CN114813744A - Multi-station image comprehensive re-judgment defect detection method - Google Patents
Multi-station image comprehensive re-judgment defect detection method Download PDFInfo
- Publication number
- CN114813744A CN114813744A CN202210324550.3A CN202210324550A CN114813744A CN 114813744 A CN114813744 A CN 114813744A CN 202210324550 A CN202210324550 A CN 202210324550A CN 114813744 A CN114813744 A CN 114813744A
- Authority
- CN
- China
- Prior art keywords
- light source
- image
- defect
- judgment
- images
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to a multistation image comprehensive re-judgment defect detection method, which comprises the following steps of S1, constructing a hardware structure required by imaging at a product imaging detection position, wherein the hardware structure comprises a plurality of cameras, lenses and light sources; s2, carrying out image acquisition on the same product for multiple times by adopting the imaging hardware set up in the step S1 to obtain multiple images; and S3, comparing the defect information of the same product on a plurality of images, and judging the detected defect type. The invention overcomes the defects of low detection rate and high false judgment rate in the prior art, adopts the comprehensive re-judgment technology of multi-station images, and distinguishes the types of the images by utilizing the characteristic expression of the same defect in a plurality of images, thereby being more accurate.
Description
Technical Field
The invention relates to the technical field of image visual inspection, in particular to a method for detecting comprehensive re-judgment defects of multi-station images.
Background
The defect detection generally refers to the detection of the surface defects of articles, the surface defect detection is to detect the defects such as spots, pits, scratches, chromatic aberration, defects and the like on the surface of a workpiece by adopting an advanced machine vision detection technology, and the defect detection technology is always a key difficult problem in the aspect of detection, but no complete solution is provided in the market at present, and the defect detection technology has high requirements, and the missing judgment rate and the misjudgment rate required by manufacturers are both extremely high, so the defect detection is always a headache problem.
Moreover, most of the existing detection technologies are single-station detection modes, such as bright field images and dark field images, and the algorithm distinguishes the types of the defects according to the defect phenotype of a single image.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method for detecting the defects of the multi-station image comprehensive re-judgment is provided, whether a defect needs to be detected or not is comprehensively judged from the defect expression of each image through the images of a plurality of stations, and the problems of detection omission and false detection in single-station image detection are solved.
The technical scheme adopted by the invention for solving the technical problems is as follows: a method for detecting defects of comprehensive multi-station images comprises the following steps,
s1, building a hardware structure required by imaging at the product imaging detection position, wherein the hardware structure comprises a plurality of cameras, lenses and light sources;
s2, carrying out image acquisition on the same product for multiple times by adopting the imaging hardware set up in the step S1 to obtain multiple images;
and S3, comparing the defect information of the same product on a plurality of images, and judging the detected defect type.
Further, the hardware structure in step S1 of the present invention includes a first camera, a first lens, a first light source, a second light source, a third light source, a second camera, and a second lens; the second light source and the third light source are respectively arranged above and below the product; the first light source is arranged above the second light source; the first camera and the first lens are arranged above the first light source; the second camera and the second lens are arranged below the third light source.
Still further, the first light source of the present invention is a backlight source.
Still further, the second light source and the third light source are both parallel coaxial light sources.
Further, in step S2 of the present invention, the image capturing the same product for a plurality of times includes the following steps,
s21, shooting a first image under a first light source;
s22, shooting a second image under a second light source;
and S23, shooting a third image under a third light source.
In step S3, the detected defect type is determined by comparing the feature representations of the same defect in the second and third images with the feature representation in the first image in gray scale and contrast.
The method has the advantages that the defects in the background technology are overcome, the comprehensive re-judgment method of the multi-station images is adopted, the types of the defects can be distinguished more accurately by utilizing the characteristic expression of the same defect in a plurality of images, the detection rate is improved, and the misjudgment rate is reduced; meanwhile, the products can be efficiently classified by using the method, the use requirements of customers are met, and the cost of manual visual inspection is greatly reduced.
