Metal reflecting film defect online identification method and storage medium

By calculating the detection accuracy of the industrial scanning camera and combining it into a minimum rectangle and resetting the pixel size, the problem of low contour screening accuracy in metal reflective film defect detection is solved, and higher detection accuracy and reliability are achieved.

CN119086571AInactive Publication Date: 2024-12-06DONGGUAN GUANGZHI PHOTOELECTRIC CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411409277.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-10
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing metal reflective film defect detection methods are not accurate when converting the image into binary images and performing contour screening, which may lead to excessive or too small contour size, affecting the accuracy of the detection results.

Method used

By determining the number and movement speed of industrial scanning cameras, calculating the horizontal and vertical detection accuracy, combining it into a minimum rectangle, resetting the pixel length and width to improve detection accuracy, and filtering the binary image according to the qualified area and threshold value, forming The area to be inspected is compared to its pixel length and width to mark defects.

Benefits of technology

It improves the accuracy and reliability of metal reflective film defect detection, ensures that the details of the detected defect can be accurately reflected and meets the high-precision detection needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119086571A_ABST
    Figure CN119086571A_ABST
Patent Text Reader

Abstract

The invention relates to a metal reflecting film defect online identification method. The method comprises the following steps: obtaining transverse detection precision and longitudinal detection precision according to the moving speed of a metal reflecting film, the number of industrial scanning cameras and the width of a product; the transverse detection precision and the longitudinal detection precision are combined into a minimum rectangle, and the length and the width of a single pixel are set according to the length and the width of the minimum rectangle; setting a qualified area and a threshold value; obtaining an original image on the metal reflecting film through an industrial scanning camera, and converting the original image into a binary image; performing contour screening on the binary image according to a set threshold value and forming a to-be-detected area; determining according to the number of pixels occupied by the to-be-detected area and the number of pixels occupied by the qualified area, and determining a minimum rectangle through transverse detection precision and longitudinal detection precision; and then the length and the width of the pixel are set according to the reset of the minimum rectangle, so that the details of the detected defect can be reflected more finely, and the accuracy and the reliability of detection are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of film material defect detection, and in particular to an online metal reflective film defect recognition method and a storage medium. Background Art

[0002] If the reflective surface of the metal reflective film is uneven or there is leakage during production, the metal reflective film will have uneven density or pinholes.

[0003] The Chinese patent with patent announcement number CN118396959A discloses a method, device, storage medium, electronic device and program product for detecting defects based on an event camera. The method includes: preprocessing the event data related to the detected object collected by the event camera to obtain the original image, wherein the types of preprocessing include: frame accumulation operation and motion compensation; converting the denoised image obtained after denoising the original image into a binary image; performing contour screening on the binary image to obtain the defect detection result. Some embodiments of the present application can improve the efficiency and accuracy of defect detection. In the above scheme, an event camera is used to obtain the original image. In the process of converting the image into a binary image and performing contour screening on the binary image, two situations may occur. The first situation is that all pixels with pixel values ​​are screened as contours, resulting in the obtained contour size being larger than the actual contour size. The second situation is that only the contour composed of pixels with the highest pixel value can be screened, resulting in the obtained size being smaller than the actual contour size. The result is that the accuracy of contour screening is not high. Summary of the invention

[0004] The purpose of the present invention is to provide a method for online identification of metal reflective film defects, in order to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the technical solution adopted by the present invention is: a method for online identification of metal reflective film defects, comprising the following steps:

[0006] The first step is to obtain the number of industrial scanning cameras based on the width of the metal reflective film.

[0007] The second step is to obtain the horizontal detection accuracy and the vertical detection accuracy according to the moving speed of the metal reflective film, the number of industrial scanning cameras and the width of the product; the horizontal detection accuracy and the vertical detection accuracy are combined into a minimum rectangle;

[0008] The third step is to set the length and width of the minimum rectangle to the length and width of a single pixel;

[0009] The fourth step is to set the qualified area according to the minimum rectangle, set the length and width of the qualified area and set the threshold;

[0010] The fifth step is to obtain the original image on the metal reflective film through an industrial scanning camera, and convert the original image into a binary image; the binary image is contour screened according to the threshold set in the fourth step to form a to-be-tested area;

[0011] The sixth step is to compare the length and width of the pixels occupied by the area to be inspected with the length and width of the pixels occupied by the qualified area.

