Aluminum foil quality defect detection method based on image processing

Abstract

The invention discloses an aluminum foil quality defect detection method based on image processing. Comprising the following steps: calculating to obtain a maximum enclosing rectangle of a maximum outline; segmenting an image in the rectangle to obtain an image; predicting the segmented image; integrally predicting the collected image; judging whether the defect detection is determined or not according to the predicted score, and performing area or length calculation on the detected defect by using halcon; and finally, finely controlling whether the defect is detected or not through secondary screening of the area or length of the defect. According to the invention, the aluminum foil defect identification speed and precision can be effectively improved, the omission ratio of manual detection and the production cost are reduced, the requirements of the manufacturing industry for development towards intelligentization and automation are met, and good economic and social benefits are brought.

Description

Technical Field

[0001] The invention relates to a method for detecting quality defects of aluminum foil. Background Art

[0002] Low-voltage chemical-formed foil is the core material for electronic components such as aluminum electrolytic capacitors. Its surface quality directly affects the capacitor's voltage resistance, capacity stability, and lifespan. With the rapid development of new energy, consumer electronics, 5G communications and other fields, the market demand for high-performance chemical-formed foil has surged, and quality requirements are becoming increasingly stringent. High-end capacitors have a "zero tolerance" for defects. For example, if high-voltage capacitors used in new energy vehicles fail due to foil defects, it may pose a safety hazard. Defects in low-voltage chemical-formed foil include pinholes, scratches, uneven oxidation, black lines, bright spots, and other defects that are difficult to accurately identify using traditional equipment. Existing manual visual inspection is inefficient: it relies on the naked eye or simple optical instruments for detection, is easily affected by fatigue and subjective judgment, and has high missed detection and false detection rates. The visual solution uses high-resolution line scan cameras and high-speed image processing chips to solve the hardware bottleneck of high-speed imaging and real-time analysis. The current machine vision solution relies on the mature application of traditional algorithms combined with deep learning algorithms in the field of defect detection, providing a technical foundation for a variety of tiny defects in complex backgrounds. It detects and marks aluminum foil defects in real time and synchronizes the detected defects to the relevant web page to realize an online real-time feedback system. It also collects statistical defect distribution data to guide process optimization and form a closed-loop quality control.

[0003] Micron-level defects (such as pinholes and microcracks) are easily missed due to the limited resolution of traditional optical systems. Highly reflective surfaces (such as the anodized layer of aluminum foil) result in low image contrast, blurred defect edges, and misidentification of reflective noise as defects. Defects can take many forms, such as uneven oxidation that manifests as color aberration, linear scratches, and black spots that represent foreign matter. Traditional algorithms that rely solely on manually set thresholds struggle to dynamically adapt to varying defect characteristics. This results in low detection efficiency, difficulty accurately identifying defects, and high labor costs. Summary of the Invention

[0004] The purpose of the present invention is to provide an aluminum foil quality defect detection method based on image processing, which can inspect and alarm the aluminum foil production line in real time through multiple camera stations and upload statistical defects to help manufacturers optimize the aluminum foil production process.

[0005] The technical solution of the present invention is: A method for detecting aluminum foil quality defects based on image processing is characterized by comprising the following steps: Step 1: First, use the algorithm to obtain the area with the largest outline in the collected aluminum foil standard image to determine the location of the aluminum foil; Step 2: Calculate the maximum circumscribed rectangle of the largest contour and segment the image within this rectangle. To prevent defects from being segmented into different images, each cut in the same direction must be made forward or upward to ensure some overlapping areas. These cut images are then passed to the model, which returns the predicted classification and score for these cut images. Step 3: The entire image is passed to the model, and the model returns the predicted classification and score for the entire image; Step 4: It is necessary to cut the edges of the object to be inspected, obtain the four points in the quadrilateral outline, and then calculate the slope of the quadrilateral, and cut the edge image in sequence according to the slope; these edge images are passed into the model, and the model returns the predicted classification and score of these edge images. The classification and score in the configuration file are used to determine which of these images are defective images in the first screening, and then different halcon image processing algorithms are used to calculate the corresponding defect area or length according to the defect type; finally, the defect area or length is screened twice to fine-tune the control of whether the defect is detected.

[0006] In the second step, the image within the rectangle is segmented into 5 segments in the X-axis direction and 8 segments in the Y-axis direction to ensure that the final segmented images have similar aspect ratios.

[0007] In the fourth step, the cropping size of the edge image is set to 200*200.

[0008] Method for capturing standard images of aluminum foil: The width of the aluminum foil is 500mm. Since double-sided inspection is required, the inspection mechanism camera captures standardized images of both sides of the aluminum foil. The image acquisition device uses an 8k line scan camera, and the triggering method adopts frame trigger plus line signal to obtain fixed-length images. The sensor is used to trigger the frame signal, and the encoder is used to continuously output the line signal to obtain the standard image.

