Electronic Image Stabilization based on Motion Classification

DOI

10.2991/icmt-13.2013.180How to use a DOI?

Keywords

electronic image stabilization, global motion estimation, motion compensation, feature points

Abstract

A novel electronic image stabilization based on motion classification is presented. It achieves image stabilization with high accuracy in complicated camera motion. Firstly, the feature points are selected evenly using the Harris operator and then matched based on the feature window. Hence, the corresponding points are obtained and the statistic features of their motions are analyzed according to different motion kinds including translation, rotation and zoom. Then, the fast motion classification method is proposed to validate all points and delete mismatched or unreliable points. Thirdly, the remained global feature points are brought to the affine model to compute global motion. Lastly, the Kalman filter is used to obtain dithering motion and each current frame is compensated. Experimental results show that the algorithm can correctly detect global motion in dynamic scenes with camera scan and various dithering. Furthermore, its estimation error is below 1/2 pixel and it also accomplishes real-time stabilization, which can greatly improve the stability and fidelity of videos.

Copyright

© 2013, 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  - Juanjuan Zhu
AU  - Baolong Guo
PY  - 2013/11
DA  - 2013/11
TI  - Electronic Image Stabilization based on Motion Classification
BT  - Proceedings of 3rd International Conference on Multimedia Technology(ICMT-13)
PB  - Atlantis Press
SP  - 1466
EP  - 1473
SN  - 1951-6851
UR  - https://doi.org/10.2991/icmt-13.2013.180
DO  - 10.2991/icmt-13.2013.180
ID  - Zhu2013/11
ER  -