Vision-Based Positioning Method Based on Landmark Using Multiple Calibration Lines

Currently, with the significant development of the personal terminal's processing rapidly, vision-based indoor positioning has become a hot area of research. Compared with the traditional algorithm, this method is deployed with lower cost. In addition, it provides more robust positioning results and extra visualized services. The positioning results of the traditional method relies heavily on the density of position fingerprinting. Location accuracy could be improved when the fingerprinting is concentrated. However, this causes a greater time delay because of the bigger database and vice versa. This paper proposes a vision-based indoor positioning method based on landmarks to solve above problems. It reduces time complex degree by using SURF-based object image feature matching and improves the location accuracy by adding extra homography matrix and projection constrain information. It leverages several landmarks instead of redundant images in database. Moreover, additional priori information, such as homography matrix constraint and multiple calibration lines' projection relations from landmark to image, could optimize the location results smoothly.