Texture Image Analysis for Larger Lattice Structure using Orthogonal Polynomial framework


  • L Ganesan Ramco Institute of Technology, Rajapalayam, Tamil Nadu, India
  • C Umarani A C Govt. College of Engineering and Technology
  • M Kaliappan Ramco Institute of Technology, Rajapalayam, Tamil Nadu
  • S Vimal National Engineering College
  • Seifedine Kadry Department of Applied Data Science, Noroff Universty College
  • Yunyoung Nam Soonchunhyang University; Asan; Republic of Korea




Texture Representation, Polynomials, Statistical Tests, Local and Global descriptors, Standard Texture Images


An Orthogonal Polynomial Framework using 3 x 3 mathematical model has been proposed and attempted for the textureanalysis by L.Ganesan and P.Bhattacharyya during 1990. They proposed this frame work which was unified to address both edgeand texture detection. Subsequently, this work has been extended for different applications by them and by different authors overa period of time. Now the Orthogonal Polynomial Framework has been shown effective for larger grid size of (5 x 5) or (7 x 7) orhigher, to analyze textured surfaces. The image region (5 x 5) under consideration is evaluated to be textured or untextured usinga statistical approach. Once the image region is concluded to be textured, it is proposed to be described by a local descriptor,called pro5num, computed by a simple coding scheme on the individual pixels based on their computed significant variances. Thehistogram of all the pro5nums computed over the entire image, called pro5spectrum, is considered to be the global descriptor.The novelty of this scheme is that it can be used for discriminating the region under consideration is micro or macro texture,based on the range of values in the global descriptor. This method works fine for many standard texture images. The works usingthe proposed descriptors for many texture analysis problems with (5 x5) including higher grid size and applications are underprogress