Sporathe knowledge network
ES
Universidad EAFIT

Ingeniería y Ciencia, Vol. 01, Núm. 02 (2005) · 2005

Edge and corner identification for tracking the line of sight

This article presents an edge and corner detector, implemented in the GEIST project domain (a Computer Aided Tourist Information System) to extract information from straight edges and their intersections (corners in the image) from camera images ( from the real world) contrasted with computer-generated images (from the Historical Monuments Database based on the position and orientation of a virtual observer). The camera and computer generated images are processed to reduce detail, find the skeleton of the image and detect edges and corners. The surviving corners of the detection and discovery process of the skeleton of the images are treated as reference points and fed to a matching algorithm, which estimates the sampling errors that usually contaminate the GPS and orientation data (fed to the generator images per computer). In this way, a closed loop control cycle is implemented, by means of which the system converges to the exact determination of position and orientation of an observer crossing a historical scenario (in this case, the city of Heidelberg). With this exact position and orientation, in the GEIST project other modules are capable of projecting historical re-creations in the observer's field of vision, which have the exact scenario (the real image seen by the observer). Thus, the tourist "sees" the scenes unfolding in material and real historical sites of the city. To do this, this article presents the modification and articulation of algorithms such as the Canny Edge Detector, "SUSAN Corner Detector", 1- and 2-dimensional filters, and so on.

Full text 19 pages with text

Read the full thesis Record in the repository

Keywords