Recently, indoor location positioning has attracted great attention in industries and research institutes. However, one of the most critical problems is difficulty of database construction when estimating the user’s position in indoor environments. This paper proposes a method for the vision-based indoor location positioning system in the hallway using a single camera without any database construction. In this proposed approach, first, the structural information of indoor environments is used to avoid the database construction. Second, the depth information is compensated because an image captured from single camera does not have the depth information. The experimental results show that the proposed system has an average error as 1.43 m.

Keywords: indoor location positioning, indoor navigation, no database construction, perspective line detection, depth compensation