Finger estimation method for hand gesture recognition

Abstract

This paper proposes a method for estimating the number of finger from sequences of image frames based on Higher-order Local Auto-Correlation (HLAC) features and multiple regression analysis(MRA). This method is based on a low computation that enables fast and automatic finger extraction for hand gesture recognition in real time. The hand palm region is detected using background subtraction method. Then the palm and fingers are segmented. HLAC features are extracted from image, and multiple regression approach is adopted to estimate number of fingers. Furthermore, in order to improve the performance of the proposed method, we have also proposed a modified version of multiple regression method, and conduct comparative experiment with a normal multiple regression analysis method. The experimental results were analysed, and show that the regularized MRA slightly improves the performance of MRA by stabilizing the results of the estimation.