Ear Recognition by using Least Mean Square Method

Biometrics is a proper way of identifying human beings based on certain physiological characteristics and behavioral characteristics. Some of the physiological characteristics are fingerprint, face, DNA and iris. Some of the useful behavioral traits are gait and voice. But ear biometrics has recently emerged as a new area in biometrics. A great possibility for use of human ears for identification exists but no automated ear biometrics system has been used yet. Automatic analysis in biometrics reduces time and helps us to accomplish tasks in a quicker and efficient manner. In the present paper , an ear recognition based on least mean square method is proposed. That is the mean square method is based on the approximate curve between the intensity and distance from the center of the ear. So overall work is broadly divided in to three parts. First segment the ear i.e. ROI(region of interest) . Second apply the method of least square and finally identify the ear by minimum average distance matching. Experimental results show that the proposed algorithm achieve better result from other 2D ear recognition techniques.