Fusion hand gesture segmentation and extraction based on CMOS sensor and 3D sensor

Abstract

Gesture recognition is one of the most promising subjects in the field of computer vision and artificial intelligence; the development of it will have a profound influence on the research of robot control and Human-Machine Interface (HMI) and image segmentation is a key step in image recognition. This paper based on Kinect sensor developed by Microsoft gives an introduction on its hardware structure and the measuring method of depth camera. At the same time we will collect the colour hand gesture images as samples, to introduce the traditional skin colour image segmentation method in HSV and YCbCr colour space. Finally, a new 3D image hand gesture segmentation method based on depth information will be proposed.