摘要
针对深度学习运用于雷达目标识别时存在的数据量欠缺和数据人工标注难等问题,提出将传统目标识别方法与人工智能技术结合,建立面向应用的新的目标识别架构,通过融合处理以及基于传统方法的机器自动标注,获得更优越的识别效果,大幅减少人工标注的工作量,确保系统在低数据量、低标注数据下仍可维持一定的识别效果。雷达实测数据证明了该识别框架的有效性。
An application oriented new target recognition architecture is proposed in this paper to solve the problem of poor data and manual annotation by combination of traditional radar automatic target recognition methods and artificial intelligence technology.Recognition performance is proved and the workload of manual annotation is reduced too by fusion processing and machine automatic labeling based on traditional method.This ensures that the system can maintain a certain recognition effect under low data quantiby and low label data.The results of radar real data verify that the new architecture is effective.
作者
李士国
张瑞国
孙晶明
孙俊
LI Shiguo;ZHANG Ruiguo;SUN Jingming;SUN Jun(Nanjing Research Institute of Electronics Technology,Nanjing 210039,China;Key Laboratory of IntelliSense Technology,CETC,Nanjing 210039,China;Taiyan Satellite Launch Center,Taiyuan 030027,China)
出处
《现代雷达》
CSCD
北大核心
2019年第11期57-61,84,共6页
Modern Radar
关键词
雷达
自动目标识别
深度学习
架构
radar
automatic target recognition(ATR)
deep learning
architecture
