By Topic

Optical flow computation on a heterogeneous platform

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Jinsoo Oh ; School of Computer Science, Kookmin University, Seoul, 136-702, Korea ; Eun-Jin Im ; Kyoungro Yoon

Unmanned Aerial Vehicles (UAV) are finding their way to a wide range of safety-critical missions. Our collaborative research team of computer scientists and aerospace engineers has worked on developing hardware and software of UAV. One of vital components in software used in UAV is image stabilization. The constantly-shaking images taken from the UAV, resulted from the vehicle's motion, need to be stabilized to perform its mission. In this paper, we present our implementation of image stabilization software. Our research is focused on using state-of-the-art Graphic Processing Unit (GPU) to improve the performance of the image stabilization software. The stabilizer estimates motion of the vehicle by calculating optical flow between successive two frames. In this study, we parallelized the calculation of the optical flow, which is identified as a computational bottleneck of the entire image stabilization process. Using the massive parallelism of NVIDIA C2060 GPU with 448 cores, we could improve the overall performance of image stabilizer.

Published in:

Ubiquitous Robots and Ambient Intelligence (URAI), 2011 8th International Conference on

Date of Conference:

23-26 Nov. 2011