现代雷达2017，Vol.39Issue(12)：24-28,5.DOI:10.16592/j.cnki.1004-7859.2017.12.005
基于卷积神经网络的高分辨距离像目标识别
High Resolution Range Profile Target Recognition Based on Convolutional Neural Network
杨予昊 1孙晶明 2虞盛康 1彭雄伟2
作者信息
- 1. 南京电子技术研究所, 南京210039
- 2. 中国电子科技集团公司智能感知技术重点实验室, 南京210039
- 折叠
摘要
Abstract
Feature extraction is the key technique for radar target recognition based on high resolution range profiles.Traditional artificial feature extraction algorithms only utilize shallow architecture features,which result in the loss of information inevitably and restrict the generalization performance of target recognition methods.Aiming at this issue,the deep learning tool is used in this paper,and a new method of target recognition based on convolutional neural network is presented.By constructing a convolutional neural network model for dealing with high resolution range profiles,and optimizing deep learning parameters,the deep property features of targets included in high resolution range profiles are fully explored,moderate automatic feature extraction is realized,and target classification is accomplished.In the end,the performance of the proposed method is validated based on measured data,and experimental results show the effectiveness of the proposed method.关键词
高分辨距离像/目标识别/深度学习/卷积神经网络Key words
high resolution range profile/target recognition/deep learning/convolutional neural network分类
信息技术与安全科学引用本文复制引用
杨予昊,孙晶明,虞盛康,彭雄伟..基于卷积神经网络的高分辨距离像目标识别[J].现代雷达,2017,39(12):24-28,5.
