Archana S. Ghotkar & Gajanan K. Kharate International Journal of Human Computer Interaction (IJHCI), Volume (3) : Issue (1) : 2012 15 Hand Segmentation Techniques to Hand Gesture Recognition for Natural Human Computer Interaction Archana S. Ghotkar archana.ghotkar@gmail.com Department of Computer Engineering Pune Institute of Computer Technology University of Pune Pune 411 043, India Gajanan K. Kharate gkkharate@yahoo.co.in Department of Electronics and Telecommunication Engineering Matoshri College of Engineering and Research Centre University of Pune Nasik 422 105, India Abstract This work is the part of vision based hand gesture recognition system for Natural Human Computer Interface. Hand tracking and segmentation are the primary steps for any hand gesture recognition system. The aim of this paper is to develop robust and efficient hand segmentation algorithm where three algorithms for hand segmentation using different color spaces with required morphological processing have were utilized. Hand tracking and segmentation algorithm (HTS) is found to be most efficient to handle the challenges of vision based system such as skin color detection, complex background removal and variable lighting condition. Noise may contain, sometime, in the segmented image due to dynamic background. An edge traversal algorithm was developed and applied on the segmented hand contour for removal of unwanted background noise. Keywords: Human Computer Interface, Hand tracking and Segmentation, Hand Gesture Recognition, Edge Traversal Algorithm. 1. INTRODUCTION Natural Human Computer Interaction (HCI) is the demand of today’s world. Survey and Sign language study shows that from various gesture communications modality, the hand gesture is the most easy and natural way of communication. Real-time vision-based hand gesture recognition is considered to be more and more feasible for Human-Computer Interaction with the help of latest advances in the field of computer vision and pattern recognition [1]. There are various applications using Hand Gesture Recognition. Zhao et al. [2] explained virtual reality system based on hand gesture. Gaussian distribution was used for building complexion model, YCbCr color space was used for segmentation purpose and Fourier descriptor as a feature vector. BP neural network was utilized for recognition and all these resulted in an improved recognition rate. High light and shadow segmentation results were, however, found to be not perfect. Guan and Zheng [3] introduced a novel approach to pointing gesture recognition based on binocular stereo vision, in which user needs to wear special clothes or markers and was found to be suitable for both left and right handed users. Freeman and Weissman [4] explained television control application by hand gesture. Here, the user uses only one gesture: the open hand, facing the camera for controlling the television. Sepehri et al. [5] proposed algorithms and applications for using hand as an interface device in virtual and physical spaces. In Real-time Vision based Hand Gesture recognition system, hand tracking and segmentation are most important and challenging steps towards gesture recognition. Uncontrolled environment,