PREDICT OF THE SMALL SIZE OBJECT POSITION ON THE VIDEO STREAM FRAMES FROM THE BOARD OF UNMANNED AERIAL VEHICLE
Abstract
Algorithms of video stabilization from unmanned aerial vehicle board and search of small size targets on the video sequence frames with the prediction are proposed. Proposed stabilization method provides a parametrization and localization time reduction to 5.2 times compared with the SURF method. Proposed search algorithm speeds up frame processing time approximately in 6.2 times in comparison with the standard covariance method.
About the Author
I. A. Baryskievic
Белорусский государственный университет информатики и радиоэлектроники
Belarus
Belarus
References
1. Lowe D.G. // Intern. Journ. of Computer Vision. 2004. Vol. 60, № 2. P. 91-110.
2. Bay H. // Computer Vision and Image Understanding. 2008. Vol. 110, № 3. P. 346-359.
3. Борискевич, И.А., Цветков В.Ю. // Докл. БГУИР. 2014. № 3. С. 33-39.
4. Porikli F., Tuzel O. // Proceedings of IEEE Int’l. Conference on Multimedia and Expo. 2005. P. 1234-1237.
5. Tuzel O., Porikli F. // Proc. 9th European Conf. on Computer Vision. 2006. Vol. 2. P. 589-600.
6. Борискевич И.А., Цветков В.Ю. // Весці Нацыянальнай акадэміі навук Беларусі. Сер. фізіка-тэхнічных навук. 2015. № 1. С. 106-111.
Review
For citations:
Baryskievic I.A. PREDICT OF THE SMALL SIZE OBJECT POSITION ON THE VIDEO STREAM FRAMES FROM THE BOARD OF UNMANNED AERIAL VEHICLE. Doklady BGUIR. 2016;(2):107-113. (In Russ.)
Views: 325
Email this article 