Texture Feature Extraction of RGB, HSV, YIQ and Dithered Images using GLCM, Wavelet Decomposition Techniques

Abstract

When changing the format of an image from simple RGB to HSV, YIQ and Dithered image, the characteristics of image also change. In this paper, the similar images in the above formats are retrieved using statistical and structural retrieving techniques i.e. GLCM (Gray Level Co-occurrence Matrix) and Wavelet Decomposition techniques. The best results are coming for dithered, HSV images by using GLCM technique for feature extraction and by using Wavelet decomposition; HSV images are giving the best results. While other formats also give the correct retrieval but from the accuracy point of view, HSV images are retrieved with better accuracy using both techniques.