At present, the work to analyze fabric structure still depends on artificial visual measurement, which is easily influenced by personal sight, mood, mental state as well as light condition. With the development of image processing technology and artificial intelligence, automatic analysis on fabric structure as a replacement of manual labor is of great possibility. In this study, features of fabric-image have been extracted by GLCM (Gray Level Co-occurrence Matrix). These features were analyzed by employing a three layer BP neural network. Three kinds of fabric structures such as plain, twill and satin was verified and the accurate recognition rate is very high to 93.45%.