A detection method for test paper surface texture defects based on gray gradient clustering

Abstract

The invention discloses a test paper surface texture defect detection method based on gray scale gradient clustering, comprising the steps of: collecting a frame image of test paper, performing gray scale and median filter preprocessing on the image; The image is binary segmented, and the test paper area is extracted by the difference method; the Gama grayscale enhancement is performed on the image and a Gaussian low-pass filter is used to filter out some periodic textures; a unidirectional Gaussian kernel function is constructed to convolve the image in the vertical direction Filtering; calculate the gradient grad_x of the image in the horizontal direction; divide the test paper area into n columns of sub-areas in the horizontal direction, and calculate the position of the maximum gradient area of ​​each sub-area; place the position of the maximum gradient area of ​​each sub-area in the vertical direction Carry out cluster calculation, and mark the region where the number of regional clusters reaches the threshold range as the texture defect region; judge whether the test paper is qualified according to the marked region. The invention has the advantages of fast detection speed, high detection precision, good robustness and the like.

Description

technical field [0001] The invention belongs to the technical field of computer vision, in particular to a method for detecting texture defects on the surface of test paper based on gray gradient clustering. Background technique [0002] At present, the automatic detection technology for texture defects on the test paper surface is relatively immature, and the missed detection rate and false detection rate for subtle texture defects are relatively high. Under normal circumstances, the detection of surface texture defects of test paper products is done manually, but the judgment of the surface texture of test paper by naked eyes is subjective, and because of the large batch of test paper, long-term detection will cause visual fatigue of the testing staff , thereby reducing the detection efficiency and accuracy. [0003] With the rapid development of computer technology, the detection efficiency can be greatly improved by applying computer vision detection technology to the i...

AI Technical Summary

Problems solved by technology

For example: Chinese invention patent CN201711262071.9, announced a fabric defect detection method based on spectral curvature analysis, which has good adaptive and anti-interference capabilities, but due to the need to convert the spatial domain image to the frequency domain and then to the spatial domain, It is difficult to guarantee real-time detection of large images; Chinese invention patent CN201310362813.0 discloses a textile defect detection algorithm based on texture gradient, which has the advantage of quickly and accurately distinguishing textile defects

[0004] However, the above methods are all aimed at the detection of textiles, and have limitations on the defects caused by the subtle texture changes on the surface of the test paper. At present, there is no effective visual detection algorithm that can effectively detect the test paper with texture defects in a periodic texture background.

Images