the strong edge pixel points, and if the edge image is +.>Middle->When the edge is a closed edge, the first ∈>Edge image of the individual images to be detected +.>Middle->The edge is marked as->Edge image of the individual images to be detected +.>It should be noted that, the closed edge in this embodiment represents: when each edge pixel point in one edge has two adjacent edge pixel points, the edge is considered to be a closed edge; when->Edge image of the individual images to be detected +.>Middle->Strong edge degree of individual edge->Satisfy->Or->Edge image of the individual images to be detected +.>Middle->When the edge is not a closed edge, the edge image is described +.>Middle->The gradient amplitude difference of all edge pixel points of each edge is larger, and the gradient amplitude difference is possibly the weak edges generated by the surface textures, the surface light spots and the surface structures of the zinc alloy workpiece, so that the edges are not subjected to edge treatment with the strong edge degree smaller than the strong edge threshold value; similarly, calculate->Edge image of the individual images to be detected +.>Strong edge degree of all edges in (a) obtaining an edge image using a strong edge threshold value +.>Is included.
So far, all suspected air hole edges in the edge image are obtained.
And S003, obtaining the degree of the near circle of the suspected air hole edge according to the perimeter of the closed area of the suspected air hole edge and the number of the pixel points, and obtaining the suspected degree of the suspected air hole edge belonging to the air hole by combining the strong edge probability of each edge pixel point in the suspected air hole edge.
It should be noted that, in the process of pouring a zinc alloy workpiece, the zinc alloy presents a liquid state under the action of high temperature, bubbles are generated when the zinc alloy impacts the workpiece grinding tool, the bubbles present circular bubbles in the liquid zinc alloy metal liquid and float to the surface of the zinc alloy, and after solidification, regular circles appear, so that in the edge image of the image to be detected, compared with the surface structure and reflection of the zinc alloy workpiece, irregular textures are generated, if the edge of a suspected air hole is more circular, the edge of the suspected air hole is more likely to be the edge of an air hole. Therefore, in this embodiment, according to the perimeter of each suspected air hole edge and the number of pixels surrounded by the edges into the connected domain, the near-circle degree of each suspected air hole edge is obtained, and the suspected degree of the suspected air hole edge belonging to the air hole is obtained by combining the strong edge degree of the suspected air hole edge.
Specifically, count the thEdge image of the individual images to be detected +.>Middle->The number of edge pixel points of the suspected air hole edge>And edge image +.>Middle->The number of pixel points in the connected domain surrounded by the edges of the suspected air holes>Will->Edge image of the individual images to be detected +.>Middle->Edge image of each suspected air hole edgeNumber of pixels->As->The perimeter of the edge of each suspected air hole is obtained according to the perimeter formula>Edge image of the individual images to be detected +.>Middle->Theoretical radius of the edge of the suspected air holes +.>The method comprises the steps of carrying out a first treatment on the surface of the According to->Edge image of the individual images to be detected +.>Middle->Theoretical radius of the edge of the suspected air holes +.>The number of pixel points in the connected domain surrounded by the edges of the suspected air holes>Obtain->Edge image of the individual images to be detected +.>Middle->The degree of rounding of the edges of the suspected pores +.>The calculation mode of (a) is as follows:
wherein,indicate->Edge image of the individual images to be detected +.>Middle->The degree of rounding of the edges of each suspected air hole,indicate->Image edge of the individual image to be detected +.>Middle->The number of pixel points in the connected domain surrounded by the edges of the suspected air holes,indicate->Edge image of the individual images to be detected +.>Middle->Theoretical radius of the edge of each suspected air hole, +.>The circumference ratio is indicated as such,represents an exponential function based on natural constants, < ->To take absolute value symbols. />Representing edge image +.>Middle->The ratio of the number of the pixels in the connected domain surrounded by the edges of the suspected pores to the number of the pixels in the connected domain obtained by using the theoretical radius calculation, wherein the number of the pixels in the connected domain obtained by using the theoretical radius calculation is the number of the pixels under the standard circle under the theoretical radius, the ratio is more towards 1 to describe the edge image +.>Middle->The closer the edge of each suspected air hole is to the standard circle, the edge image +.>Middle->The more likely the edge of a suspected air hole is the edge of an air bubble, the edge image +.>Middle->The degree of rounding of the edges of the suspected pores +.>The larger the value is.
