基于谱峰个数的无人机与武装单兵分类识别算法  

A Recognition Algorithm for Distinguishing UAVs and Armed Individuals Based on the Number of Spectrum Peaks

作　　者：陈尹翔 王育东[1] 李硕[1] 张都川 CHEN Yinxiang;WANG Yudong;LI Shuo;ZHANG Duchaun(Xi'an Electronic Engineering Research Institute,Xi'an 710100)

机构地区：[1]西安电子工程研究所,西安710100

出　　处：《火控雷达技术》2024年第2期55-61,共7页Fire Control Radar Technology

摘　　要：本文基于无人机与武装单兵分类识别问题,提出了基于多普勒谱峰个数的目标识别算法,该算法用于战场环境侦察雷达快速准确地鉴别无人机与武装单兵。本文解决了低空无人机与地面武装单兵难以区分的问题;算法中充分利用了无人机旋翼转动与武装单兵四肢摆动在多普勒谱上的差异,提出了多普勒谱峰个数特征来进行分类识别。解决了传统方法难以识别无人机与武装单兵的问题。This paper proposes a target recognition algorithm using the number of Doppler spectrum peaks to solve the problem of classification and recognition of UAVs and armed individuals.The proposed algorithm can be used by battlefield reconnaissance radar to quickly and accurately identify UAVs and armed individuals,significantly reducing the difficulty in distinguishing low-altitude UAVs and ground armed individuals.It makes full use of the differences in the Doppler spectrum caused by the rotation of the UAVs rotors and the swing of the limbs of the armed individuals.By using the number of Doppler spectrum peaks for classification and recognition,the proposed algorithm can distinguish UAVs and armed individuals which are difficult to be distinguished using traditional methods.

关 键 词：窄带雷达 多普勒谱峰个数 支持向量机 

分 类 号：TN95[电子电信—信号与信息处理]