An Automatic and Detachable Fabric Defect Detection and Grading Device for Apparel Industry

This project will develop an automatic fabric defect detection and grading device using various artificial intelligence (AI) techniques for automating the current manual fabric inspection process in apparel industry. The proposed device, which can be integrated with the currently used fabric inspection machines, can automatically locate the defects and measure the defect sizes on woven fabrics with solid and strip pattern, as well as calculate the graded points according to current inspection standards.

In apparel manufacturing process, fabric inspection is a vital step to ensure the quality of fabric and locate defects before spreading, cutting and sewing of making up an apparel product. In the market, although a few number of automatic fabric inspection machines exist, they can only be used in textile manufacturing process with the limitation of inspecting single type of fabric, e.g. greige fabric, and the price is extremely expensive. Almost all apparel manufacturing plants are still relying on workers’ visual inspection on the moving fabric rolls loaded on the fabric inspection machine. However workers are subject to fatigue or boredom and thus unreliable and inconsistent inspection results are often generated. As a result, fabric inspection is highly prone to errors and it allows defects to go undetected.