Factor Analysis Assisted Classification of Ear Images Based on GLCM Features

Abstract It is well known fact the Gray Level co-occurrence matrix (GLCM) is immensely used to extract second order statistical textual features of an image. These features have been variously used by researchers. In this paper, GLCM features of 800 right ear images have been used. The case in point is ear biometrics; the 22 features extracted from the images were subjected to factor analysis for delineating the most predominant textural attributes and the factors involved with them. The result shows that three factors qualified by Kaiser’s criterion were used to develop a person identification system, which showed 97% of recognition accuracy.