FAST system approach to discriminate the characterized generic hand of fabrics

Abstract

An attempt has been made to discriminate different characterized generic hands of cotton, linen, wool, and silk woven fabrics using discriminant analysis and neural network method. Ten physical properties based on the FAST system have been selected for the analysis. It is observed that the cotton, linen, wool, and silk groups of fabric can be characterized and discriminated by discriminant analysis and neural network method with 91.67 % and 98.33 % classified accuracy. Model test results show that the cotton type polyester, linen-textured rayon, wool type polyester, and silk-like polyester fabrics can be classified accurately by the neural network method. The confusion coefficient is found to be 100%.