摘要
提出一种基于卷积神经网络(Convolution Neural Network,CNN)的高分辨率雷达目标识别方法.首先针对小样本应用于深度CNN时训练过程中损失函数值收敛速度慢的问题,利用结合批归一化算法的改进CNN网络对高分辨距离像(High Resolution Range Profile,HRRP)进行自动特征提取;再利用支持向量机(Support Vector Machine,SVM)对距离像特征进行分类.使用军事车辆高保真电磁仿真数据对提出的方法进行验证,识别结果证明了该方法的有效性.
A new method of high resolution radar target recognition based on Convolution Neural Network(CNN)was presented.To solve the problem of slow convergence of loss function values during the training process when small samples are applied to the deep CNN,High Resolution Range Profile(HRRP)features were firstly extracted by using the improved CNN combined with the Batch Normalization(BN)algorithm,and then classified by using a Support Vector Machine(SVM).The experimental results using high-fidelity electromagnetic simulation data of military vehicles validate the effectiveness of the proposed method.
作者
何松华
张润民
欧建平
张军
HE Songhua;ZHANG Runmin;OU Jianping;ZHANG Jun(College of Computer Science and Electronic Engineering,Hunan University,Changsha 410082,China;ATR Laboratory,National University of Defense Technology,Changsha 410073,China)
出处
《湖南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2019年第8期141-148,共8页
Journal of Hunan University:Natural Sciences
基金
国家自然科学基金资助项目(61471370)~~
关键词
高分辨距离像
雷达目标识别
卷积神经网络
批归一化
支持向量机
High Resolution Range Profile(HRRP)
radar target recognition
Convolution Neural Network(CNN)
Batch Normalization(BN)
Support Vector Machine(SVM)