Further, according to the firstEdge image of the individual images to be detected +.>Middle->Near-circular degree of each suspected air hole edgeAnd edge image->Middle->The strong edge degree of the suspected pore edge, obtain +.>Edge image of the individual images to be detected +.>Middle->The suspected degree of the edge of the suspected stomata>The calculation mode of (a) is as follows:
wherein,indicate->Edge image of the individual images to be detected +.>Middle->The degree of plausibility of the edges of the individual plausible pores,indicate->Edge image of the individual images to be detected +.>Middle->The degree of rounding of the edges of the suspected pores, +.>Indicate->Edge image of the individual images to be detected +.>Middle->Strong edge extent of each suspected air hole edge. />The larger the value is, the more the edge image is indicated>Middle->The more likely the edges of the suspected air holes are strong edges, the more consistent the edges of the air bubbles are, the more suspected the edges of the suspected air holes areDegree of similarity (I)>The larger the value is; />The larger the value is, the more the edge image is indicated>Middle->The closer the edge of each suspected air hole is to the standard circle, the more the edge of each suspected air hole is in line with the recess generated by the air bubble, the suspected degree of the edge of each suspected air hole is +.>The larger the value is.
Similarly, the theoretical radius of each suspected air hole edge in the edge image of the image to be detected is obtained, the near-circle degree of each suspected air hole edge is further obtained, and the suspected degree of each suspected air hole edge in the edge image is obtained by combining the strong edge degree of each suspected air hole edge.
Thus, the suspected degree of each suspected air hole edge in the edge image is obtained.
And S004, obtaining the real edges of the air holes in the image to be detected according to the suspected degree and the distribution aggregation degree of each suspected air hole edge of each edge image in the image to be detected.
It should be noted that, in the image to be detected, not only circular air holes exist, but also structural components on the surface of the zinc alloy workpiece exist, and the structural components are circular in the image to be detected, so that the values of the suspected degrees of the structural components and the suspected degrees of the air holes are larger, but compared with the situation that the air holes are distributed in a clustered mode and the suspected probability of each air hole is higher, the distribution of the structural components is independent, therefore, according to the position aggregation relation of the edge of each suspected air hole compared with the edges of other suspected air holes, the suspected degree of each suspected air hole edge is corrected to obtain the corrected suspected degree of the edge of each suspected air hole, and further the real edge of the air hole in the image to be detected is obtained.
Specifically, obtain the firstEdge image of the individual images to be detected +.>Centroid position of each suspected pore edge +.>Edge image of the individual images to be detected +.>The centroid positions of all suspected pore edges form a clustering space, the sample radius and the minimum sample number of a DBSCAN clustering algorithm are preset, the DBSCAN clustering algorithm is a known technology, and excessive description is not made in the embodiment; in this embodiment, a sample radius of 20 and a minimum sample number of 5 are used to obtain the +.>Edge image of the individual images to be detected +.>A plurality of clusters and a plurality of discrete centroids, and setting the suspected degree of the suspected pore edge represented by the discrete centroids to 0, the method is in +.>Edge image of the individual images to be detected +.>Middle->Obtaining the +.>Edge image of the individual images to be detected +.>Middle->Correction of the suspected pore edges>The calculation mode of (a) is as follows:
wherein,is->Edge image of the individual images to be detected +.>Middle->Correction of the suspected air hole edges, +.>Indicate->Edge image of the individual images to be detected +.>Middle->The degree of plausibility of the edges of the plausible pores, +.>Is->Edge image of the individual images to be detected +.>Middle->The average value of the suspected degrees of all the suspected pore edges in the cluster to which the suspected pore edges belong; />Is->Edge image of the individual images to be detected +.>Middle->In the cluster to which each suspected pore edge belongs, the centroid of each suspected pore edge reaches +.>Euclidean distance mean of the edge centroids of the suspected air holes, < ->An exponential function based on a natural constant is represented. In->In the cluster to which the edge of each suspected air hole belongs, the average value of the suspected degree and the +.>Difference in the degree of plausibility of the edges of the plausible air holes +.>The smaller the value, the description of +.>Edge image of the individual images to be detected +.>Middle->The clusters to which the edges of the suspected pores belong are all relatively regular circles and are equal to +.>The shape of the edge of the suspected air hole is similar, the +.>The more likely the edges of the suspected air holes are edges where air bubbles are generated; />The smaller the value is, the>Edge images of the individual images to be detectedMiddle->The higher the aggregation level of clusters to which the edges of the suspected air holes belong, the +.>The more likely the edge of a suspected air hole is the edge where an air bubble is generated. And similarly, obtaining correction suspected coefficients of all suspected air hole edges in the edge image of the image to be detected, and carrying out linear normalization on the correction suspected coefficients of each suspected air hole edge in the edge image of the image to be detected, wherein the obtained result is recorded as the correction suspected degree of each suspected air hole edge in the edge image of the image to be detected.
