Hand Posture Segmentation, Recognition and Application for Human-Robot Interaction

Abstract

Vision-based hand gesture recognition provide a more nature and powerful way for humancomputer interaction. In the chapter, we present some new approaches for hand image segmentation, 2D hand posture recognition and 3D hand posture reconstruction. We segment hand images using the color segmentation approach which is based on the RCE neural network. Then we extract topological features of the hand from the binary image of the segmented hand region, and recognize 2D hand postures base on the analysis of these features. We also propose to use the stereo vision and 3D reconstruction techniques to recover 3D hand postures and present a new method to estimate the fundamental matrix from uncalibrated stereo hand images in this chapter. A human-robot interaction system has been developed to demonstrate the application of our hand posture recognition approaches.