[0012] Compared with the existing technology, the minimum rectangle is determined by the horizontal detection accuracy and the vertical detection accuracy; then the length and width of the pixel are reset according to the minimum rectangle, which can more finely reflect the details of the detected defects, thereby improving the accuracy and reliability of detection.

[0013] According to a preferred technical solution of the present invention, the number of the industrial scanning cameras is determined according to the field of view of the industrial scanning cameras and the width of the metal reflective film.

[0014] According to the preferred technical solution of the present invention, the field of view of the industrial scanning camera is determined according to the shooting distance from the lens to the defect, the model of the industrial scanning camera and the focal length of the industrial scanning camera.

[0015] According to the preferred technical solution of the present invention, the lateral detection accuracy is determined according to the detection width, the number of industrial scanning cameras and the working accuracy of the industrial scanning cameras.

[0016] In a preferred technical solution of the present invention, the detection width is the sum of the width of the metal reflective film, the reserved width and the degree of image overlap.

[0017] In the preferred technical solution of the present invention, the longitudinal detection accuracy is determined according to the moving speed of the metal reflective film and the camera scanning frequency.

[0018] According to the preferred technical solution of the present invention, the longitudinal detection accuracy is determined according to the moving speed of the metal reflective film, the camera clock frequency and the lateral detection accuracy.

[0019] The preferred technical solution of the present invention is to mark according to the result of the determination in the sixth step.

[0020] If the length and width of the pixels occupied by the inspected area are less than or equal to the length and width of the pixels occupied by the qualified area, the defect will not be marked or will be marked as qualified;

[0021] If one of the length and width of the pixels occupied by the area to be inspected is greater than the length and width of the pixels occupied by the qualified area, the defect will be marked as unqualified or as a warning mark.

[0022] A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that the computer program implements the film material defect detection method when executed by a processor.

[0023] In addition to the technical problems solved by the present invention, the technical features that constitute the technical solution, and the advantages brought by the technical features of the technical solution described above, other technical problems that can be solved by the present invention, other technical features included in the technical solution, and the advantages brought by these technical features will be further described in detail in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flow chart of the steps of the present invention.

[0025] Figure 2 It is a schematic diagram of the inspection width of the present invention.

[0026] Figure 3 It is a schematic diagram of the qualified area of ​​the present invention.

[0027] Figure 4 It is a schematic diagram of the original image enlargement of the present invention. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0029] See also Figure 1-4 As shown, the metal reflective film defect online identification method of the present invention is used to perform online detection of the metal reflective film during the process of cutting and producing the metal reflective film.

[0030] Since the metal reflective film is in a state of constant speed movement during the cutting process, the industrial scanning camera (linear camera) and the light source are used to scan the moving metal reflective film to obtain defects on the metal reflective film.

[0031] First, the number of industrial scanning cameras is determined according to the width of the metal reflective film, and the multiple industrial scanning cameras are placed on the same plane.

[0032] The number of industrial scanning cameras is determined according to the field of view of the industrial scanning camera and the width of the metal reflective film. The field of view of the industrial scanning camera is determined according to the shooting distance from the lens to the defect, the model of the industrial scanning camera and the focal length of the industrial scanning camera.

[0033] The relationship between the field of view, the distance from the lens to the defect, the model of the industrial scanning camera, and the focal length of the industrial scanning camera is as follows:

[0034]

[0035] Camera model size refers to the pixel size.

[0036] The number of industrial scanning cameras is obtained by dividing the width of the metal reflective film by the field of view width.

