WiseEye Intelligent Fabric Defect Detection System Developed by Hong Kong Polytechnic University

The hong kong polytechnic University (PolyU) recently developed an intelligent cloth defect detection system, known as “WiseEye”, that leverages technologies as well as artificial intelligence (AI) and Deep Learning within the method of quality control (QC) in textile industry.

Supported by AI-based machine-vision technology, the novel “WiseEye” will be put in in an exceedingly weaving machine to assist cloth makers to find defects within the production method. Through the automated scrutiny system, the assembly line manager will find the defects, therefore serving to them to spot the reason behind the issues and fix them.

“WiseEye” is developed by the Textile and apparel computer science (TAAI) analysis Team, that is spearheaded by academic Calvin Wong, Cheng Yik decorated academician in Fashion of Institute of Textiles and consumer goods, PolyU.

The analysis team has applied huge knowledge and Deep Learning technologies in “WiseEye”. By inputting knowledge of thousands yards of materials into the system, the team has trained “WiseEye” to find concerning forty common cloth defects with exceptionally high accuracy resolution of up to zero.1 mm/pixel.

At the instant, “WiseEye” will be applied to most kinds of materials with totally different weaving structures and solid colors. The analysis team plans to additional train and extend the system to find defects in materials with tougher patterns, like sophisticated strip and check patterns. the final word target is to hide all common sorts of cloth in 5 years’ time.

Prof Wong and also the TAAI analysis team are conducting elementary and applied analysis on AI, pc vision and machine learning, specifically for the style and textile trade since 2012. The team has earlier introduced the first-of-its-kind “FashionAI Dataset” that integrates fashion and machine learning for systematic analysis of fashion pictures through the utilization of AI.

Leave a Reply

Your email address will not be published. Required fields are marked *