Non-woven fabric flaw detection and classification method and system based on machine vision

Abstract

The invention relates to a non-woven fabric flaw detection and classification method and system based on machine vision. The method comprises the following steps: (1) image acquisition; (2) image preprocessing; (3) extracting detail signals of the image: analyzing and extracting the detail signals of the defect image of the non-woven fabric by adopting wavelet transform; (4) calculating a texture feature image and reconstructing a feature image: calculating texture information of detail signals of the image to obtain the feature image with the capability of distinguishing a normal image and a defect image of the non-woven fabric, and reconstructing the feature image to obtain the reconstructed feature image; (5) calculating an abnormal score: calculating the abnormal score of the non-woven fabric feature image so as to judge whether flaws appear or not; (6) differential calculation of a flaw region: performing differential calculation on the original to-be-detected image and the reconstructed feature image to obtain the flaw region, and performing post-processing on the flaw to obtain a complete flaw region; and (7) flaw region classification: constructing a feature vector and training a classifier, and classifying the flaw region.

Description

technical field [0001] The invention relates to the field of machine vision detection, in particular to a non-woven fabric defect detection and classification method and system based on machine vision. Background technique [0002] With the continuous expansion of the non-woven market in the global field, the research and development and innovation of non-woven fabrics have never stopped, and new varieties have emerged one after another. Therefore, their applications in various fields are becoming more and more extensive: disposable surgical masks, protective clothing, etc. used in the medical industry. Waterproof membranes used in the construction industry, industrial filtration materials, etc. However, to produce high-quality non-woven fabrics, it is necessary to strictly control the procedures and quality of each process to ensure the final production of excellent non-woven products. [0003] At present, most factories still use manual inspection to check whether there a...

AI Technical Summary

Problems solved by technology

[0005] 1. Based on the method of grayscale binarization, due to the uneven brightness and darkness of non-woven fabric imaging, the adaptability of simply binarizing images is very poor, and there will be situations where no defects can be completely separated;

[0006] 2. Based on the method of template matching, the existing non-woven fabric defects are established as templates, so that the system can identify the defects. However, due to the wide range of changes in various defects of non-woven fabrics, the established templates usually fail, which is industrially applicable. poor;

[0007] 3. The non-woven fabric defect location method based on Gabor filtering uses multiple Gabor filtering to extract fabric surface features, which requires a large amount of calculation and good accuracy, but poor real-time performance

Images