Recognizing Human Iris by Modified Empirical Mode Decomposition
With the increasing needs in security systems, iris recognition is reliable as one important solution for biometrics-based identification systems. Empirical Mode Decomposition (EMD), a multi-resolution decomposition technique, is adaptive and appears to b
- PDF / 710,736 Bytes
- 13 Pages / 430 x 660 pts Page_size
- 3 Downloads / 164 Views
Department of Electrical and Electronic Engineering, Institute of Technology, National Defense University, Taoyuan, Taiwan, Republic of China 2 Department of Electronic Engineering, Ming Chuan University, Taoyuan, Taiwan, Republic of China {i923002, cpchang, tutm}@yahoo.com.tw
Abstract. With the increasing needs in security systems, iris recognition is reliable as one important solution for biometrics-based identification systems. Empirical Mode Decomposition (EMD), a multi-resolution decomposition technique, is adaptive and appears to be suitable for non-linear, non-stationary data analysis. This paper presents an effective approach for iris recognition using the proposed scheme of Modified Empirical Mode Decomposition (MEMD) to analyze the iris signals locally. Since MEMD is a fully data-driven method without using any pre-determined filter or wavelet function, MEMD is used as a low-pass filter to extract the iris features for iris recognition. To verify the efficacy of the proposed approach, three different similarity measures are evaluated. Experimental results show that those three metrics have achieved promising and similar performance. Therefore, the proposed method demonstrates to be feasible for iris recognition and MEMD is suitable for feature extraction. Keywords: Biometrics, iris recognition, Empirical Mode Decomposition (EMD), multi-resolution decomposition.
1 Introduction Biometrics is inherently a more reliable and capable technique to identity human's authentication by his or her own physiological or behavioral characteristics. The features used for personnel identification by current biometric applications include facial features, fingerprints, iris, palm-prints, retina, handwriting signature, DNA, gait, etc. [1], [2] and the lowest error recognition rate is achieved by iris recognition [3]. With the increasing interests, more and more researchers gave their attention into the field of iris recognition. Recently, iris recognition approaches can be roughly divided into four categories: phase-based approaches [4], zero-crossing representation [5], texture analysis [6], [7], and intensity variation analysis [8], [9]. Daugman’s algorithm [4] adopted the 2D Gabor filters to demodulate phase information of iris. Each phase structure is quantized to one of the four quadrants in the complex plane. The Hamming distance was further used to measure the 2048-bits of iris code. Boles and Boashash [5] D. Mery and L. Rueda (Eds.): PSIVT 2007, LNCS 4872, pp. 298 – 310, 2007. © Springer-Verlag Berlin Heidelberg 2007
Recognizing Human Iris by Modified Empirical Mode Decomposition
299
proposed the zero-crossing of 1D wavelet transform to represent distinct levels of a concentric circle for an iris image, and then two dissimilarity functions were used for matching. Wildes et al. [6] analyzed the iris texture using the Laplacian pyramids to combine features from four different resolutions. Normalized correlation is selected to decide whether the input image and the enrolled image belong to the same class. L. Ma e
Data Loading...
