A DSP-Based Finger Vein Authentication System

摘要：

A DSP-based design method of finger vein recognition was proposed, and a finger vein authentication system FV-1 was developed. In this system, finger vein images were captured with infrared, and normalized through rotation correction, then they were segmented and thinned to wipe off useless segments. Finally, features on vein points were obtained and two finger vein images were matched hi view of a feature distance. AD these processes were completed with the high performance DSP called DM6437 of TI Corporation. The system is small and the recognition rate is good with the high matching speed.

关键词： Biometric authentication Finger vein DSP Rotation correction Feature extraction

作者: Xiao Sun Chun-Yi Lin Ming-Zhong li Hai-Wu Lin Qing-Wu Chen

作者单位: Sun Yat-Sen University, Guangzhou, Guangdong,,510080, China South China University of Technology, Guangzhou, Guangdong, 510640, China Xinsitong Communication Technology Co. Ltd, Guangzhou, Guangdong, 510640, China

会议类型: 国际会议

会议名称: 2011 Fourth International Conference on Intelligent Computation Technology and Automation(2011年第四届智能计算技术与自动化国际会议 ICICTA 2011)

会议地点: 深圳

会议语种:英文

页码: 333-336

在线出版日期: 2011-03-28（万方平台首次上网日期，不代表论文的发表时间）