Textile woven fabric recognition using Fourier image analysis techniques: Part II texture analysis for crossed-states detection

In this paper, the problem of crossed-states detection is investigated and solved. The identification of the weave types in the fabric is realized by analyzing the texture information in the extracted crossed-points. This analysis is termed as 'crossed-states detection'. Finally, to evaluate visually our proposed method, it is applied to plain woven fabric without skewness as well as with skewness. The obtained results demonstrate the ability to solve such crossed-states detection problems. This non-destructive method will be a useful tool that can decrease the human intervention in analysing fabric weave types, especially in the context of ancient textiles.