Further, the first step isEdge image of the individual images to be detected +.>Middle->The corrected suspected degree of the edge of each suspected air hole is marked +.>Presetting a real edge threshold +.>The present embodiment employs +.>To describe, when->Edge image of the individual images to be detected +.>Middle->Correction of the suspected pore edges>Satisfy->Description of the->Edge image of the individual images to be detected +.>Middle->The suspected pore edges are clustered, the more likely they are to be real pore edges, the +.>Edge image of the individual images to be detected +.>Middle->The edge of the suspected air holes is marked as->A real edge of the image to be detected; when->Edge image of the individual images to be detected +.>Middle->Correction of the suspected pore edges>Satisfy->Description of the->Edge image of the individual images to be detected +.>Middle->The more likely the distribution difference between the suspected pore edges and other suspected pore edges in the cluster is large, the more likely the suspected pore edges are structural components of the zinc alloy workpiece, the no treatment is performed. And similarly, judging the correction suspected degree of each suspected pore edge in the edge image of the image to be detected by using the real edge threshold value, and obtaining all real edges with the correction suspected degree of all the suspected pore edges in the edge image of the image to be detected being greater than or equal to the real edge threshold value.
So far, all real edges in the edge image of the image to be detected are obtained.
And S005, detecting surface air hole defects of the zinc alloy workpiece according to the real edges of the air holes in the image to be detected.
After all real edges in the edge images of the images to be detected are obtained, counting the total number of the real edges in all the images to be detected of the zinc alloy workpiece, and presetting a defect evaluation valueThe defect determination value is obtained according to the empirical value, and the embodiment adoptsTo be described, when the total number of the real edges in the edge image of the image to be detected is equal to or greater than the total number of the defect determination values +.>When the zinc alloy workpiece is marked as a defective product; when the total number of the real edges in the edge image of the image to be detected is smaller than the defect determination value number +.>And when the zinc alloy workpiece is used, marking the zinc alloy workpiece as a qualified product.
The following examples were usedThe model only represents the negative correlation and the result of the constraint model output is +.>Within the interval>For the input of the model, other models with the same purpose can be replaced in the implementation, and the embodiment is only to +.>The model is described as an example, and is not particularly limited.
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. The method for detecting the surface defects of the metal workpiece based on computer vision is characterized by comprising the following steps of:
collecting a surface image of a zinc alloy workpiece to obtain an image to be detected;
obtaining an edge image of each image to be detected according to the image to be detected; obtaining the strong edge degree of each edge according to the gradient amplitude value of each edge pixel point of each edge in the edge image; obtaining all suspected air hole edges of the edge image of the image to be detected according to the strong edge degree of each edge in the image to be detected; obtaining the suspected degree of each suspected air hole edge in the edge image of the image to be detected according to the number of all edge pixel points of each suspected air hole edge in the edge image of the image to be detected and the strong edge degree of each suspected air hole edge; clustering centroids of all suspected air hole edges in an edge image of an image to be detected to obtain a plurality of clusters, and obtaining the correction suspected degree of each suspected air hole edge in the edge image of the image to be detected according to the aggregation relation between each suspected air hole edge in the edge image of the image to be detected and each suspected air hole edge in the cluster to which the suspected air hole edge belongs; the correction suspected degree of each suspected air hole edge in the edge image of the image to be detected is judged by presetting a real edge threshold value, and all the real edges of the air holes in the edge image of the image to be detected are obtained;
and performing defect detection on the zinc alloy workpiece by using all the real edges of the air holes in the edge image of the image to be detected.
