Real-time detection method for defect points of cloth

Abstract

The invention provides a real-time detection method for defect points of cloth. According to the real-time detection method, the real-time detection efficiency of defect points of cloth is improved when the accuracy is guaranteed, and particularly, the problems that a cloth pattern is difficult to be accurately positioned and then the pattern and the defect points cannot be accurately recognized by reasons of malposition movement due to mechanical sliding and the like after the cloth which is smaller than a standard size and is expected to be printed or jetted with a complex pattern is subjected to multiple printing or jetting processes are avoided. When the detection accuracy of the defect points which are 1mm or over higher than a peripheral region is guaranteed, the real-time detection of the cloth on a conveying belt in a real-time conveying period can be carried out, the detection speed is high, and the detection accuracy is high.

Description

technical field [0001] The invention relates to the technical field of textile production, in particular to a method for real-time detection of cloth defects. Background technique [0002] For various objective reasons, defects are unavoidable for cloth. However, as people's requirements for the beauty and complexity of cloth patterns increase, it is expected that if there are some defects in the printed or spray printed patterns on the cloth that are larger in size or protrude higher in the cloth, for example, the defects are larger than the surrounding area. 1mm and above, resulting in less and less quality degradation. Traditional manual inspection technology is costly and slow, and human eyes are prone to fatigue after long-term work, resulting in false inspections, and it is very easy to cause accidents where defects appear while looms and printing and dyeing equipment are still working, resulting in a decline in cloth quality. [0003] In the prior art, cloth pattern...

AI Technical Summary

Problems solved by technology

The Chinese invention patent application with the application number CN201410467061.9 discloses a cloth defect detection method and device based on adaptive orthogonal wavelet transform, which replaces manual visual inspection and the original manual selection of wavelet bases, and then conducts wavelet-based detection based on wavelet transform. The defect automatic detection method overcomes the defects of traditional manual visual detection, such as slow speed, low efficiency, false detection, and high rate of missed detection, as well as the poor detection accuracy caused by the failure to optimize the wavelet base in the original defect detection method based on wavelet transform. high problem

[0004] However, these existing solutions have high requirements for the quality of image collection, especially for cloth that is smaller than the standard cloth and needs to be printed or printed with complex patterns. Or in the printing process, unexpected mechanical slippage and displacement will greatly cause the image acquisition to be unable to be accurately aligned, and the collected images often have angles or lack of clarity, which brings difficulties to the timely detection and accurate identification of patterns and cloth defects. come more difficult

In particular, this image analysis method requires a large amount of calculation for image processing, and it is difficult to meet the time requirements for defect detection under the requirements of modern high-speed printing and printing processes.

Images