Texture classification based on Markov modeling in wavelet feature space

One difficulty of texture analysis and classification in the past was the l ack of adequate tools to characterize textures over different scales. Recen t developments in multiresolution analysis, such as the wavelet transform, promise ways to overcome this difficulty. In this paper, we present a textu re classification methodology that is based on a stochastic modeling of tex tures in the wavelet domain. The model captures significant intrascale and interscale statistical dependencies between wavelet coefficients, which are typically disregarded by wavelet-based statistical signal processing techn iques. It provides an accurate multiscale texture representation and underl ies a highly discriminative texture classification algorithm. (C) 2000 Else vier Science B.V. All rights reserved.