An online fabric defect detection and alarm system based on machine vision

Abstract

The invention provides an on-line fabric flaw detection and alarming system based on machine vision. According to the invention, a high-resolution and high-speed camera is used for acquiring a fabric on-line video; the obtained on-line video is subjected to video AD digital processing, and is collected and pre-processed by using a video processor; the processed video is outputted to a video terminal through the video processor; the video is subjected to intelligent video processing by using a main processor; flaw detection is carried out with a multi-mode self-adaptive flaw detection method; if a flaw is detected, the main processor notifies a control device for stopping the machine from production, and alarms; in case of false detection, with a human-machine interaction interface, the production device is notified of continuing production; otherwise, production is stopped such that the amount of fabric with flaws can be reduced. With the on-line fabric flaw detection and alarming system based on machine vision, an on-line video detection method is introduced. With the adopted multi-mode self-adaptive flaw detection method, flaw identification efficiency is improved, and defect rate is reduced.

Description

technical field [0001] The invention belongs to the field of detection systems, and in particular relates to an online fabric defect detection and alarm system based on machine vision. Background technique [0002] In the textile industry, the defect detection of fabric products plays an important role. Improving the accuracy and efficiency of fabric defect detection will contribute to the customer satisfaction of the entire product and the overall benefit of the product. Generally speaking, after the fabric becomes a defective product due to fabric defects, its value will be reduced by 40% to 50%, or even lower. At present, there are mainly two methods for fabric defect detection: [0003] Manual detection mainly relies on human vision to detect defects. This method mainly consumes a lot of manpower, the efficiency of defect detection is low, and there is a certain rate of missed detection due to human visual fatigue and other reasons. [0004] The automatic detection me...

AI Technical Summary

Problems solved by technology

[0005] But at present, the above two methods are mainly used for the detection and qualitative of the produced fabrics, that is, they can only show whether there are defects in the fabrics, and cannot interfere with the fabric production.

Images