Suitability analysis of techniques for flaw detection in textiles using texture analysis

This paper addresses issues concerning image analysis algorithms for the vi sual quality inspection of textile fabrics. An overview of a number of flaw detection techniques and analysis of their suitability fur detecting! the presence of weaving defects is presented. The Raw detection algorithms are based on a local analysis of spatial and spatial-frequency domain features, exploring both the statistical and structural texture analysis of the defe ctive image samples. The algorithms are compared using two criteria: accura cy and computational efficiency. Four methods are applied to the flaw detec tion problem, and their accuracy is assessed experimentally on a number of common textile defects in this preliminary comparison The methods are spati al grey level co-occurrence; normalised cross-correlation; texture blob det ection; and spectral approaches. Of these methods, the correlation approach appears to be the must promising candidate for a real-time, high accuracy flaw detection algorithm.