Article Preview
Top1. Introduction
Texture analysis is very much important in a wide range of image processing applications from medical imaging, remote sensing to defect detection. It has been an active research area in the field of computer vision for more than two decades. It is found to be very hard because one is usually not aware of the number and type of texture classes that an image may contain. Many techniques have been proposed in the literature for texture analysis (Haralick, 1973; He, 1990; Chang, 1993; Yoshimura, 1997; Van de Wouwer, 1999; Vese, 2003; Targhi, 2006; Liao, 2009; Li, 2010; Chakraborty, 2012; Karthikeyan, 2012; Zingman, 2013). The ultimate aim is to detect, characterize and represent texture feature in images. But, most of the texture analysis techniques end up with texture segmentation (Chen, 1995; Ojala, 2001; Li C. T., 2003; Liu, 2006; Ranjan, 2014). Only a few techniques perform texture detection, characterization and representation which lead to texture classification (Haralick, 1973; Chang, 1993; Liao, 2009; Chakraborty, 2012; Karthikeyan, 2012) in images. Thus, there is a need for a suitable scheme for texture characterization and representation in images.