Hand Segmentation Based on Improved Gaussian Mixture Model

Abstract

In the process of human computer interaction, hand tracking and hand gesture recognition are of great importance. Hand segmentation is the first step of hand gesture recognition. Among several common methods, the background subtraction method is chosen for detecting moving hands. An improved Gaussian mixture model is used to establish the background model. In order to adapt to a changing scene, parameters of the background model are updated in real time by implementing an on-line K-means approximation. Then, the moving hands are segmented by the preset threshold. Experimental results demonstrate that the proposed method can segment hand region in a complex background. The proposed method can be used in the fields of human computer interaction and augmented reality.