检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:吴春光[1] 张文林 孙宝全 Wu Chunguang;Zhang Wenlin;Sun Baoquan(Unit 92941,PLA,Huludao 125000,China;Beijing Institute of Environmental Characteristics,Beijing 100854,China)
机构地区:[1]中国人民解放军92941部队,葫芦岛125000 [2]北京环境特性研究所,北京100854
出 处:《电子测量技术》2020年第22期77-81,共5页Electronic Measurement Technology
摘 要:针对雷达目标特性研究中目标分类识别需求,因复杂目标散射特性可等效为多类型体目标单元散射特性的合成,进而采用单元目标历程图数据训练神经网络(CNN)的方法,用以完成对平板、二面角、三面角、圆柱、圆形顶帽等5类典型体目标的分类识别。其中设计了基于历程图数据训练CNN方法及流程,构建了典型仿真案例,实现了训练准确率为99%,验证准确率为97%,且在测试集上表现良好。设计方法能够准确完成典型体目标分类识别,为进一步开展复杂目标分类识别研究提供一种新的思路。Aiming at the requirement of target classification and recognition in the research of radar target characteristics, as the scattering characteristics of complex targets can be equivalent to the synthesis of scattering characteristics of the multiple types volume target unit, the method of convolutional neural network(CNN) trained by the process chart data of unit target is used to realize the classification and recognition of five typical volume targets, including flat plate, dihedral angle, trihedral angle, cylinder and circular top hat. The method and process of CNN training based on process chart data are designed, typical simulation cases are constructed, 99% training accuracy and 97% verification accuracy are achieved, and good performance is achieved in the test set. The method designed and adopted is able to accurately classify and recognize typical volume targets, and the study provides a new idea for further research of the classification and recognition of complex targets.
分 类 号:TN955[电子电信—信号与信息处理]
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:216.73.216.90