상세정보
학술지
Multimodal Biometric Authentication based on the Fusion of Finger Vein and Finger Geometry
- 저자
- 강병준, 박강령
- 발행일
- 200909
- 출처
- Optical Engineering, v.48 no.9, pp.1-3
- ISSN
- 0091-3286
- 출판사
- SPIE
- 초록
- We propose a new multimodal biometric recognition based on the fusion of finger vein and finger geometry. This research shows three novelties compared to previous works. First, this is the first approach to combine the finger vein and finger geometry information at the same time. Second, the proposed method includes a new finger geometry recognition based on the sequential deviation values of finger thickness extracted from a single finger. Third, we integrate finger vein and finger geometry by a score-level fusion method based on a support vector machine. Results show that recognition accuracy is significantly enhanced using the proposed method.© 2009 Society of Photo-Optical Instrumentation Engineers.
- KSP 제안 키워드
- Geometry information, Multimodal biometric authentication, Multimodal biometric recognition, Optical instrumentation, Recognition Accuracy, Support VectorMachine(SVM), finger vein, fusion method, score-level fusion