Motion detection using multiscale optical flow in infrared imagery

Wen-shuai Yu; Xu-chu Yu; Bing Chen; Yue Chang

doi:10.1117/12.832958

30 October 2009 Motion detection using multiscale optical flow in infrared imagery

Wen-shuai Yu, Xu-chu Yu, Bing Chen, Yue Chang

Proceedings Volume 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis; 74951W (2009) https://doi.org/10.1117/12.832958
Event: Sixth International Symposium on Multispectral Image Processing and Pattern Recognition, 2009, Yichang, China

Abstract

Motion detection is one of the basic operations in dynamic image analysis. In thermal image sequence processing, motion detection is a difficult problem. In this paper, we employ multi-scale optical flow methods to solve the problem. After analyzing the difficulties in detecting motions, multi-resolution framework is recognized as an useful approach for the process. Phase information has been demonstrated its robustness and stability in disparity and flow estimation. Velocity-tuned filtering integrates the multi-scale framework with the phase information. Using a bank of Gabor filters, image velocities, which form the flow field, are estimated with the output phase responses. Comparing with Lucas-Kanade pyramid method, our experiment shows that the approach could compute the optical flow more accurately, and it is effective for motion detection in infrared images.

Citation Download Citation

Wen-shuai Yu, Xu-chu Yu, Bing Chen, and Yue Chang "Motion detection using multiscale optical flow in infrared imagery", Proc. SPIE 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis, 74951W (30 October 2009); https://doi.org/10.1117/12.832958

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available