Blue film surface defect classification method

By using the principle of time-sharing exposure and taking pictures with light sources at different angles to acquire images, the problem of high cost and inability to classify defects on the surface of blue film in existing technologies has been solved, and efficient and accurate defect detection and classification have been achieved.

CN118967582APending Publication Date: 2024-11-15BEIJING FOCUSIGHT TECH
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202410966047.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-18
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

In existing technologies, 3D line laser inspection of blue film surface defects is costly and cannot accurately distinguish between pits, bubbles and foreign objects.

Method used

Using the principle of time-sharing exposure, two images are captured in one shot using two light sources at different angles (white linear light and infrared light). The defects are then classified by calculating the vertical contrast and the internal and external contrast.

Benefits of technology

It enables accurate detection and classification of defects on the blue film surface, reducing detection costs and improving detection accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118967582A_ABST
    Figure CN118967582A_ABST
Patent Text Reader

Abstract

The invention relates to a blue film surface defect classification method. The method comprises the following steps: constructing an imaging system hardware structure; two images of a product are simultaneously obtained through one-time photographing according to the time-sharing exposure principle, namely, a first image is obtained under a first light source, and a second image is obtained under a second light source; performing feature extraction on the defect on the first image, and calculating the upper and lower contrast of the defect; performing feature extraction on the defect on the second image, and calculating internal and external contrast of the defect; and comparing the upper and lower contrast ratios and the internal and external contrast ratios with set threshold values, thereby classifying the defects. According to the method, time-sharing exposure is utilized, a low-angle white light source and a high-angle infrared light source in two different directions are used for photographing at a time to obtain two images of a product, one image is used for representing the concave-convex feeling of the surface of the product, the other image is used for representing the coating color information of the product, and the two images are comprehensively utilized for defect detection; the detection precision is improved, and the defects can be accurately detected and classified.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of visual inspection, in particular to a method for classifying blue film surface defects. Background Art

[0002] In the existing production process, 3D line laser is generally used to detect surface defects of blue film products, and the concave and convex defects on the product surface are detected through the height information of the blue film surface.

[0003] However, this approach has the following problems:

[0004] 1. 3D line laser is expensive;

[0005] 2. Unable to accurately classify defects: Pits appear as concave features under 3D imaging. Although foreign matter and bubbles can be detected under 3D imaging, they are all convex features, and it is impossible to distinguish whether the defect is a foreign matter or a bubble. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a method for classifying blue film surface defects, which can detect pits, bubbles and foreign matter defects on the blue film surface and can accurately distinguish these defects.

[0007] The technical solution adopted by the present invention to solve the technical problem is: a method for classifying surface defects of a blue film, comprising the following steps:

[0008] S1. Build the hardware structure of the imaging system;

[0009] S2, using the time-sharing exposure principle to take a photo at a time to simultaneously obtain two images of the product, that is, obtain a first image under a first light source and obtain a second image under a second light source;

[0010] S3, extracting features of defects on the first image, and calculating upper and lower contrasts of the defects;

[0011] S4, extracting features of defects on the second image, and calculating the contrast between inside and outside of the defects;

[0012] S5. Compare the upper and lower contrast and the inner and outer contrast with the set thresholds respectively, so as to classify the defects.

[0013] Furthermore, in step S1 described in the present invention, the hardware structure of the imaging system includes a first light source, a second light source and an imaging component; the second light source is vertically above the product, and the angle between the second light source and the product is 0°; the first light source is arranged on one side of the second light source, and the angle between the first light source and the second light source is 25°; the imaging component is arranged on the other side of the second light source, and the angle between the imaging component and the second light source is also 25°; the first light source is white line light; and the second light source is infrared light.

[0014] Furthermore, in step S2 of the present invention, the defects respectively included in the first image and the second image are bubbles, foreign matter and / or pits; bubbles show a bright upper and dark lower feature under the first light source, and a bright feature under the second light source; foreign matter shows a bright upper and dark lower feature under the first light source, and a bright outer circle and dark inner feature under the second light source; pits show a bright upper and dark lower feature under the first light source, and have no obvious features under the second light source.

[0015] Furthermore, in step S3 of the present invention, the upper and lower contrast ratios are calculated as follows:

[0016] 1) Calculate the X and Y coordinates of the center of gravity of the defect in the first image; divide the defect into upper and lower parts with the row coordinate Y of the center of gravity as the boundary;

[0017] 2) Divide the upper part of the defect into three equal areas: upper, middle and lower. Take the middle area and calculate the average grayscale Gray1 of the area.

