TEXTURE CLASSIFICATION USING RIDGELET TRANSFORM

Abstract

Texture classification has long been an important research topic in image processing. Classification of images based on textural features is one of the fast growing technique in CBIR. Textural features corresponding to human visual perception are very useful for optimum feature selection and texture analyzer design. Textural properties are defined for a region or sub-image, not for a point. Pairs of regions are often used in psychological experiments and in fundamental studies of computational measurements. In this situation, if two adjacent regions differ only in scale, contrast, orientation, or shape of repetitive elements, we may be able to perceive two different textures and detect the texture boundary between these two regions. Tamura features classifies images based on six textural features namely coarseness, contrast, directionality, line-likeness, Regularity and Roughness. It is required to compute every feature separately through this approach. Computing each feature separately is a difficult task. Now a days classification based on wavelet transform is being very popular. Wavelet transform method is simple and faster method than that of Tamura features. In wavelet transform method the image is decomposed in to sub bands. After decomposition feature vectors can be constructed using the mean and standard deviation of the energy distribution of each sub-band at each level. Wavelets are very effective in representing objects with isolated point singularities, but failed to represent line singularities. Recently, ridgelet transform which deal effectively with line singularities in 2-D is introduced. It allows representing edges and other singularities along lines in a more efficient way. With this objective, in the present work, ridgelet transform to represent texture property of an image is presented. Features have to derive from the sub-bands of the ridgelet decomposition and are used for classification of a data set containing 1000 texture images.