2. The method for detecting surface defects of a metal workpiece based on computer vision according to claim 1, wherein the obtaining the strong edge degree of each edge according to the gradient magnitude of each edge pixel point of each edge in the edge image comprises:
will be the firstThe edge image of the individual images to be detected is denoted +.>First->Edge image of the individual images to be detected +.>Middle->No. H of the edge>The calculation mode of the strong edge probability of each edge pixel point is as follows:
wherein,is->Edge image of the individual images to be detected +.>Middle->No. H of the edge>Gradient magnitude of each edge pixel, +.>Is->Edge image of the individual images to be detected +.>Middle->The number of edge pixel points contained in each edge;
is->Edge image of the individual images to be detected +.>Middle->In the edges, except->The +.>Gradient magnitude of each edge pixel point;
representing a linear normalization function, ++>Represents an exponential function based on natural constants, < ->Taking absolute value symbols;
computing edge imagesIs>Strong edge probability mean value of all edge pixels in the edge, said mean value being denoted +.>Edge image of the individual images to be detected +.>Middle->Strong edge degree of individual edge->。
3. The method for detecting surface defects of a metal workpiece based on computer vision according to claim 1, wherein the obtaining all suspected pore edges of the edge image of the image to be detected comprises:
presetting a strong edge thresholdWhen->Edge image of the individual images to be detected +.>Middle->Strong edge degree of individual edge->Satisfy->At the time, and edge image->Middle->When the edge is a closed edge, the first ∈>Edge image of the individual images to be detected +.>Middle->The edge is marked as->Edge image of the individual images to be detected +.>Is a suspected air hole edge; get->Edge image of the individual images to be detected +.>The strong edge degree of the satisfying edge is equal to or higher than the strong edge threshold, and all edges of the closed edge are taken as the +.>Edge image of the individual images to be detected +.>Is included.
4. The method for detecting surface defects of a metal workpiece based on computer vision according to claim 1, wherein the obtaining the suspected degree of each suspected pore edge in the edge image of the image to be detected comprises:
will be the firstEdge image of the individual images to be detected +.>Middle->The strong edge degree of the edge of the suspected air hole is marked +.>Will->Edge image of the individual images to be detected +.>Middle->The degree of rounding of the edges of the suspected pores is denoted +.>First->Edge image of the individual images to be detected +.>Middle->The calculation mode of the suspected degree of each suspected air hole edge is as follows:
wherein,indicate->Edge image of the individual images to be detected +.>Middle->The degree of plausibility of the edges of the plausible pores.
5. The method for detecting surface defects of a metal workpiece based on computer vision according to claim 4, wherein the obtaining method of the degree of near-circle comprises the following steps:
will be the firstEdge image of the individual images to be detected +.>Middle->The number of edge pixels of the edge of each suspected air hole is marked as +.>Will->The individual images to be detected and the edge image +.>Middle->The number of the pixel points in the connected domain surrounded by the edges of the suspected air holes is +.>First->Edge image of the individual images to be detected +.>Middle->The degree of rounding of the edges of the suspected pores +.>The calculation mode of (a) is as follows:
wherein,indicate->Edge image of the individual images to be detected +.>Middle->Theoretical radius of the edge of each suspected air hole, +.>Representing the circumference ratio>The representation being based on natural constantsExponential function of>To take absolute value symbols.
6. The method for detecting surface defects of a metal workpiece based on computer vision according to claim 5, wherein the method for obtaining the theoretical radius of the suspected pore edge comprises the following steps:
will be the firstEdge image of the individual images to be detected +.>Middle->The number of edge pixels of the edge of each suspected air hole is marked as +.>The number of the edge pixel points is +.>As->Edge image of the individual images to be detected +.>Middle->The perimeter of the edge of each suspected air hole is obtained according to the perimeter formula>Edge image of the individual images to be detected +.>Middle->Theoretical radius of the edge of the suspected air holes +.>。
7. The method for detecting surface defects of a metal workpiece based on computer vision according to claim 1, wherein the obtaining the corrected suspected degree of each suspected pore edge in the edge image of the image to be detected comprises:
first, theEdge image of the individual images to be detected +.>Middle->The calculation method for correcting the suspected degree of each suspected air hole edge is as follows:
wherein,is->Edge image of the individual images to be detected +.>Middle->Correction of the suspected air hole edges, +.>Indicate->Edge image of the individual images to be detected +.>Middle->The degree of plausibility of the edges of the individual plausible pores;
is->Edge image of the individual images to be detected +.>Middle->In the cluster to which the suspected pore edges belong, the average value of the suspected degrees of all the suspected pore edges;
is->Edge image of the individual images to be detected +.>Middle->In the cluster to which each suspected pore edge belongs, each suspected pore edgeCentroid to->Euclidean distance average value of the center of mass of each suspected air hole edge;
an exponential function that is based on a natural constant;
obtaining correction suspected coefficients of all suspected air hole edges in an edge image of an image to be detected, and carrying out linear normalization on the correction suspected coefficients of each suspected air hole edge in the edge image of the image to be detected, wherein the obtained result is recorded as the correction suspected degree of each suspected air hole edge in the edge image of the image to be detected.