[0018] 3) Divide the lower area of ​​the defect into three equal areas: upper, middle and lower. Take the middle area and calculate the average grayscale Gray2 of the area.

[0019] 4) Upper and lower contrast Contrast1 = Gray1 - Gray2.

[0020] Furthermore, in step S4 of the present invention, the internal and external contrast is calculated as follows:

[0021] a. In the defect area, use threshold segmentation to segment the dark area in the defect;

[0022] b. Calculate the average grayscale Gray3 of the dark area;

[0023] c. Subtract the dark area from the defect area to obtain the surrounding bright area, and calculate the average grayscale Gray4 of the bright area;

[0024] d. Calculate the internal and external contrast of the defect: Contrast2 = Gray4 - Gray3.

[0025] Furthermore, in step S5 described in the present invention, if the upper and lower contrast of the defect area is less than 0, the defect is judged as a pit; if the upper and lower contrast of the defect area is greater than 0, and the inner and outer contrast is greater than 5, the defect is judged as a foreign body; if the upper and lower contrast of the defect area is greater than 0, and the inner and outer contrast is less than 5, the defect is judged as a bubble.

[0026] The beneficial effects of the present invention are that the defects existing in the background technology are solved.

[0027] By utilizing time-sharing exposure, two low-angle white light sources and high-angle infrared light sources in two different directions are used to obtain two images of the product at one time. One image is used to show the concave-convex feeling of the product surface, and the other image is used to show the color information of the product packaging. The two images are used in combination for defect detection; this not only improves the detection accuracy, but also can accurately detect and classify defects. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a hardware optical path diagram of the imaging system of the present invention;

[0029] Figure 2 Schematic diagrams of imaging results of bubbles, foreign matter and pits under the first light source of the present invention;

[0030] Figure 3 Schematic diagrams of imaging results of bubbles, foreign matter and pits under the second light source of the present invention;

[0031] Figure 4 It is a schematic diagram of the process of the present invention;

[0032] In the picture: 1. Lithium battery blue film product; 2. Camera; 3. First light source; 4. Second light source. DETAILED DESCRIPTION

[0033] The present invention will now be described in further detail in conjunction with the accompanying drawings and preferred embodiments. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0034] A method for classifying blue film surface defects, targeting surface defects of a lithium battery blue film product 1, comprises the following steps:

[0035] S1. Hardware Construction

[0036] like Figure 1 As shown, the imaging system has two light sources, the first light source 3 has an angle of 25 degrees, the second light source 4 has an angle of 0 degrees, and the camera 2 has an angle of 25 degrees. The 25-degree first light source is white line light, and the second light source is infrared light. Using the time-sharing exposure principle, two images are generated in one scan.

[0037] S2. Scan and print

[0038] like Figure 2 and Figure 3 As shown, bubbles appear bright on top and dark on the bottom under the first light source, and bright on the bottom under the second light source; foreign matter appears bright on top and dark on the bottom under the first light source, and bright on the outer circle and dark on the inside under the second light source; pit defects appear bright on top and dark on the bottom under the first light source, and have no obvious features under the second light source.

[0039] S3. First image feature extraction

[0040] The first image mainly shows the concave and convex features of the defect. The concave features appear dark on the top and bright on the bottom in the first image, and the convex features appear bright on the top and dark on the bottom in the first image. The upper and lower contrast of the defect is used for judgment. The upper and lower contrast is calculated as follows:

[0041] 1) Calculate the X and Y coordinates of the center of gravity of the defect in the first image. Divide the defect into upper and lower parts using the Y coordinate of the center of gravity as the boundary.

[0042] 2) Divide the upper part of the defect into three equal areas: upper, middle and lower. Take the middle area and calculate the average grayscale Gray1 of the area.

[0043] 3) Divide the lower part of the defect into three equal areas: upper, middle and lower. Take the middle area and calculate the average grayscale Gray2 of the area.

[0044] 4) Upper and lower contrast Contrast1 = Gray1 - Gray2.

[0045] S4. Second image feature extraction

[0046] The second image mainly shows the characteristics of bubbles and foreign matter. Bubbles appear as bright features in the second image, and foreign matter appears as bright features in the outer circle and dark features in the outer circle in the second image. The internal and external contrast of the defect is calculated to make a judgment. The calculation method of the internal and external contrast is as follows:

[0047] a. In the defect area, use threshold segmentation to segment the dark area within the defect.

[0048] b. Calculate the average grayscale Gray3 of the dark area.

