An intelligent system for supporting design of fashion oriented personalized fabric products

Abstract

An intelligent system for supporting the design of fashion oriented personalized fabric samples has been proposed. Based on fuzzy logic and semantic network, it permits to model the relationship between fabric parameters and fashion design elements via fashion images. This system can effectively help textile producers and designers to determine parameters of new fabrics to be produced according to fashion requirements of garments and predict garment fashion styles for given fabrics. A sensory evaluation on fabric hand has been used to determine fabric parameters of a collection of samples. Principal component analysis has been used to reduce the complexity of the model. A set of fashion images has been selected in order to extract abstract fashion design elements and identify relationship between fabric samples and fashion design elements. This system is helpful for textile companies to realize mass customization, i.e. design and production of personalized products with very low costs