Texture Classification Using Cyclic Spectral Function
Mehdi Chehel Amirani, Ali Asghar Beheshti Shirazi · 2008
In this paper, a new feature extraction technique for texture classification is proposed. Features are energy and standard deviation of spectral correlation function (SCF) of signals got from image at different regions of bifrequency plane. This scheme shows high performance in the classification of Brodatz texture images. Experimental results indicate that the proposed method improves correct classification rate in comparing with traditional discrete wavelet transform approaches.