Journal of Signal and Information Processing, 2015, 6, 191-202
Published Online August 2015 in SciRes. http://www.scirp.org/journal/jsip
http://dx.doi.org/10.4236/jsip.2015.63018
How to cite this paper: Borisagar, V.H. and Zaveri, M.A. (2015) Census and Segmentation-Based Disparity Estimation Algo-
rithm Using Region Merging. Journal of Signal and Information Processing, 6, 191-202.
http://dx.doi.org/10.4236/jsip.2015.63018
Census and Segmentation-Based Disparity
Estimation Algorithm Using Region Merging
Viral H. Borisagar
1
, Mukesh A. Zaveri
2
1
Computer Engineering Department, Government Engineering College, Gandhinagar, India
2
Computer Engineering Department, Sardar Vallabhbhai National Institute of Technology, Surat, India
Email: viralborisagar@gmail.com , mazaveri@coed.svnit.ac.in
Received 12 May 2015; accepted 11 July 2015; published 15 July 2015
Copyright © 2015 by authors and Scientific Research Publishing Inc.
This work is licensed under the Creative Commons Attribution International License (CC BY).
http://creativecommons.org/licenses/by/4.0/
Abstract
Disparity estimation is an ill-posed problem in computer vision. It is explored comprehensively
due to its usefulness in many areas like 3D scene reconstruction, robot navigation, parts inspec-
tion, virtual reality and image-based rendering. In this paper, we propose a hybrid disparity gen-
eration algorithm which uses census based and segmentation based approaches. Census trans-
form does not give good results in textureless areas, but is suitable for highly textured regions.
While segment based stereo matching techniques gives good result in textureless regions. Coarse
disparities obtained from census transform are combined with the region information extracted
by mean shift segmentation method, so that a region matching can be applied by using affine
transformation. Affine transformation is used to remove noise from each segment. Mean shift
segmentation technique creates more than one segment of same object resulting into non-smooth-
ness disparity. Region merging is applied to obtain refined smooth disparity map. Finally, multila-
teral filtering is applied on the disparity map estimated to preserve the information and to smooth
the disparity map. The proposed algorithm generates good results compared to the classic census
transform. Our proposed algorithm solves standard problems like occlusions, repetitive patterns,
textureless regions, perspective distortion, specular reflection and noise. Experiments are per-
formed on middlebury stereo test bed and the results demonstrate that the proposed algorithm
achieves high accuracy, efficiency and robustness.
Keywords
Stereo Vision, Census Transform, Mean Shift Segmentation, Affine Transform, Region Merging
1. Introduction
Stereo vision is a fundamental problem in computer vision. An extensive analysis of stereo matching algorithms