This paper proposes an innovative contactless palm print and knuckle print recognition system. We present a novel palm print and knuckle print tracking approach to automatically detect and capture these features from low-resolution video stream. Besides, we introduce a simple yet robust directional coding technique to encode the palm print feature in bit string representation. The bit string representation offers speedy template matching and enables more effective template storage and retrieval. Apart from that, we present a new scheme to extract knuckle print feature via ridgelet transform. Our method is different from the others in the sense that we do not resize the knuckle print images to standard size. The scores output by the palm print and knuckle print experts are fused using Support Vector Machine. The fusion of these features yields promising result for practical implementation.
All Science Journal Classification (ASJC) codes
- Signal Processing
- Computer Vision and Pattern Recognition
- Artificial Intelligence