Image Abnormal Region Recognition with Fuzzy Clustering Based on Multi-Characteristic Variable Window

This paper proposes a new pattern recognition algorithm with fuzzy clustering based on multi-characteristic variable window. This algorithm can recognize the abnormal region for dynamic image. The image is divided into some windows by this algorithm, and multi-characteristic vector of each window is constructed. The weight factor vector is introduced so that each window characteristic may be primary or secondary according to different image feature. The window coefficient is introduced so that the recognition speed and precision can be adjusted. In this paper, the objective function, membership function and clustering center calculation function of fuzzy clustering algorithm with weight factor and window coefficient is proposed. At last, this paper takes example for fabric defects recognition with this algorithm. Experimental results show that this algorithm can recognize more categories of image abnormal regions with high-accuracy, high-speed, no-training and extensive application.