Development of Human Identification System Based on Simple Finger-Vein Pattern-Matching Method for Embedded Environments

Abstract

Biometric techniques for authentication using body parts such as a fingerprint, face, iris, voice, and finger-vein have become increasingly important in the personal security field, including door access control, finance security, and electronic passport. Finger-vein images can be captured under various conditions, such as different temperatures and illumination, and noise in the acquisition camera. Difficulties in recognizing finger-vein images include the use of a complex algorithm for noise reduction, image re-construction, and rotation invariance in the pattern-matching algorithm. In this paper, we use a compact CMOS camera with a penetrating infrared LED light source. In addition, we suggest a simple pattern matching method to reduce the calculation time for embedded environments. The experimental results show that our simple system has good results in terms of speed and accuracy for personal identification.