[0009] When identifying edge defects, there is background interference. After obtaining the vertices of the aluminum foil quadrilateral, the slope value of each edge is calculated. The edge image is then cut more finely according to the slope to reduce background interference, thereby improving the defect detection rate and reducing the false alarm rate.

[0010] During the image segmentation process, defects are separated into different images, causing the defect features to be scattered and leading to missed detection. By overlapping the images, the defect features are segmented into a single image, improving the defect detection rate.

[0011] The present invention can effectively improve the speed and accuracy of aluminum foil defect recognition, reduce the missed detection rate and production cost of manual inspection, meet the requirements of the manufacturing industry towards intelligent and automated development, and will bring good economic and social benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The present invention will be further described below with reference to the accompanying drawings and examples.

[0013] Figure 1 is a flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0014] A method for detecting aluminum foil quality defects based on image processing is characterized by comprising the following steps: Step 1: First, use the traditional algorithm to obtain the area with the largest contour in the collected aluminum foil standard image to determine the location of the aluminum foil; Method for capturing standard images of aluminum foil: The width of the aluminum foil is 500mm. Since double-sided inspection is required, the inspection mechanism camera captures standardized images of both sides of the aluminum foil. The image acquisition device uses an 8k line scan camera, and the triggering method adopts frame trigger plus line signal to obtain fixed-length images. The sensor is used to trigger the frame signal, and the encoder is used to continuously output the line signal to obtain the standard image.

[0015] Step 2: Calculate the maximum circumscribed rectangle of the largest contour and segment the image within this rectangle. To prevent defects from being segmented into different images, each cut in the same direction must be made forward or upward to ensure some overlapping areas. These cut images are then passed into the model (a mathematical function and parameter set formed through image data training statistics that can map input data to output prediction results). The model returns the predicted classification and score for these cut images. Step 3: The entire image is passed to the model, and the model returns the predicted classification and score of the entire image; Step 4: It is necessary to cut the edges of the object being inspected, because the object is tilted. If the edge is cut directly by the circumscribed rectangle of the largest contour, a large amount of background will be cut into the image. Therefore, the four points in the quadrilateral contour are obtained, and then the slope of the quadrilateral is calculated, and the edges are cut in sequence according to the slope; these edge images are passed into the model, and the model returns the predicted classification and score of these edge images. The classification and score in the configuration file are used to determine which of these images are defective images in the first screening. For example, when the threshold is set to 0.5, it is considered that images greater than 0.5 are credible in the classification, that is, if the classification is a defect classification, it is considered a defect, otherwise it is not a defect. Different halcon image processing algorithms are used to calculate the corresponding defect area or length according to the defect type; finally, the defect area or length is screened twice to fine-tune control of whether the defect is detected.

[0016] In the second step, the image within the rectangle is segmented into 5 segments in the X-axis direction and 8 segments in the Y-axis direction to ensure that the final segmented images have similar aspect ratios.

[0017] In the fourth step, the cropping size of the edge image is set to 200*200.

Claims

1. A method for detecting aluminum foil quality defects based on image processing, characterized by: The following steps are involved: Step 1: First, use the algorithm to obtain the area with the largest outline in the collected aluminum foil standard image to determine the location of the aluminum foil; Step 2: Calculate the maximum circumscribed rectangle of the largest contour and segment the image within this rectangle. To prevent defects from being segmented into different images, each cut in the same direction must be made forward or upward to ensure some overlapping areas. These cut images are then passed to the model, which returns the predicted classification and score for these cut images. Step 3: The entire image is passed to the model, and the model returns the predicted classification and score for the entire image; Step 4: It is necessary to cut the edges of the object to be inspected, obtain the four points in the quadrilateral outline, and then calculate the slope of the quadrilateral, and cut the edge image in sequence according to the slope; these edge images are passed into the model, and the model returns the predicted classification and score of these edge images. The classification and score in the configuration file are used to determine which of these images are defective images in the first screening, and then different halcon image processing algorithms are used to calculate the corresponding defect area or length according to the defect type; finally, the defect area or length is screened twice to fine-tune the control of whether the defect is detected.

2. The aluminum foil quality defect detection method based on image processing according to claim 1 is characterized in that: in the second step, the image within the rectangle is segmented into 5 segments in the X-axis direction and 8 segments in the Y-axis direction to ensure that the final segmented images have similar aspect ratios.

3. The aluminum foil quality defect detection method based on image processing according to claim 1 or 2, characterized in that: In the fourth step, the cropping size of the edge image is set to 200*200.

4. The aluminum foil quality defect detection method based on image processing according to claim 1 or 2, characterized in that: Method for capturing standard images of aluminum foil: The width of the aluminum foil is 500mm. Since double-sided inspection is required, the inspection mechanism camera captures standardized images of both sides of the aluminum foil. The image acquisition device uses an 8k line scan camera, and the triggering method adopts frame trigger plus line signal to obtain fixed-length images. The sensor is used to trigger the frame signal, and the encoder is used to continuously output the line signal to obtain the standard image.

Images