A Novel Approach for Iris Recognition Using Local Edge Patterns
This paper presents an effective approach for iris recognition by analyzing the iris patterns. We propose an iris classification method that divides the normalized iris image into several regions to avoid the iris image with several noise factors (eyelids
- PDF / 451,090 Bytes
- 10 Pages / 430 x 660 pts Page_size
- 17 Downloads / 207 Views
Department of Electrical and Electronic Engineering, Institute of Technology, National Defense University, Taoyuan, Taiwan 2 Department of Electronic Engineering, Ming Chuan University, Taoyuan, Taiwan {g980301,cpchang,tutm}@ccit.edu.tw
Abstract. This paper presents an effective approach for iris recognition by analyzing the iris patterns. We propose an iris classification method that divides the normalized iris image into several regions to avoid the iris image with several noise factors (eyelids and eyelashes) and reduce the error rates. In every region, effective features are extracted by the proposed method of local edge pattern (LEP) for edge and corner detection. Feature vectors are linearly combined into a two dimensional matrix that represents every iris image for further recognition. Then 2D linear discriminant analysis (2DLDA) is used to identify the person. We use two public and freely available iris image databases for evaluation, organized in training and test sets respectively. Experimental results show that the recognition rate of the two iris image databases have achieved similar performance more than 98% and the proposed method has an encouraging performance and robustness.
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 lower error recognition rate is achieved by iris recognition [3]. In general, iris recognition approaches can be roughly divided into four main categories: phase-based approaches [4], zero-crossing representation [5], intensity variation analysis based methods [6], [7] and texture analysis [8], [9]. Daugman’s algorithm [4] adopted the 2D Gabor filters to demodulate phase information of iris. Boles and Boashash [5] proposed the zero-crossing of 1D wavelet transform to represent distinct levels of a concentric circle for an iris image. L. Ma et al. [6], [7] proposed a local intensity variation analysis-based method and adopted the GaussianHermite moments and dyadic wavelet to characterize the iris image for recognition. Wildes et al. [8] analyzed the iris texture using the Laplacian pyramids to combine features from four different resolutions. More recently, Tisse et al. [9] constructed the analytic image to demodulate the iris texture. However, there are still limited capabilities in recognizing the identity of person accurately and efficiently, and much space is needed for improving performance in a practical viewpoint. The main G. Bebis et al. (Eds.): ISVC 2007, Part II, LNCS 4842, pp. 479–488, 2007. © Springer-Verlag Berlin Heidelberg 2007
480
J.-C. Lee et al.
difficulty of iris recognition is that it is hard to find apparent feature points in the iris image and to keep high classification rate in an efficient way. In this paper, we propose a robust
Data Loading...
