基于多特征融合的导弹类目标分类识别方法
作者:
作者单位:

1.中国电子科技集团公司,第三十八研究所,安徽 合肥 230088;2.中国电子科技集团公司,孔径阵列与空间探测安徽省重点实验室,安徽 合肥 230088

作者简介:

王 凤(1991-),女,博士,工程师,主要研究方向为雷达目标识别.email:1290555870@qq.com.
王 斌(1987-),男,博士,高级工程师,主要研究方向为雷达目标识别、人工智能、机器学习.

通讯作者:

基金项目:

伦理声明:



A classification and recognition method based on multi-feature fusion for missile targets
Author:
Ethical statement:

Affiliation:

1.The 38th Research Institute ;2.Anhui Provincial Key Laboratory of Aperture Array and Space Detection ,China Electronics Technology Group Corporation,Hefei Anhui 230088,China

Funding:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
    摘要:

    常规的巡航导弹、空地导弹、制导炸弹3种导弹类目标的分类识别多依赖于目标的运动特性,往往利用目标的高度、速度等运动特征参数作为分类判决的依据,随着导弹武器研制技术的发展,单维度的运动特性已经无法满足精细化识别的需求。针对这一问题,设计了一种导弹类目标分类识别方法,通过融合目标运动特性和雷达截面积(RCS)序列特性,利用多特征层次化识别方法实现对3种导弹类目标的分类识别。仿真试验结果表明该方法对3种导弹目标的分类准确率优于95%,验证了该方法的有效性。

    Abstract:

    The conventional classification and identification of cruise missiles, air-to-ground missiles, and guided bombs rely heavily on the kinematic characteristics of the targets, often using parameters such as the target's altitude and speed as the basis for classification judgment. With the advancement of missile weapon development technology, single-dimensional kinematic characteristics can no longer meet the demand for refined identification. To address this issue, a classification and identification method for missile targets has been designed, which integrates the kinematic characteristics and Radar Cross-Section(RCS) series characteristics of the targets. By employing a multi-feature hierarchical identification approach, it achieves the classification and identification of the three types of missile targets. Simulation test results indicate that the classification accuracy of this method for the three types of missile targets is better than 95%, validating the effectiveness of the method.

    参考文献
    相似文献
    引证文献
引用本文

王凤,王斌.基于多特征融合的导弹类目标分类识别方法[J].太赫兹科学与电子信息学报,2024,22(11):1262~1269

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
历史
  • 收稿日期:2023-01-16
  • 最后修改日期:2023-07-07
  • 录用日期:
  • 在线发布日期: 2024-12-11
  • 出版日期:
关闭