8. The method for detecting surface defects of a metal workpiece based on computer vision according to claim 7, wherein the cluster acquisition mode comprises:
acquisition of the firstEdge image of the individual images to be detected +.>Centroid position of each suspected pore edge +.>Edge image of the individual images to be detected +.>The centroid positions of all suspected pore edges in the wafer form a clustering space, the sample radius and the minimum sample number of a DBSCAN clustering algorithm are preset, and the DBSCAN clustering algorithm is used for obtaining the +.>Edge image of the individual images to be detected +.>A number of clusters in the matrix.
9. The method for detecting surface defects of a metal workpiece based on computer vision according to claim 1, wherein the obtaining all real edges of the air holes in the edge image of the image to be detected comprises:
will be the firstEdge image of the individual images to be detected +.>Middle->The corrected suspected degree of the edge of each suspected air hole is marked +.>Presetting a real edge threshold +.>When->Edge image of the individual images to be detected +.>Middle->Correction of suspected pore edgesSatisfy->When it is, will be->Edge image of the individual images to be detected +.>Middle->The edge of the suspected air holes is marked as->A real edge of the edge image of the image to be detected; and judging the correction suspected degree of each suspected air hole edge in the edge image of the image to be detected by using the real edge threshold value, and obtaining all real edges with the correction suspected degree of all the suspected air hole edges in the edge image of the image to be detected being greater than or equal to the real edge threshold value.
10. The method for detecting surface defects of a metal workpiece based on computer vision according to claim 1, wherein the defect detection of the zinc alloy workpiece using all real edges of the air holes in the edge image of the image to be detected comprises:
counting the total number of real edges in all to-be-detected images of the zinc alloy workpiece, and presetting a defect evaluation valueWhen the total number of the real edges in all the images to be detected is greater than or equal to the defect determination value +.>When the zinc alloy workpiece is marked as a defective product; when the total number of the real edges in all the images to be detected is smaller than the defect determination value +.>And when the zinc alloy workpiece is used, marking the zinc alloy workpiece as a qualified product.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311620035.0A CN117314925B (en) | 2023-11-30 | 2023-11-30 | Metal workpiece surface defect detection method based on computer vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311620035.0A CN117314925B (en) | 2023-11-30 | 2023-11-30 | Metal workpiece surface defect detection method based on computer vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117314925A true CN117314925A (en) | 2023-12-29 |
| CN117314925B CN117314925B (en) | 2024-02-20 |
Family
ID=89288828
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311620035.0A Active CN117314925B (en) | 2023-11-30 | 2023-11-30 | Metal workpiece surface defect detection method based on computer vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117314925B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117830312A (en) * | 2024-03-05 | 2024-04-05 | 陕西长空齿轮有限责任公司 | Alloy crack nondestructive testing method based on machine vision |
| CN117934456A (en) * | 2024-03-20 | 2024-04-26 | 大连建峰印业有限公司 | Packaging box printing quality detection method based on image processing |
| CN117934456B (en) * | 2024-03-20 | 2024-05-28 | 大连建峰印业有限公司 | Packaging box printing quality detection method based on image processing |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114820631A (en) * | 2022-07-04 | 2022-07-29 | 南通中豪超纤制品有限公司 | Fabric defect detection method capable of resisting texture interference |
| CN115082419A (en) * | 2022-07-14 | 2022-09-20 | 江苏诺阳家居科技有限公司 | Blow-molded luggage production defect detection method |
| CN115239735A (en) * | 2022-09-23 | 2022-10-25 | 探长信息技术(苏州)有限公司 | Communication cabinet surface defect detection method based on computer vision |
