Textured image segmentation using autoregressive model and artificial neural network

This paper presents a region growing technique for texture segmentation. The technique is implemented by comparing local region properties, which are represented by a 2-D autoregressive (AR) model in a hierarchical manner. It is able to grow all regions in a textured image simultaneously starting from initially decided internal regions until smooth boundaries are formed between all adjacent regions. A multilayer neural network is used in the local region identification procedure to establish a 2-D AR model for a textured region and to compute the region properties for the segmentation. The performance of the segmentation technique is shown by experiments on natural textured images