Design of appropriate wavelet bases for texture discrimination

Texture analysis and discrimination are an important part of computer vision systems and active areas of research. In this research, we have analyzed different regions of Pakistan on the basis of their textural properties. We propose a design of wavelet bases using genetic optimization, which will provide excellent discrimination between the multiple texture images. Our objective function is based on maximization of distinguishability measures involving the computation of finer details subject to some wavelet constraints. In contrast to well-known orthogonal wavelet families, we have used an extra degree of freedom in the design of the wavelet function for best possible texture discrimination. In the genetic optimization process, design parameters of the wavelet are optimized according to the characteristics of the texture images under a defined set of constraints. Classification results of designed wavelets when compared with existing wavelet families show much better texture discrimination