CONTROL OF ENERGY EFFICIENCY IN INDUSTRY AND HOUSING AND COMMUNAL SERVICES
Sign | Register
UDC 004.021
Using of the scaled disparity map for the first iteration of the matching cost HMI calculation process
A.V. Argutin, South Ural State University, Chelyabinsk, Russian Federation, alex.argutin@gmail.com
Abstract
The article describes the algorithm of the matching cost calculation based on mutual information between pixels of stereo-images. The second part explains mathematical formulation of the mutual information based on image entropies. Mutual information application and optimization approaches in disparity calculation are described next. The «Algorithm peculiarities » paragraph analyses possible algorithm conditions, errors and their reasons.
Keywords
stereo-vision, mutual information, matching cost, probability distribution
References
1. Scharstein, D. A Taxonomy and Evaluation of Dense Two Algorithms / D. Scharstein, R. Szeliski // IJCV.
2. Hirshmuller, H. Evaluation of Cost Functions for Stereo Matching / H. Hirshmuller // CVPR. 2007. – 17–2(6). – P. 1–8.
3. Birchfield, S. Depth Discontinuities by Pixel-to-pixel Stereo / S. Birchfield, C. Tomasi // IJCV. – 1999. – 35(3) – P. 1–25.
4. Egnal, G. Mutual Information as a Stereo Correspondence Measure / G. Egnal // CIS Technical Reports. – 2009. – 18(2). – P. 124–132.
5. Viola, P. Alignment by maximization of mutual information / P. Viola, W. M. Wells // IJCV, 24(2). – 1997. – P. 137–154.
6. Kim, J. Visual Correspondence Using Energy Minimization and Mutual Information / J. Kim, V. Kolmogorov, R. Zabih // ICCV. – Vol. 2. – 2003. – P. 1033–1040.
7. Hirschmuller, H. Accurate and Efficient Stereo Processing by Semi-Global Matching and Mutual Information // CVPR. – 2005. – 20–26(6). – Vol. 2. – P. 807–814.
Source
Bulletin of the South Ural State University. Ser. Computer Technologies, Automatic Control, Radio Electronics, 2013, vol. 13, no. 2, pp. 118-121. (in Russ.) (Brief Reports)