摘要
针对地面及低空雷达目标识别中存在的雷达散射截面积小、回波幅度低、多普勒频移不明显等因素引起的目标特征被淹没难以检测识别的问题,提出了一种微多普勒特征与距离-多普勒特征融合识别方法.首先基于距离-多普勒二维矩阵,提出一种新型十字开窗目标检测手段,然后设计微动特征与距离-多普勒特征等多元特征提取方法,最后根据多元特征融合结果设计分类器进行目标分类识别.实验结果表明,相对于传统的运动、微动特征提取与分类,基于多元特征融合的分类器设计可以有效地提升地面及低空慢速小目标的识别准确率.
In the ground and low altitude radar target recognition, it is difficult to detect and recognize the target features submerged due to the small radar cross section area, low echo amplitude, insignificant Doppler frequency shift and other factors. In order to solve this problem, the paper proposes a fusion recognition method of micro-Doppler feature and range Doppler feature. Firstly, based on the range-Doppler two-dimensional matrix, the paper presents a novel cross-windowing detection method, then designs the extraction method of multi-features such as fretting feature, distance Doppler feature, etc., and finally accords to the result of multi-feature fusion to design a classifier for target classification and recognition. Experimental results show that compared with traditional motion and fretting feature extraction and classification, the classifier design based on multi-feature fusion can effectively improve the recognition accuracy of ground and low-altitude low-speed small targets.
作者
杨永昌
王曙光
YANG Yongchang;WANG Shuguang(No.38 Research Institute,China Electronic Technology Group Corporation,Hefei 230088,China;Key Laboratory of Aperture Array and Space Application,Hefei 230088,China)
出处
《空军预警学院学报》
2021年第6期416-420,共5页
Journal of Air Force Early Warning Academy
关键词
地面及低空目标
特征融合
距离-多普勒
目标识别
ground and low altitude targets
feature fusion
range-Doppler
target recognition
