CHARACTERIZATION OF VISUALLY SIMILAR DIFFUSE DISEASES FROM B-SCAN LIVER IMAGES USING NONSEPARABLE WAVELET TRANSFORM

This paper describes a new approach for texture characterization, base d on nonseparable wavelet decomposition, and its application for the d iscrimination of visually similar diffuse diseases of liver. The propo sed feature-extraction algorithm applies nonseparable quincunx wavelet transform and uses energies of the transformed regions to characteriz e textures. Classification experiments on a set of three different tis sue types show that the scale/frequency approach, particularly one bas ed on the nonseparable wavelet transform, could be a reliable method f or a texture characterization and analysis of B-scan liver images. Com parison between the quincunx and the traditional wavelet decomposition suggests that the quincunx transform is more appropriate for characte rization of noisy data, and practical applications, requiring descript ion with lower rotational sensitivity.