DocumentCode :
583704
Title :
Local stereo matching using an variable window, census transform and an edge-preserving filter
Author :
Dinh, Vinh Quang ; Nguyen, Vinh Dinh ; Vinh Dinh Nguyen ; Jeon, Jae Wook
Author_Institution :
Dept. of Electr. & Comput. Eng., Sungkyunkwan Univ., Suwon, South Korea
fYear :
2012
fDate :
17-21 Oct. 2012
Firstpage :
523
Lastpage :
528
Abstract :
In this paper, we propose an alternative blocking-matching approach to the correspondence problem in stereo matching. In blocking-matching algorithms, a local window is used to measure the similarity (or dissimilarity) between pixels of a stereo pair. Although some area-based stereo matching methods have been developed and work well in many kinds of regions such as textureless or object boundary regions, their performance can degrade when working in some types of radiometric conditions. Our proposed algorithm is an improved method that uses a non-parametric transform in the pre-processing step and an edge-preserving filter in the post-processing step. Input images are first pre-processed by the census transform, which makes the proposed method more robust when the image pair is captured in different light sources or camera exposure conditions. The window cost in our approach is computed from the transformed images using the Hamming distance, and the correspondence is finally chosen by a Winner-Takes-All strategy. The experimental results for the Middleburry images show that the proposed method outperforms test local stereo methods in radiometrically different images.
Keywords :
cameras; edge detection; filtering theory; image matching; light sources; stereo image processing; transforms; Hamming distance; Middleburry images; Winner-Takes-All strategy; area-based stereo matching methods; blocking-matching approach; camera exposure conditions; census transform; edge-preserving filter; image pair; light sources; local stereo matching; local stereo methods; local window; nonparametric transform; object boundary regions; post-processing step; radiometric conditions; radiometrically different images; stereo pair pixel similarity measurement; textureless regions; variable window; Computational efficiency; Image color analysis; Lighting; Measurement errors; Radiometry; Shape; Transforms; Stereo matching; radiometric variation; variable window;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Control, Automation and Systems (ICCAS), 2012 12th International Conference on
Conference_Location :
JeJu Island
Print_ISBN :
978-1-4673-2247-8
Type :
conf
Filename :
6393501
Link To Document :
بازگشت