An efficient ear recognition system using DWT & BLPOC

This paper proposes an efficient ear recognition algorithm using band limited phase only correlation (BLPOC) based local block matching. Phase information over the frequency field of an image has been acquired by discrete wavelet transform (DWT). The DWT may decompose an ear image into a set of basis function. The BLPOC is strong to resilient noise and rescaling variables may efficiently upgrade the performance of an ear recognition system for ear biometric applications. The phase matching technique has been adapted between the registered image and query image. The proposed algorithm works for a local image deformation in local block matching, respectively. Experimental analysis has been done using the IIT Delhi Ear database which exhibit the effectual recognition.