An Augmented Reality Navigation System Based On Android

Augmented reality is a new technology developed on the basis of virtual reality.It combines the technology of computer-generated graphs and images, space positioning technology, visualization technology, interaction technology and other related technologies. It overlays computer-generated virtual items, scenes or system's prompt messages on real scenes, thus achieving the augment of real environment and providing users with brand-new sensory experience. In recent years, the hardwares of mobile smart phones develop rapidly. With the advantages of fast processors,multi-sensors, GPS, built-in cameras with high resolution and large storage capacity and so on, they are the ideal platform to achieve augmented reality. There have been some attempts at home and abroad to introduce augmented reality into the fields of outdoor and indoor guidance and navigation in order to improve users' sensory experience of diversified augmented information.Based on Android system, this paper deeply explores technologies related to the mobile augmented reality system. It combines two methods of tracking registration respectively based on sensors and image's natural features. Taking the campus of Nanjing University of Information Science & Technology for example,it designs and realizes an augmented reality guidance system suitable for indoor and outdoor scenes.The main research contents in this paper include the following points:Firstly, this paper gives a detailed analysis on the functional requirements on the mobile augmented reality guidance system and on that basis the system's overall framework and main modules are designed. The system is designed with outdoor and indoor modes. The methods of tracking registration respectively based on sensors and image's natural features are combined to make full use of their advantages of tracking registration in different scenes.With regard to the tracking registration of interest points under the outdoor mode,this paper combines sensors and GPS, and extends the interval of GPS positioning to ensure real-time registration during tracking while reducing battery consumption. With regard to the tracking registration under the indoor mode, this paper combines the concepts of ORB algorithm and FREAK algorithm. It utilizes ORB algorithm to detect image's feature points, utilizes FREAK algorithm to calculate the descriptors of feature points, and matches the descriptors via the brute matcher to complete the tracking registration of interest points. Related tests are conducted and the fastness and robustness of this method are verified.This paper realizes a campus guidance system based on augmented reality. The augmented interest points involve all teaching buildings and landmarks on campus and some special indoor interest points; the augmented information includes texts,images and videos and so on. Finally, it conducts tests on the system's overall performance and verifies the real-time performance, stability and practicability of the system.