30 October 2009 Airborne target tracking algorithm against oppressive decoys in infrared imagery
Author Affiliations +
Proceedings Volume 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis; 74953I (2009) https://doi.org/10.1117/12.832481
Event: Sixth International Symposium on Multispectral Image Processing and Pattern Recognition, 2009, Yichang, China
Abstract
This paper presents an approach for tracking airborne target against oppressive infrared decoys. Oppressive decoy lures infrared guided missile by its high infrared radiation. Traditional tracking algorithms have degraded stability even come to tracking failure when airborne target continuously throw out many decoys. The proposed approach first determines an adaptive tracking window. The center of the tracking window is set at a predicted target position which is computed based on uniform motion model. Different strategies are applied for determination of tracking window size according to target state. The image within tracking window is segmented and multi features of candidate targets are extracted. The most similar candidate target is associated to the tracking target by using a decision function, which calculates a weighted sum of normalized feature differences between two comparable targets. Integrated intensity ratio of association target and tracking target, and target centroid are examined to estimate target state in the presence of decoys. The tracking ability and robustness of proposed approach has been validated by processing available real-world and simulated infrared image sequences containing airborne targets and oppressive decoys.
© (2009) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xiechang Sun, Xiechang Sun, Tianxu Zhang, Tianxu Zhang, } "Airborne target tracking algorithm against oppressive decoys in infrared imagery", Proc. SPIE 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis, 74953I (30 October 2009); doi: 10.1117/12.832481; https://doi.org/10.1117/12.832481
PROCEEDINGS
7 PAGES


SHARE
Back to Top