雷达科学与技术Issue(6)：633-638,6.DOI:10.3969/j.issn.1672-2337.2015.06.013
基于特征融合的雷达视频运动目标检测
Moving Target Detection in Radar Video Based on Feature Fusion
摘要
Abstract
In order to solve the problem of high false track rate of radar automatic detection and tracking in dense clutter and complex multi-target situation,the suspected targets from a single-frame radar image are firstly detected using the iterative threshold segmentation method,and the accumulation results of the single-frame detection results are gained.Then,the area change ratio and the energy density of targets are extracted based on single-frame detection results and multi-frame accumulation results,and their basic probability as-signment functions are built respectively.Finally,the optimal threshold is obtained using decision fusion based on the D-S evidence theory to realize the automatic detection of moving targets.The effectiveness of the algorithm is verified by the collected marine radar data.关键词
雷达视频/运动目标检测/D-S 证据理论/特征融合Key words
radar video/moving target detection/D-S evidence theory/feature fusion分类
信息技术与安全科学引用本文复制引用
于晓涵,周伟,关键..基于特征融合的雷达视频运动目标检测[J].雷达科学与技术,2015,(6):633-638,6.基金项目
国家自然科学基金(No.61179017,61201445,61002045,61302008) （No.61179017,61201445,61002045,61302008）
“泰山学者”建设工程专项经费资助项目 （）
