Optimization Method for Stereo Matching Based on Minimum Spanning Tree

DOI

10.2991/ncce-18.2018.177How to use a DOI?

Keywords

Stereo matching; Disparity refinement; MST; image up-sampling.

Abstract

This paper describes the application of minimum spanning tree algorithm in process of stereo matching and optimization of depth map. A stereo matching method for the real-time application scene of binocular vision is proposed, which the stereo matching is completed under the sample of the image and reduced the image size by the minimum spanning tree of original image. On the one side, the method reduces the amount of the image under the sample, and then uses the minimum spanning tree to form the relationship between the original image and the lower sample image. On the other side, the lower sampling image uses the minimum spanning tree to the cost aggregation, so that the cost aggregation can through the local limit. This method can generate global disparity map without mismatched black voids, and it also has a good effect on optimizing depth map obtained by other methods.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jingyu Yi
AU  - Binwen Fan
AU  - Xiaopeng Cui
PY  - 2018/05
DA  - 2018/05
TI  - Optimization Method for Stereo Matching Based on Minimum Spanning Tree
BT  - Proceedings of the 2018 International Conference on Network, Communication, Computer Engineering (NCCE 2018)
PB  - Atlantis Press
SP  - 1059
EP  - 1063
SN  - 1951-6851
UR  - https://doi.org/10.2991/ncce-18.2018.177
DO  - 10.2991/ncce-18.2018.177
ID  - Yi2018/05
ER  -