Rough Set-Based Decision Tool for Classification of Cotton Yarn Neps

Abstract

Neps in cotton yarn and fabric are considered as blemishes which can severely downgrade the product and economically affect the textile industry. Identification of neps in cotton yarn is a prerequisite to control their generation in the yarn formation process. Hence, a real-time nep identification technique would lead to more effective control and reduction in nep levels in cotton. In recent years, rough set theory has evolved as one of the most promising classification techniques. One of the cardinal uses of rough set theory is its application for rule generation. Our approach focuses on the classification of seed coat neps and fibrous neps using the effective decision rules envisaged by rough set theory. In this work, 60 images were captured and processed in rough set technique to classify neps in cotton yarn. The validation results ascertain that 11 out of 12 testing data are correctly predicted by the rough set technique. The framed decision rules provide an insight about the classification tool which ensures that the prediction accuracy of the tool can be raised further by framing more robust decision rules with the help of large training dataset. Thus, this technique is potent to get recognition from the modern textile industry as an automated neps classification technique.