期刊
INFRARED PHYSICS & TECHNOLOGY
卷 57, 期 -, 页码 68-75出版社
ELSEVIER
DOI: 10.1016/j.infrared.2012.12.011
关键词
Infrared point target; Template matching; Non-linear correlation coefficient; Target tracking; Kalman prediction; Principal component analysis
资金
- Natural Science Foundation of Jiangsu Province of China through Huaihai Institute of Technology [BK2012663]
- National Nature Science Foundation of China [61174013]
- Natural Science Foundation for Colleges and Universities in Jiangsu Province [10KJB510002]
For a long time, tracking IR point targets is a great challenge task. We propose a tracking framework based on template matching combined with Kalman prediction. Firstly, a novel template matching method for detecting infrared point targets is presented. Different from the classic template matching, the projection coefficients obtained from principal component analysis are used as templates and the non-linear correlation coefficient is used to measure the matching degree. The non-linear correlation can capture the higher-order statistics. So the detection performance is improved greatly. Secondly, a framework of tracking point targets, based on the proposed detection method and Kalman prediction, is developed. Kalman prediction reduces the searching region for the detection method and, in turn, the detection method provides the more precise measurement for Kalman prediction. They bring out the best in each other. Results of experiments show that this framework is competent to track infrared point targets. (C) 2012 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据