Hand posture recognition using HOW and homogenous kernel

Abstract

In this paper, we present a static hand gestures recognition system of Arabic Sign Language alphabets. The proposed method uses Histogram Of visual Words (HOW) descriptor and Support Vector Machine (SVM). First, the images of static hand gestures are converted into HOW features and grouped using k-means clustering to create histograms. Then they are converted from non linear space into linear space using Chi-squared kernel. The result is fed into One-vs-All SVM classifier to build signs models. Training and test stages of this technique are implemented on hand postures images using cluttered backgrounds for different lighting conditions, scales and rotations. The proposed method shows a satisfactory recognition rate and achieves good real-time performance regardless of the image resolution.

Keywords: arabic sign language - static hand gesture – how - one-vs-all support vector machine (svm)