One Point Perspective Vanishing Point Estimation for Mobile Robot Vision Based Navigation System

Gideon Kanji Damaryam

Abstract: Where it can be found, the vanishing point in a one point perspective image can facilitate analysis of the image. For the purpose of a vision system for a small mobile robot self-navigating within a rectilinear indoor environment such as along the corridor of a faculty building with one point perspective, knowing where the vanishing point in an image captured for navigation purpose is, can be very helpful in recognition of high level features such as corridors and doors. This paper presents a parameterised algorithm developed to estimate the vanishing point in such an image of the inside of a corridor. It takes advantage of certain characteristics of vanishing points from one point perspective, and the nature of the images encountered in this work. This algorithm is based on likelihoods. It has worked in most cases, been undecided in other cases.

Keywords: feature recognition, image processing, path detection, self-navigation, vanishing point estimation

How to Cite?: Gideon Kanji Damaryam, "One Point Perspective Vanishing Point Estimation for Mobile Robot Vision Based Navigation System", Volume 5 Issue 2, February 2016, International Journal of Science and Research (IJSR), Pages: 930-934, https://www.ijsr.net/getabstract.php?paperid=NOV161279, DOI: https://dx.doi.org/10.21275/NOV161279