Research Of Augmented Reality Man-machine Interaction System

Augmented Reality (AR) is a new area of research extending from Virtual Reality (VR) which is a combination of virtual environment and real environment. It is a technology that adds virtual objects generated by the computer to the real world to construct the effect of the actual situation. It uses the known information in the real world to track and register, then integrates the virtual objects into the real world, displays the final image by means of display device at last. AR will make us to have a better understanding of our real world.AR has its wide application prospect in medicine, machinery manufacturing, military, education and business. Many users have PC and webcam. This meets the basic requirement for desk-top AR system. The registration based on markers has a high accuracy and requires no complex track hardware. The mark is easy to track. So the research of AR system based on markers is somewhat practical and significant.The research goal is to achieve an identity-based augmented reality human-computer interaction system. We use ARToolKit open source software library to develop identity-based Augmented Reality interaction system. ARToolKit marker for the marker test block, ARToolkit tools for analysis and on this basis, aspects of the interaction of its development, so that users can have some interaction with gestures and face. The main task and contribution of this paper are as follows:1. The paper introduce the main applications of augmented reality and analyze three key technologies in the development of AR:display technology, tracking up technology and human-computer interaction technique. Each technology has its own advantages and disadvantages.2. The paper deduce the mathematical principles used in the identity-based technology, select ARToolKit as study target which is the most commonly used tools, and introduce its building blocks, analysis the implementation process of the registration module carefully. Then we focus on the identification and tracking template research and have a detailed description of design in several modules, such as image segmentation, connected component analysis, the quadrilateral selection based on the linear fitting, regional standardization and template recognition. Finally, we give the experimental results. 3. The paper design and implement an expand AR interactive system mainly on the face and gesture recognition algorithm Adaboost algorithm and the methods generating the classifier with opencv and then give a human face and gesture recognition experimental results and statistical data. The results illustrate that the human-computer interaction in this paper is very effective.