| CN116758061A (en) * | 2023-08-11 | 2023-09-15 | 山东优奭趸泵业科技有限公司 | Casting surface defect detection method based on computer vision |
| CN116977358A (en) * | 2023-09-22 | 2023-10-31 | 玖龙智能包装(天津)有限公司 | Visual auxiliary detection method for corrugated paper production quality |
-
2023
- 2023-11-30 CN CN202311620035.0A patent/CN117314925B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114820631A (en) * | 2022-07-04 | 2022-07-29 | 南通中豪超纤制品有限公司 | Fabric defect detection method capable of resisting texture interference |
| CN115082419A (en) * | 2022-07-14 | 2022-09-20 | 江苏诺阳家居科技有限公司 | Blow-molded luggage production defect detection method |
| CN115239735A (en) * | 2022-09-23 | 2022-10-25 | 探长信息技术(苏州)有限公司 | Communication cabinet surface defect detection method based on computer vision |
| CN116758061A (en) * | 2023-08-11 | 2023-09-15 | 山东优奭趸泵业科技有限公司 | Casting surface defect detection method based on computer vision |
| CN116977358A (en) * | 2023-09-22 | 2023-10-31 | 玖龙智能包装(天津)有限公司 | Visual auxiliary detection method for corrugated paper production quality |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117830312A (en) * | 2024-03-05 | 2024-04-05 | 陕西长空齿轮有限责任公司 | Alloy crack nondestructive testing method based on machine vision |
| CN117830312B (en) * | 2024-03-05 | 2024-05-28 | 陕西长空齿轮有限责任公司 | Alloy crack nondestructive testing method based on machine vision |
| CN117934456A (en) * | 2024-03-20 | 2024-04-26 | 大连建峰印业有限公司 | Packaging box printing quality detection method based on image processing |
| CN117934456B (en) * | 2024-03-20 | 2024-05-28 | 大连建峰印业有限公司 | Packaging box printing quality detection method based on image processing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117314925B (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115082467B (en) | Building material welding surface defect detection method based on computer vision | |
| CN106407502B (en) | Blade section profile parameter evaluation method based on optimal matching | |
| CN105067638B (en) | Tire fetal membrane face character defect inspection method based on machine vision | |
| CN108764229B (en) | Water gauge image automatic identification method based on computer vision technology | |
| CN115601362B (en) | Welding quality evaluation method based on image processing | |
| CN114092403A (en) | Grinding wheel wear detection method and system based on machine vision | |
| CN115797361B (en) | Aluminum template surface defect detection method | |
| CN111833306A (en) | Defect detection method and model training method for defect detection | |
| CN115131359B (en) | Method for detecting pitting defects on surface of metal workpiece | |
| CN116735612A (en) | Welding defect detection method for precise electronic components | |
| CN115601347A (en) | Steel plate surface defect detection method based on gray texture analysis | |
| CN106651831B (en) | Bamboo block defect detection method and system | |
| CN116758077A (en) | Online detection method and system for surface flatness of surfboard | |
| CN115311262A (en) | Printed circuit board defect identification method | |
| CN114119603A (en) | Image processing-based snack box short shot defect detection method | |
| CN116630304A (en) | Lithium battery mold processing detection method and system based on artificial intelligence | |
| CN113781413B (en) | Electrolytic capacitor positioning method based on Hough gradient method | |
| CN113298775B (en) | Self-priming pump double-sided metal impeller appearance defect detection method, system and medium | |
| CN117314925B (en) | Metal workpiece surface defect detection method based on computer vision | |
| CN117152129B (en) | Visual detection method and system for surface defects of battery cover plate | |
| CN112396580B (en) | Method for detecting defects of round part | |
| CN116993726B (en) | Mineral casting detection method and system | |
| CN111815575B (en) | Bearing steel ball part detection method based on machine vision | |
| CN110428916B (en) | Thickness detection method and device for coated particles and computing equipment | |
| CN116385440A (en) | Visual detection method for arc-shaped blade |
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 |