[0037] Secondly, the imaging principle of industrial scanning cameras is pixel-based image stitching technology. At the same time, the metal reflective film is in a moving state during the detection process. Therefore, it is necessary to determine the horizontal detection accuracy (D) and the vertical detection accuracy (H) based on the moving speed of the metal reflective film, the number of industrial scanning cameras, and the width of the product.

[0038] The horizontal detection accuracy (D) is determined by the detection width, the number of industrial scanning cameras, and the working accuracy of the industrial scanning cameras. The working accuracy is also the resolution. The detection width is the sum of the width of the metal reflective film, the reserved width, and the degree of image overlap (such as Figure 2 shown).

[0039] The relationship between lateral detection accuracy (D), detection width, number of industrial scanning cameras, and working accuracy of industrial scanning cameras is as follows:

[0040]

[0041] When determining the lateral detection accuracy (D), the degree of image overlap (over l ap) is added. By adjusting the size of the overlap l ap, a smoothing filter process is implemented to the image. This process is mainly a low-frequency enhanced spatial domain filtering technology, which aims to blur the image or eliminate noise. The smoothing filter in the spatial domain is usually performed by a simple averaging method, that is, calculating the average brightness value of adjacent pixel points to improve the image quality.

[0042] At the same time, the correct over l ap setting also helps to calibrate the scanner to correct the problem that the scanner carriage head cannot accurately read the image position due to mechanical tolerance. Through the calibration process, the offset value of the scanner can be calculated and stored, so that the defect position can be correctly captured in subsequent scans, improving the scanning quality.

[0043] At the same time, the reserved width is increased so that the metal reflective film can be fully inspected in the width direction during operation.

[0044] The longitudinal detection accuracy (H) is determined according to the moving speed of the metal reflective film (the moving speed of the metal reflective film) and the camera scanning frequency or determined by the moving speed of the metal reflective film, the camera clock frequency and the lateral detection accuracy. Since the metal reflective film is in a moving state during the detection process, the camera scanning frequency is determined according to the camera clock frequency and the lateral detection accuracy (D).

[0045] The longitudinal detection accuracy (H) is determined by the moving speed of the metal reflective film and the camera scanning frequency, and the relationship is as follows:

[0046] At the same time, the longitudinal detection accuracy (H) can also be determined according to the moving speed of the metal reflective film, the camera clock frequency and the lateral detection accuracy. The relationship is as follows:

[0047]

[0048] The combination of the above-mentioned transverse detection accuracy (D) and longitudinal detection accuracy (H) forms a minimum rectangle, which represents the minimum defect size that can be detected.

[0049] Next, set the pixel length and width based on the length and width of the smallest rectangle.

[0050] This is because in defect detection, the size of the pixel directly affects the precision and accuracy of the detection. When the pixel size is too large, it will not be able to accurately capture smaller defects, while when the pixel size is small enough, it can more finely reflect the details of the detected defects, thereby improving the accuracy and reliability of the detection. Specifically, if the length and width of the design pixel are designed according to the detection accuracy, it means that each pixel can cover a smaller area, so that smaller defects can be identified more accurately. This design principle helps ensure that even tiny defects can be clearly identified during image processing and analysis, thereby meeting high-precision detection needs.

[0051] Again, the qualified area is set according to the minimum rectangle, the length and width of the qualified area are set, and the threshold is set.

[0052] Then, the original image on the metal reflective film is obtained by an industrial scanning camera, and the original image is converted into a binary image; the binary image is contour screened according to the set threshold value to form the area to be inspected.

[0053] Finally, the length and width of the pixels occupied by the area to be inspected are compared with the length and width of the pixels occupied by the qualified area.

[0054] If the number of pixels occupied by the inspected area is less than the number of pixels occupied by the qualified area, the defect will not be marked or will be marked as qualified;

[0055] If the number of pixels occupied by the area to be inspected is greater than the number of pixels occupied by the qualified area, the defect will be marked as unqualified or as a warning mark.

