Texture Analysis and Classification Based on Fuzzy Triangular Greylevel Pattern and Run-Length Features

Abstract

Your Texture analysis is one of the most important techniques used in the analysis and interpretation of images, consisting of repetition or quasi repetition of some fundamental image elements. The present paper derived Fuzzy Triangular Greylevel Pattern (FTGP) to overcome the disadvantages of LBP and other local approaches. The FTGP is a 2 x 2 matrix that is derived from a 3 x 3 neighborhood matrix. The proposed FTGP scheme reduces the overall dimension of the image while preserving the significant attributes, primitives, and properties of the local texture. From each 3 x 3 matrix a Local Grey level Matrix (LGM) is formed by subtracting local neighborhoods by the gray value of its center. The 2 x 2 FTGP is generated from LGM by taking the average value of the Triangular Neighbor Pixels (TNP) of the 3 x 3 LGM. A fuzzy logic is applied to convert the Triangular Neighborhood Matrix (TNM) in to fuzzy patterns with 5 values {0, 1, 2, 3 and 4} instead of patterns of LBP which has two values {0, 1}. On these fuzzy patterns a set of Run Length features are evaluated for an efficient classification. The proposed method is experimented with wide variety of textures, and exhibited with a high classification rate. The proposed FTGP with run length features shown its supremacy and efficacy over the various existing methods in classification of textures.