30 October 2009 A novel algorithm for small dim target detection in distorted infrared image sequence
Author Affiliations +
Proceedings Volume 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis; 749520 (2009) https://doi.org/10.1117/12.833007
Event: Sixth International Symposium on Multispectral Image Processing and Pattern Recognition, 2009, Yichang, China
Abstract
A novel algorithm is proposed in this paper to detect dim moving point target with several pixels size in infrared image sequence, taken by line scan camera in high speed moving platform, with heavy background clutter and distortion. In the algorithm, the regional information of candidate points, which are extracted from the images after background suppression by spatial filter, is employed to track candidate points. After tracking candidate points, the velocity of every candidate points can be calculated easily, and false candidate points which are extracted from clutter background, can be find out by velocity classification and removed. Then, Double Hough Transform is applied on associating the remnant candidate points in multi-frame sequence to detect the target. The algorithm can get a high detection probability and reduce false alarm rate dramatically.
© (2009) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xiaolei Qu, Xiaolei Qu, Zhengrong Zuo, Zhengrong Zuo, } "A novel algorithm for small dim target detection in distorted infrared image sequence", Proc. SPIE 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis, 749520 (30 October 2009); doi: 10.1117/12.833007; https://doi.org/10.1117/12.833007
PROCEEDINGS
8 PAGES


SHARE
Back to Top