14 November 2016 Person identification by using 3D palmprint data
Author Affiliations +
Abstract
Person identification based on biometrics is drawing more and more attentions in identity and information safety. This paper presents a biometric system to identify person using 3D palmprint data, including a non-contact system capturing 3D palmprint quickly and a method identifying 3D palmprint fast. In order to reduce the effect of slight shaking of palm on the data accuracy, a DLP (Digital Light Processing) projector is utilized to trigger a CCD camera based on structured-light and triangulation measurement and 3D palmprint data could be gathered within 1 second. Using the obtained database and the PolyU 3D palmprint database, feature extraction and matching method is presented based on MCI (Mean Curvature Image), Gabor filter and binary code list. Experimental results show that the proposed method can identify a person within 240 ms in the case of 4000 samples. Compared with the traditional 3D palmprint recognition methods, the proposed method has high accuracy, low EER (Equal Error Rate), small storage space, and fast identification speed.
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xuefei Bai, Shujun Huang, Nan Gao, Zonghua Zhang, "Person identification by using 3D palmprint data", Proc. SPIE 10025, Advanced Sensor Systems and Applications VII, 100250Q (14 November 2016); doi: 10.1117/12.2246195; https://doi.org/10.1117/12.2246195
PROCEEDINGS
8 PAGES


SHARE
Back to Top