A vision-based positioning system

Abstract

Technology in image processing had always been improving over the past few decades. From the simplest face detection, to the face recognition programs the world have now, the ability to break down and process images to each individual’s requirement had helped mankind in many areas, such as tracking and identification.

In this project, different approaches to identify obstacles as objects from an image are investigated and experimented. The identified obstacles in the image is then plotted into a grid, using x-y co-ordinates. This is further improved from a 2D surface to a 2D surface on a 3D grid, and a 3D structure from the 2D surface. This can help the user to identify the relative position of the obstacle in the image. This can be used to help plot a good estimate of the location of an object of interest within a confined space. If further improved, this can be used as a good tool for terrain mapping.