The Design And Implementation Of Augmented Reality Client Based On Android

Augmented reality,which is based on the virtual reality from the development, is also known as mixed reality. Through the perception of the real world, augmented information which generated by computer is superimposed in the proper position of the real scence image. This achieves the augmented reality effect. In recent years, with the rapid development of the mobile terminals and the mobile Internet, the mobile augmented reality has caused the wide attention of social all circles. The convenience of the mobile terminal makes augmented reality a wider use, so that users have proposed more demands of augmented reality, especially in the perception experience and the diversification of augmented information which continuously drives the development of mobile augmented reality.In this thesis, we study an augmented reality system based on the most mainstream system Android that can apply to many scences. First, we clear the research background and significance of augmented reality and analyze the current research status and the related results at home and abroad. Then we propose a scheme based on visual camera and non-visual sensors that can implement the recognition and tracking of target object under the indoor scene and street scene. Considering the limitations of the mobile terminal’s CPU, memory performance, we use the server to complete the matching algorithm. Under the indoor scene, we use visual camera to extract the feature points, and then send them to the server so that can get the response of the target positioning results which will be the input of tracking registration module to complete target tracking. Under the street scene, with a large geographic information database and the complex scenes, we use non-visual sensor technique to perceive the scene information. The server also works out the positioning information by sensors’ data. And then we use visual tracking registration techniques for target tracking. Finally, inorder to realize virtual-real fusion, use OpenGL ES to render the augmented information on the target object. After testing the function and performance of the system, we have evaluated its performance and summarized the whole development process of the system. What’s more, we also have discussed the future research direction. The main contributions and innovation points of our system are as follows:1、This system realizes multi-scenario mobile augmented reality that could be used in indoor mode and street mode.2、According to different scenarios, the fusion of visual camera and non-visual sensors sensing technologies make the tracking registration more efficient and accurate. It also reduces the incidence of network traffic.3 、 We design and implement the feature detection algorithm and tracking registration algorithm which can apply to mobile terminal. The Fast corner detection algorithm based on FastCV is fast in feature point extraction. And the ICMT algorithm is fast and accurate in tracking registration.4、Using the cross-platform open source OpenGL ES technology to complete the rendering of multiple augmented information can improve the efficiency of rendering on mobile terminals.