Classification and detection of printed product surface scratches based on machine vision

Aiming at the surface defects of packaging and printing products caused in the process of packaging and printing production, a method based on machine vision to detect the scratch defects on the surface of printing products was proposed, and the common scratch defects on the surface of printing products were classified and defined. First, the image to be detected on the surface of the printed product is obtained through the industrial camera, and the image to be detected is grayed using the weighted value method. In terms of image denoising, Gaussian filtering is used for effective processing. Then, image binarization and canny edge detection operators are used to segment and extract the defect features of the pre-processed image. Finally, Fourier fast transform is used for image registration. The effectiveness of the proposed method is verified by using the matable software. The experimental results show that the accuracy of defect detection reaches 98.7% Compared with other defect detection methods, the proposed method has certain advantages in terms of convenience and accuracy and can realize rapid detection of scratches on the surface of printed products.