Abstract
Hand gesture recognition has become a hot topic of research in recent years due to its wide applicability in the fields of human machine interaction, robotics, virtual reality, medical therapy and sign language recognition etc. This paper presents a novel method for static hand gesture (hand posture) recognition using visual features. The image features are extracted using spatial histogram coding of nonsubsampled contourlet transform coefficients, and the dimensionality of feature values are further reduced by principal component analysis. The classification results on the publicly available NUS (National University of Singapore) hand posture dataset using support vector machine gives promising results when comparing with a recent approach on the same dataset.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have