Warp knitting machine cloth flaw on-line visual inspection method based on Gabor transformation

Abstract

The invention provides a warp knitting machine cloth flaw on-line visual inspection method comprising Gabor transformation, character extraction, and flaw identification. With the method, warp knitting machine cloth real-time monitoring and machine stopping on detecting flaw can be realized. With the method, defects such as large data calculation amount and information redundancy when traditional Gabor multi-scale decomposition-fusion is used in the field of flaw detection can be solved. Texton cycle is automatically searched, such that window automatic separation can be carried out by the system, characteristic extraction is facilitated, and non-precision caused by different object manual window size input is avoided. With an offline detection algorithm, fast Gabor transformation, fast characteristic extraction and flaw identification can be realized under the guidance of the offline data. Therefore, requirements of high timeliness and high accuracy of warp knitting machine cloth flaw on-line detection can be completely satisfied.

Description

technical field [0001] The invention relates to the field of on-line detection of warp knitting machine defects by using machine vision, and specifically refers to an image processing method for timely shutdown of warp knitting machine cloth with high real-time requirements when defects occur during the weaving process, which is applied to industrial sites. Background technique [0002] Internationally, the requirements for quality testing methods are increasing day by day. Textiles are a necessity in people's daily life, and their quality directly affects people's quality of life. The quality of textiles is reflected in reliability, aesthetics, safety and service life. High-quality textiles can better meet the needs of consumers and society, thereby increasing the value of the textiles. Therefore, in the textile industry, how to ensure the quality of textiles is a very critical link for manufacturers. During the weaving process of the warp knitting machine, due to mechanic...

AI Technical Summary

Problems solved by technology

The traditional manual detection technology is costly, slow, and the human eye is prone to fatigue due to long-term work, resulting in false detection. In this way, the automatic detection technology of fabric defects has emerged as the times require, and has gradually become an urgent problem that restricts the textile level.

For

Images