Drawings
FIG. 1 is a schematic view of an imaging configuration of the present invention;
FIGS. 2A-2C are schematic diagrams of defect comprehensive re-judgment images according to the present invention;
in the figure: 1. producing a product; 2. a second light source; 3. a third light source; 4. a first light source; 5. a first lens; 6. a second lens; 7. a first camera; 8. a second camera.
Detailed Description
The invention will now be described in further detail with reference to the drawings and preferred embodiments. These drawings are simplified schematic views illustrating only the basic structure of the present invention in a schematic manner, and thus show only the constitution related to the present invention.
The method for detecting the multi-station image comprehensive re-judgment defects as shown in fig. 1-2 comprises the following steps,
s1, building a hardware structure required by imaging at the product imaging detection position, wherein the hardware structure comprises a plurality of cameras, lenses and light sources;
s2, carrying out image acquisition on the same product for multiple times by adopting the imaging hardware set up in the step S1 to obtain multiple images;
and S3, comparing the defect information of the same product on a plurality of images, and judging the detected defect type.
The hardware structure in step S1 includes a first camera 7, a first lens 5, a first light source 4, a second light source 2, a third light source 3, a second camera 8, and a second lens 6; the second light source 2 and the third light source 3 are respectively arranged above and below the product 1; the first light source 4 is arranged above the second light source 2; the first camera 7 and the first lens 5 are arranged above the first light source 4; the second camera 8 and the second lens 6 are disposed below the third light source 3. The first light source 4 is used as a common light source and is a backlight source; the second light source 2 and the third light source 3 are parallel coaxial light sources as special light sources. The common light source can be arranged on one side of any special light source and the lens; backlights as well as parallel coaxial light sources are directly available from the market.
In step S2, the image capturing of the same product for a plurality of times includes the steps of,
s21, shooting a first image under a first light source; the shooting picture of the common light source is a picture A in figure 2;
s22, shooting a second image under a second light source; one of the special light sources above the product is taken as diagram B in figure 2,
s23, shooting a third image under a third light source; another special light source located below the product is photographed as C in fig. 2.
In step S3, the feature expression of the same defect in the second image and the third image is compared with the feature expression in the first image in terms of gray scale and contrast, so as to determine the type of the detected defect.
The detection principle is specifically explained as follows:
as shown in fig. 2, A, B, C are images of the same product at three stations (at the same physical location), where the three stations represent three shooting processes, not three different shooting locations; circles and squares in the figure are representative of defects at three stations.
The traditional algorithm detection method is to use a single-station image for analysis, when only an A image is used for algorithm analysis, the gray levels (gray level: the gray level of a region itself, in the figure, it can be seen that the circular and the square are both black, so that the gray levels thereof are the same) and the contrast (contrast: the absolute value of the difference between the gray levels of the region and the gray levels of the surrounding environment, as shown in FIG. 2, the gray levels of the circular and the square are the same, and the gray levels of the surrounding environment are the same, that is, the white, that is, the contrast thereof is the same) are the same, and when the two attributes are used for judging the defects at the algorithm level, the two defects belong to the same type of defect.
When a plurality of station images are adopted for comprehensive judgment, by utilizing the characteristic phenotype of the same defect in the B, C image, the gray scale and the contrast of the defect of the circular defect in the image A are different from those of the defect of the common light source (the image A), the phenotype of the square defect in the A, B, C three images is solid, the gray scale and the contrast of the defect of the square defect in the image B and the image C are the same as those of the defect of the common light source (the image A), and the circular defect and the square defect are obviously different from the defect of the same type, the circular defect is a defect to be detected, and the square defect is an interference defect. Therefore, if only a single image A is used for analysis, wrong judgment can be caused, and the accuracy rate of comprehensive judgment by using the characteristic phenotype of the defect in the three images is far higher than that of a single image.
While particular embodiments of the present invention have been described in the foregoing specification, various modifications and alterations to the previously described embodiments will become apparent to those skilled in the art from this description without departing from the spirit and scope of the invention.
Claims (6)
1. A multi-station image comprehensive re-judgment defect detection method is characterized by comprising the following steps: comprises the following steps of (a) carrying out,
s1, building a hardware structure required by imaging at the product imaging detection position, wherein the hardware structure comprises a plurality of cameras, lenses and light sources;
s2, carrying out image acquisition on the same product for multiple times by adopting the imaging hardware set up in the step S1 to obtain multiple images;
and S3, comparing the defect information of the same product on a plurality of images, and judging the detected defect type.