[0056] The following are examples

[0057] For example, take a silver reflective film with a width of 1300mm. The distance from the lens to the metal reflective film is 1225mm, the camera size is 6.4, the focal length of the industrial scanning camera lens is 6mm, the moving speed of the metal reflective film is 30m / min, the industrial scanning camera is an 8K camera (working accuracy is 8192), the single-side reserved width is 50mm, and the camera scanning frequency is 3000kz.

[0058] First, determine the field of view of the industrial scanning camera lens.

[0059]

[0060] According to the above calculation, the field of view width is 1306.7 mm. Since the width of the silver reflective film is 1300 mm, an industrial scanning camera is required. The field of view of an industrial scanning camera is 1306.7 mm, which is greater than the width of the metal reflective film.

[0061] Secondly, the horizontal detection accuracy (D) and the vertical detection accuracy (H) are determined according to the moving speed of the metal reflective film and the product width;

[0062]

[0063] The lateral detection accuracy (D) is approximately equal to 0.17;

[0064] In this embodiment, the longitudinal detection accuracy (H) is determined according to the moving speed of the metal reflective film and the camera scanning frequency.

[0065]

[0066] The longitudinal detection accuracy (H) is approximately equal to 0.17;

[0067] The width combination minimum rectangle of the horizontal detection accuracy (D) and the vertical detection accuracy (H) has a length of 0.17 and a width of 0.17;

[0068] From this point on, the length and width of the smallest rectangle are set to a single pixel length and width.

[0069] Again, the length and width of the qualified area are set according to the length and width of the minimum rectangle. Therefore, the length and width of the qualified area are set to be twice the minimum rectangle, with a length of 2H and a width of 2D (such as Figure 3 As shown), the rectangular length and width are respectively a square of 0.34 mm.

[0070] The threshold value 190 is set for the industrial scanning camera.

[0071] The original image is obtained by scanning the metal reflective film with an industrial scanning camera, and the original image is converted into a binary image. After the original image is enlarged, it can be found that the pixel values ​​within the pixel are different (such as Figure 4 As shown in the figure). Because during the inspection process, light cannot pass through the qualified area on the metal reflective film, but light can pass through the defective area, so the pixel values ​​in the pixels on the original image are different. As a result, the pixel values ​​corresponding to the pixels occupied by the qualified area are higher than the pixel values ​​corresponding to the pixels occupied by the defect.

[0072] like Figure 4 As shown, defects of different shapes are named A defect, B defect, C defect, D defect, and E defect respectively.

[0073] Since the threshold of the industrial scanning camera is 190, all pixels with a pixel value less than 190 are included in the defect size calculation range. Thus, the length and width of the pixels occupied by each defect in the inspection area are obtained.

[0074] Finally, the length and width of the pixels occupied by the area to be inspected are compared with the length and width of the pixels occupied by the qualified area.

[0075] From this we can see that

[0076] The length of the inspected area formed by the pixels occupied by defect A is 2H and the width is 1D, while the length of the qualified area is 2H and the width is 2D. Comparing the two, we can see that the lengths are the same, but the width of the inspected area 1D is smaller than the width of the qualified area 2D, so it is determined to be a qualified area. Therefore, defect A is determined to be a qualified area.

[0077] The length of the inspected area formed by the pixels occupied by defect B is 2H and the width is 2D, while the length of the qualified area is 2H and the width is 2D. Comparing the two, we can see that the length and width are the same, and it is determined to be a qualified area. Therefore, defect A is determined to be a qualified area.

[0078] The length of the inspected area formed by the pixels occupied by the C defect is 2H and the width is 3D, while the length of the qualified area is 2H and the width is 2D. Comparing the two, we can see that the lengths are the same. However, the width 3D of the inspected area is greater than the width 2D of the qualified area, so it is judged as an unqualified area. Therefore, an unqualified mark or warning mark is made at the C defect.

