Share Email Print
cover

Proceedings Paper

The electronic image stabilization technology research based on improved optical-flow motion vector estimation
Author(s): Chao Wang; Ming Ji; Ying Zhang; Wentao Jiang; Xiaoyan Lu; Jiaoying Wang; Heng Yang
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The electronic image stabilization technology based on improved optical-flow motion vector estimation technique can effectively improve the non normal shift, such as jitter, rotation and so on. Firstly, the ORB features are extracted from the image, a set of regions are built on these features; Secondly, the optical-flow vector is computed in the feature regions, in order to reduce the computational complexity, the multi resolution strategy of Pyramid is used to calculate the motion vector of the frame; Finally, qualitative and quantitative analysis of the effect of the algorithm is carried out. The results show that the proposed algorithm has better stability compared with image stabilization based on the traditional optical-flow motion vector estimation method.

Paper Details

Date Published: 26 January 2016
PDF: 7 pages
Proc. SPIE 9796, Selected Papers of the Photoelectronic Technology Committee Conferences held November 2015, 97962U (26 January 2016); doi: 10.1117/12.2229882
Show Author Affiliations
Chao Wang, Xi'an Institute of Applied Optics (China)
Ming Ji, Xi'an Institute of Applied Optics (China)
Ying Zhang, Xi'an Institute of Applied Optics (China)
Wentao Jiang, Xi'an Institute of Applied Optics (China)
Xiaoyan Lu, Xi'an Institute of Applied Optics (China)
Jiaoying Wang, Xi'an Institute of Applied Optics (China)
Heng Yang, Xi'an Institute of Applied Optics (China)


Published in SPIE Proceedings Vol. 9796:
Selected Papers of the Photoelectronic Technology Committee Conferences held November 2015
Weimin Bao; Yueguang Lv, Editor(s)

© SPIE. Terms of Use
Back to Top