| Nowadays, the Human-Computer Interactive Interface (HCI) has become more and more smart and easy to use with the evolving trend. The future interactions will mainly concentrated on two aspects:Natural User Interface and Ubiquitous Computing. Natural user interactive system is currently a hot research area, which was believed to be the next development of the HCI. Natural user interactive system requires computers to interact man with people's natural interactive way. Thus, it could achieve more channels and methods in the HCI according to people's gesture, voice, expression and movement. The purpose of this research is to achieve rapid and convenient man-machine interaction.In this thesis, hand tracking and gesture recognition are research topics and computer vision, the HCI and identify ways have been deeply analyzed. Furthermore, this thesis points out that area limited and the skin color detection could benefit the research on design and implementation of interactive system based on gesture of computer vision. The main results are as follows:1. The thesis has studied gestures and expressions identification and recognition, more touch of intellectual input and output devices as well as the asynchronous functions of operating process. It also inspects interaction with more channels, scalable beyond the desktop application;2. It studies how to take the sequence of the hand movement, analysis the reasons of changing caused by the complex shape of the hand, and through experiment, it identifies the methods to determine the hand area by color;3. Further study, analysis and comparison with the general gesture recognition and tracking algorithm with the HCI, the thesis points out those existing methods used today still need improving. Then this article presents a solution of high speed and robust gesture interactive system based on the skin color detection and movement recognition which has been proved to be validated by experiments. |
PDF Full Text Request