Perceptual Resemblance of Facial Images: a near Set Approach

In this paper, we introduce a near set approach to image analysis. Near sets result from generalization of rough set theory. One set X is near another set Y to the extent that the description of at least one of the objects in X matches the description of at least one of the objects in Y. Near set Evaluation And Recognition (NEAR) system is used to measure the degree of resemblance between facial images. The goal of the NEAR system is to extract perceptual information from images using near set theory, which provides a framework for measuring the perceptual nearness of objects. In this work, we have used images from Japanese Female Facial Expression (JAFFE) database. The images were first converted into Local Binary Patterns (LBP) images and then divided into non-overlapping blocks. The degree of nearness of histograms of all the blocks of one image is measured with the corresponding blocks of another image by using NEAR system.