2. The multi-station image comprehensive re-judgment defect detection method of claim 1, wherein: the hardware structure in step S1 includes a first camera, a first lens, a first light source, a second light source, a third light source, a second camera, and a second lens; the second light source and the third light source are respectively arranged above and below the product; the first light source is arranged above the second light source; the first camera and the first lens are arranged above the first light source; the second camera and the second lens are arranged below the third light source.
3. The multi-station image comprehensive re-judgment defect detection method according to claim 2, characterized in that: the first light source is a backlight source.
4. The multi-station image comprehensive re-judgment defect detection method according to claim 2, characterized in that: the second light source and the third light source are both parallel coaxial light sources.
5. The multi-station image comprehensive re-judgment defect detection method according to claim 2, characterized in that: in step S2, the image capturing for multiple times for the same product includes the following steps,
s21, shooting a first image under a first light source;
s22, shooting a second image under a second light source;
and S23, shooting a third image under a third light source.
6. The multi-station image comprehensive re-judgment defect detection method of claim 5, wherein: in step S3, the feature expression of the same defect in the second image and the third image is compared with the feature expression in the first image in terms of gray scale and contrast, so as to determine the type of the detected defect.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210324550.3A CN114813744A (en) | 2022-03-30 | 2022-03-30 | Multi-station image comprehensive re-judgment defect detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210324550.3A CN114813744A (en) | 2022-03-30 | 2022-03-30 | Multi-station image comprehensive re-judgment defect detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114813744A true CN114813744A (en) | 2022-07-29 |
Family
ID=82532976
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210324550.3A Pending CN114813744A (en) | 2022-03-30 | 2022-03-30 | Multi-station image comprehensive re-judgment defect detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114813744A (en) |
-
2022
- 2022-03-30 CN CN202210324550.3A patent/CN114813744A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI603074B (en) | Optical film defect detection method and system thereof | |
| CN111970506B (en) | Lens dirt detection method, device and equipment | |
| CN101329281B (en) | System and method for testing image sensing wafer stain and | |
| TW555969B (en) | Method for non-destructive inspection, apparatus thereof and digital camera system | |
| CN109685760B (en) | MATLAB-based SLM powder bed powder laying image convex hull depression defect detection method | |
| WO2007074770A1 (en) | Defect inspection device for inspecting defect by image analysis | |
| KR20160090359A (en) | Surface defect detection method and surface defect detection device | |
| CN109488888B (en) | Metal pipeline leakage monitoring method based on infrared temperature field multivariate analysis | |
| CN107515481B (en) | Display panel detection method and device | |
| CN111047655A (en) | High-definition camera cloth defect detection method based on convolutional neural network | |
| CN110021012B (en) | Mobile phone lens window glass defect detection method based on machine vision technology | |
| CN109932369B (en) | Method and device for detecting special-shaped display panel | |
| CN105388165A (en) | Defect monitoring method of polarizing film | |
| KR101867015B1 (en) | Device and method for inspecting defect of glass and inspection system | |
| CN108072664B (en) | Appearance detection system and method | |
| CN111551559A (en) | LCD (liquid Crystal display) liquid crystal screen defect detection method based on multi-view vision system | |
| CN111426693A (en) | Quality defect detection system and detection method thereof | |
| CN106846294B (en) | Visual detection method, device and equipment | |
| JPH08189904A (en) | Surface defect detector | |
| KR20140122938A (en) | Method and apparatus of panel inspection | |
| CN113866182A (en) | Detection method and system for detecting defects of display module | |
| CN114813744A (en) | Multi-station image comprehensive re-judgment defect detection method | |
| CN110766649B (en) | Defect judging device and method for automatic repair system | |
| CN116091506A (en) | Machine vision defect quality inspection method based on YOLOV5 | |
| CN112326681B (en) | Method for correcting and detecting lens cleanliness by utilizing defocusing difference flat field |
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 |