Texture analysis using 2D wavelet transform: theory and applications

In this paper a survey of recently proposed methods for the analysis of textured images is given. These new methods are based on multiresolution properties of the two-dimensional wavelet transform, which is used to extract the features needed to discriminate textures. A special emphasis is given to the use of nonseparable multiresolution systems. The possible application areas of new methods are also summarized. A large bibliography is also provided for further reading.