Contribution towards a fast stereo dense matching.

Date of Award


Publication Type

Master Thesis

Degree Name



Computer Science


Computer Science.



Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.


Stereo matching is important in the area of computer vision as it is the basis of the reconstruction process. Many applications require 3D reconstruction such as view synthesis, robotics... The main task of matching uncalibrated images is to determine the corresponding pixels and other features where the motion between these images and the camera parameters is unknown. Although some methods have been carried out over the past two decades on the matching problem, most of these methods are not practical and difficult to implement. Our approach considers a reliable image edge features in order to develop a fast and practical method. Therefore, we propose a fast stereo matching algorithm combining two different approaches for matching as the image is segmented into two sets of regions: edge regions and non-edge regions. We have used an algebraic method that preserves disparity continuity at the object continuous surfaces. Our results demonstrate that we gain a speed dense matching while the implementation is kept simple and straightforward.Dept. of Computer Science. Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis2005 .Z42. Source: Masters Abstracts International, Volume: 44-03, page: 1420. Thesis (M.Sc.)--University of Windsor (Canada), 2005.