Our Security-Conscious Society

2967 Words6 Pages

1 Introduction

In today’s security-conscious society, a reliable, robust and convenient approach for automated user authentication is becoming a strong requirement. Since September 2001 (i.e. World Trade Center blast), public awareness about the need for security has been increased considerably and lead to a massive rise in demand for the personal authentication systems (Wang et al., 2005). Biometrics plays a major role in today’s security applications. A biometric system is essentially a pattern recognition system that recognizes a person based on his/her physiological or behavioral characteristic that the person possesses (Prabhakar et al., 2003). During the past few decades, researches have been carried out on utilizing various biometrics for personal authentication such as fingerprint, face, iris, palm, signature, voice, etc (Bolle et al., 2003). Amongst these biometrics, recently vein pattern has gained more and more research attention for personal authentication (Kang., 2012; Mirmohamadsadeghi et al., 2011; Miura et al., 2004, & 2007; Toh et al., 2006; Wang et al., 2006; Wang et al., 2012; Watanabe et al., 2005; Yanagawa et al., 2007; Zhang et al., 2007) due to its distinct characteristics like ease of feature extraction, spoofing resistant, high accuracy and liveness detection etc.

A typical vein pattern biometric system consists of five individual processing stages: Image Acquisition, ROI extraction, Image enhancement and Vein Pattern Segmentation, Feature Extraction and Matching, as shown in Figure 1. The acquisition of hand veins is generally done using infrared (IR) imaging. The IR imaging for veins is of two types namely near infrared (NIR) in the range of 0.75μm to 2μm and far infrared (FIR) in...

... middle of paper ...

... et al. (2005)

48 Users

FAR = 0%, FRR = 0.9%

Wang et al. (2008)

47 Users

FAR = 0%, FRR = 0%

Kumar et al. (2009)

100 Users

FAR = 1.14%, FRR = 1.14%

This Paper

60 Users

FAR = 1.26%, FRR = 1.21%

7 Conclusions

This paper proposed an efficient and new approach for vein pattern authentication using connected neighbors of minutiae. The dynamic ROI extraction algorithm retrieves maximum possible features from the hand which is not possible with static ROI algorithm. The use of minutiae neighbor information in the proposed algorithm reduced the false

matching of minutiae with imposter images (i.e. reduces FAR) and improved the performance (EER=1.2%) of the authentication system. The experimental results using minutiae neighbor information are promising and suggested as better alternative for human authentication.

Open Document