Abstract
The M-band wavelet decomposition, which is a direct generalization of the standard 2-band wavelet decomposition, has been applied to the problem of an unsupervised segmentation of two texture systems. Standard wavelets are not suitable for the analysis of high frequency signals with relatively narrow bandwidth. Therefore, in the present work we were motivated to use the decomposition scheme based on M-band wavelets that yield improved segmentation accuracies. Unlike the standard wavelet decomposition, which gives a logarithmic frequency resolution, the M-band decomposition gives a mixture of a logarithmic and linear frequency resolution. Further motivation to use M-band wavelet filter for our texture analysis scheme is because, this decomposition yields a large number of subbands, which is required for good quality segmentation.