Automatic location and extraction of woven fabric blocks based on Gaussian blur and maximization thought

The automatic identification of the woven fabric structures is usually based on the study of the fabric structure blocks. Generally, in the identification process of a textile fabric image which contains several different woven fabric structures, blocks of different woven fabric structures need to be located and extracted manually. This defect extremely limits the development of the fabric image´s automatic identification. A set of method about the automatic location and extraction of woven fabric blocks from the whole fabric image is researched and implemented in this paper. This method consists of Gaussian blur, K-means clustering with human-computer interaction and the thought of maximization. It can automatically locate and extract several different woven fabric blocks in a textile fabric image.