Home // ICDT 2013, The Eighth International Conference on Digital Telecommunications // View article

Performance Analysis of Stereo Matching Using Segmentation Based Disparity Map

Authors:
Arti Khaparde
Apurva Naik
Manini Deshpande
Sakshi Khar
Kshitija Pandhari
Mayura Shewale

Keywords: Disparity Map; PSNR; Mean Square Error (MSE); Compression ratio; Particle Swarm Optimization.

Abstract:
Stereo vision has been studied extensively due to its usefulness in many applications like 3D scene reconstruction, robot navigation, etc. Rather than finding out the disparity between two original stereo images, various segmentation techniques are used to segment the images and the disparity between the resulting segmented images is calculated. The comparison between the disparity of the original stereo image pair and that of the segmented image pair is done on the basis of compression ratio and Peak-Signal to Noise Ratio (PSNR), which is calculated for image quality measurement. Segmentation techniques like Mean Shift Algorithm, K-means Algorithm and Particle Swarm Optimization (PSO) are used and their results are compared on the basis of subjective and objective parameters. The experimental results show that PSO based 3D image reconstruction gives a good compromise between subjective quality and compression ratio.

Pages: 38 to 43

Copyright: Copyright (c) IARIA, 2013

Publication date: April 21, 2013

Published in: conference

ISSN: 2308-3964

ISBN: 978-1-61208-262-2

Location: Venice, Italy

Dates: from April 21, 2013 to April 26, 2013