|
Augmented Reality Indoor Navigation System |
|---|---|
| รหัสดีโอไอ | |
| Creator | บุณฑริกา โพชฌงค์เดช |
| Title | Augmented Reality Indoor Navigation System |
| Contributor | เอื้อมพร รักกำเหนิด, สมเกียรติ วังศิริพิทักษ์ |
| Publisher | Faculty of Information Science and Technology, Mahanakorn University of Technology |
| Publication Year | 2561 |
| Journal Title | Journal of Information Science and Technology |
| Journal Vol. | 8 |
| Journal No. | 1 |
| Page no. | 1-15 |
| Keyword | Indoor navigation, augmented reality, natural marker, smartphone |
| URL Website | https://tci-thaijo.org/index.php/JIST |
| Website title | Journal of Information Science and Technology |
| ISSN | 2651-1053 |
| Abstract | Augmented reality based indoor navigation system running on a smartphone is proposed to be used for in-building navigation. The system uses a built-in camera to capture the image of surroundings, detects a natural marker in the image, and calculates the pose of the camera with respect to the marker. The position and orientation of the camera (which are the same as the smartphone itself) with respect to the indoor map are then determined using the pose information of that markernote that each marker must be pre-registered with pose information in the system. Once the destination is specified by the user, the shortest path to that destination will be calculated and the arrow pointing along the path to the destination will be augmented on the scene. The information message explaining the route will also be annotated on the screen and be read out to help guide users to the destination. In addition, the system can display a top download map of building, showing current position and facing direction of the user, and drawing the route to the destinationthe top-download mode provides a better understanding and experience for the user. The accuracy of marker detection in the proposed system depends on the distance from the marker, the downloading angle, the type of the camera, and characteristics of the marker. Experimental results in real environments show that more than 70% of detection accuracy is achieved when the marker has high details and uniqueness, regardless of camera type. The downloading angle, on the other hand, has less impact on detection accuracy; except when many other irrelevant scene components appear in the download. The detection error is mostly caused by the 'no matches found'not the 'mismatch'; a slight movement of the camera normally helps the system correctly recognize the place. The calculation of the shortest path to the destination, the display of route and arrow, and the voice guidance work perfectly without error. |
