In this paper, a recursive higher order statistics algorithm is proposed for small target detection in temporal domain. Firstly, the background of image sequence is normalized. Then, the higher order statistics are recursively solved in image sequence to obtain the feature image. Finally, the feature image is segmented with threshold to detect the small target. To validate the algorithm proposed in this paper, five simulated and one semi-simulation image sequences are created. The ROC curves are employed for evaluation of experimental results. Experiment results show that our method is very effective for small target detection.
Detection of small target using recursive higher order statistics
Selected Papers from Conferences of the Photoelectronic Technology Committee of the Chinese Society of Astronautics: Optical Imaging, Remote Sensing, and Laser-Matter Interaction 2013 ; 2013 ; SuZhou,China
Proc. SPIE ; 9142
2014-02-21
Conference paper
Electronic Resource
English
Detection of small target using recursive higher order statistics [9142-12]
British Library Conference Proceedings | 2014
|CORRESPONDENCE - Small Target Detection in Clutter Using Recursive Nonlinear Prediction
Online Contents | 2000
|PAPERS - On Detection Using Filter Banks and Higher Order Statistics
Online Contents | 2000
|