OPTIMIZED TECHNIQUES AND TECHNOLOGIES FOR TEXTILE FABRIC DEFECT DETECTION

Abstract

Defect detection in textile fabrics is critical for ensuring product quality in a highly competitive market. Traditional manual inspection methods are error-prone and inefficient, leading to increased costs and reduced productivity. This study explores advanced computational techniques and technologies, including computer vision, neural networks, and IoT-integrated systems, to automate fabric defect detection. Utilizing a systematic approach, this paper investigates methods like NIR spectroscopy, kernel methods, and neural networks to identify and classify textile defects effectively. The proposed solutions demonstrate high accuracy, reduced inspection times, and significant cost efficiency. These findings support the transformative potential of automation in the textile industry, ensuring sustainability and operational excellence.