[0079] The length of the inspected area formed by the pixels occupied by the D defect is 3H and the width is 2D, while the length of the qualified area is 2H and the width is 2D. Comparing the two, we can see that the widths are the same. However, the length of the inspected area 3H is greater than the length of the qualified area 2H, and it is determined to be an unqualified area. Therefore, an unqualified mark or warning mark is made at the D defect.

[0080] The length of the inspected area formed by the pixels occupied by the E defect is 4H and the width is 1D, while the length of the qualified area is 2H and the width is 2D. Comparing the two, it can be seen that although the width 1D of the inspected area is smaller than the width 2D of the qualified area, the length 4H of the inspected area is greater than the width 2H of the qualified area. It is judged as an unqualified area, so an unqualified mark or warning mark is made at the E defect. Similarly, when the length of the inspected area is 3H and the width is 1D, since the length 3H of the inspected area is greater than the width 2H of the qualified area, it is also judged as an unqualified area.

[0081] If the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0082] The above implementation modes are merely descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering and technical personnel in the field shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for online identification of metal reflective film defects, characterized in that: The following steps are included: The first step is to obtain the number of industrial scanning cameras based on the width of the metal reflective film. The second step is to obtain the horizontal detection accuracy and the vertical detection accuracy according to the moving speed of the metal reflective film, the number of industrial scanning cameras and the width of the product; the horizontal detection accuracy and the vertical detection accuracy are combined into a minimum rectangle; The third step is to set the length and width of the minimum rectangle to the length and width of a single pixel; The fourth step is to set the length and width of the qualified area and the threshold value according to the length and width of the minimum rectangle; The fifth step is to obtain the original image on the metal reflective film through an industrial scanning camera, and convert the original image into a binary image; the binary image is contour screened according to the threshold set in the fourth step to form a to-be-tested area; The sixth step is to compare the length and width of the pixels occupied by the area to be inspected with the length and width of the pixels occupied by the qualified area.

2. The method for online identification of metal reflective film defects according to claim 1, characterized in that: The number of the industrial scanning cameras is determined according to the field of view of the industrial scanning cameras and the width of the metal reflective film.

3. The method for online identification of metal reflective film defects according to claim 2, characterized in that: The field of view of an industrial scanning camera is determined by the shooting distance from the lens to the defect, the model of the industrial scanning camera, and the focal length of the industrial scanning camera.

4. The method for online identification of metal reflective film defects according to claim 2, characterized in that: The lateral detection accuracy is determined based on the detection width, the number of industrial scanning cameras, and the working accuracy of the industrial scanning cameras.

5. The method for online identification of metal reflective film defects according to claim 4, characterized in that: The detection width is the sum of the width of the metal reflective film, the reserved width and the degree of image overlap.

6. The method for online identification of metal reflective film defects according to claim 5, characterized in that: The longitudinal detection accuracy is determined by the moving speed of the metal reflective film and the camera scanning frequency.

7. The method for online identification of metal reflective film defects according to claim 5, characterized in that: The longitudinal detection accuracy is determined by the moving speed of the metal reflective film, the camera clock frequency, and the lateral detection accuracy.

8. The method for online identification of metal reflective film defects according to claim 1, characterized in that: The result determined in the sixth step is marked. If the length and width of the pixels occupied by the inspected area are less than or equal to the length and width of the pixels occupied by the qualified area, the defect will not be marked or will be marked as qualified; If one of the length and width of the pixels occupied by the area to be inspected is greater than the length and width of the pixels occupied by the qualified area, the defect will be marked as unqualified or as a warning mark.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the defect detection method for a film material as described in any one of claims 1 to 8 is implemented.

Citation Information

Patent Citations

  • Method and device for detecting defects based on event camera, storage medium, electronic equipment and program product

    CN118396959A

  • Image acquisition method of on-line detection system for surface defects of ultra-wide thick plate

    CN110376207A

  • Image acquisition system and method based on gear shaft visual detection, product, equipment and terminal

    CN112903693A

  • Automatic optical detection method and system

    CN117825385A

  • Circuit pattern inspecting device and inspecting method thereof

    US20130271595A1