[0049] c. Subtract the dark area from the defect area to obtain the surrounding bright area, and calculate the average grayscale Gray4 of the bright area.

[0050] d. Calculate the internal and external contrast of the defect: Contrast2 = Gray4 - Gray3.

[0051] S5. Defect Classification

[0052] If the upper and lower contrast ratio contrast1 of the defect area is less than 0, the defect is judged as a pit;

[0053] If the upper and lower contrasts of the defect area, contrast1, are greater than 0 and contrast2 is greater than 5, the defect is judged as a foreign body.

[0054] If the upper and lower contrasts of the defect area, contrast1, are greater than 0 and contrast2 is less than 5, the defect is judged as a bubble.

[0055] The above description only describes the specific implementation mode of the present invention. Various examples do not limit the essential content of the present invention. After reading the description, ordinary technicians in the relevant technical field can modify or deform the specific implementation modes described above without departing from the essence and scope of the invention.

Claims

1. A method for classifying surface defects of a blue film, characterized by: The following steps are included: S1. Build the hardware structure of the imaging system; S2, using the time-sharing exposure principle to take a photo at a time to simultaneously obtain two images of the product, that is, obtain a first image under a first light source and obtain a second image under a second light source; S3, extracting features of defects on the first image, and calculating upper and lower contrasts of the defects; S4, extracting features of defects on the second image, and calculating the contrast between inside and outside of the defects; S5. Compare the upper and lower contrast and the inner and outer contrast with the set thresholds respectively, so as to classify the defects.

2. A method for classifying blue film surface defects as claimed in claim 1, characterized in that: In the step S1, the hardware structure of the imaging system includes a first light source, a second light source and an imaging component; the second light source is vertically above the product, and the angle between the second light source and the product is 0°; the first light source is arranged on one side of the second light source, and the angle between the first light source and the second light source is 25°; the imaging component is arranged on the other side of the second light source, and the angle between the imaging component and the second light source is also 25°; the first light source is white line light; the second light source is infrared light.

3. A method for classifying blue film surface defects as claimed in claim 1, characterized in that: In step S2, the defects included in the first image and the second image respectively are bubbles, foreign matter and / or pits; bubbles appear bright on the top and dark on the bottom under the first light source, and appear bright under the second light source; foreign matter appears bright on the top and dark on the bottom under the first light source, and appears bright on the outer circle and dark on the inside under the second light source; pits appear bright on the top and dark on the bottom under the first light source, and have no obvious features under the second light source.

4. A method for classifying blue film surface defects as claimed in claim 1, characterized in that: In step S3, the upper and lower contrast ratios are calculated as follows: 1) Calculate the X and Y coordinates of the center of gravity of the defect in the first image; divide the defect into upper and lower parts with the row coordinate Y of the center of gravity as the boundary; 2) Divide the upper part of the defect into three equal areas: upper, middle and lower. Take the middle area and calculate the average grayscale Gray1 of the area. 3) Divide the lower area of ​​the defect into three equal areas: upper, middle and lower. Take the middle area and calculate the average grayscale Gray2 of the area. 4) Upper and lower contrast Contrast1 = Gray1 - Gray2.

5. A method for classifying blue film surface defects as claimed in claim 1, characterized in that: In step S4, the internal and external contrast is calculated as follows: a. In the defect area, use threshold segmentation to segment the dark area in the defect; b. Calculate the average grayscale Gray3 of the dark area; c. Subtract the dark area from the defect area to obtain the surrounding bright area, and calculate the average grayscale Gray4 of the bright area; d. Calculate the internal and external contrast of the defect: Contrast2 = Gray4 - Gray3.

6. A method for classifying blue film surface defects as claimed in claim 1, characterized in that: In step S5, if the upper and lower contrast of the defect area is less than 0, the defect is judged as a pit; if the upper and lower contrast of the defect area is greater than 0, and the inner and outer contrast is greater than 5, the defect is judged as a foreign body; if the upper and lower contrast of the defect area is greater than 0, and the inner and outer contrast is less than 5, the defect is judged as a bubble.

Citation Information

Patent Citations

  • Surface flaw detection method, surface flaw detection device, and manufacturing method for steel material

    CN107735673A

  • Transparent material internal defect detection method based on time-sharing exposure image synthesis

    CN110441321A

  • Defect detection method, system and device and computer-readable storage medium

    CN110687119A

  • Method for detecting surface defects of perovskite battery film

    CN116256373A

  • Defect detection method and device for battery cell blue film

    CN116626053A