Paper wrinkle defect detection method and device

Abstract

The invention discloses a paper wrinkle defect detection method and device. The method comprises the steps of obtaining a to-be-detected image and a to-be-trained image; preprocessing the to-be-detected image and the to-be-trained image to extract a paper area image range of the to-be-detected image and a paper area image range of the to-be-trained image; calculating the classification of each pixel point in the paper area image range of the current to-be-detected image by adopting a preset detection model and trained model parameters, analyzing the probability value that each pixel point is wrinkled, and obtaining a wrinkled area according to the obtained probability value that each pixel point is wrinkled; wherein the detection model comprises a feature map, and each pixel value of the feature map is defined as a probability value identifying whether the pixel is a wrinkle or not. According to the paper wrinkle defect detection method, pixel-by-pixel classification is adopted, the advantage of extremely low missing detection risk is achieved, and meanwhile high detection precision and online detection real-time performance are achieved.

Description

technical field [0001] The invention belongs to the technical field of image detection, and in particular relates to an online paper wrinkle defect detection method and device. Background technique [0002] Paper wrinkles are creases caused by uneven pressure when the paper passes through the cylinder during the printing process. Wrinkled areas are prone to serious printing defects such as ink shortage and printing deformation, which pose a great risk to printing quality. The traditional method to detect paper wrinkle mainly has the following steps: [0003] Step 1, use a linear array CCD camera, a lens, a special infrared light source and a flattening device to perform infrared imaging on the printed matter moving on the roller; [0004] Step 2, use the convolution operator to extract the wrinkle features from the collected image, and eliminate the shadow caused by the unevenness of the paper; [0005] Step 3, the detection image is consistent with the detection template...

AI Technical Summary

Problems solved by technology

[0007] figure 1 is the paper wrinkle image, figure 2 It is the processing result diagram of the traditional paper wrinkle detection method using the above steps. As shown in the figure, since the convolution operator is used to extract edge features, it is easily disturbed by the image texture, and it is impossible to accurately determine whether there are wrinkles at the image texture. Detect Weaknesses

Images