Abstract—Aim of this paper is to develop a texture recognition system for browsing and retrieval of image data. Many features have been proposed to precisely define the natural texture properties. Tamura proposed six features. Those features are coarseness, contrast, directionality, line-likeness, regularity and roughness. Haralick extract some features from gray level co-occurrence matrix (GLCM). Gabor features and wavelet features are widely used in image retrieval system and gives good result. In this paper combination of gray-level co-occurrence matrix, Daubechies filters and rotated wavelet filters are used to get a high quality feature set. A new algorithm is proposed to get rotated wavelet filter from Daubecheis wavelet coefficients. Experimental results demonstrate that the propose method is very efficient and superior to some other existing method.

Index Terms—Content based image retrieval, rotated wavelet filter, texture retrieval, wavelet.

Dipankar Hazra is a Senior Lecturer in the department of Computer Science & Engineering, Dr..B.C.Roy Engineering College, Fuljhore,Durgapur, West Bengal, India.(e-mail: dipankar1998@rediffmail.com).

Cite: Dipankar Hazra, "Texture Recognition with combined GLCM, Wavelet and Rotated Wavelet Features," International Journal of Computer and Electrical Engineering vol. 3, no. 1, pp. 146-150, 2011.