高旋掠飞弹载激光雷达对地面装甲目标的分割与识别
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

装备重点预研项目(No.301070201)资助


Segmentation and recognition of ground armor target for the hedgehopping munition with high rotational speed carrying laser radar
Author:
Affiliation:

Fund Project:

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

    为了完成搭载单元激光雷达的高旋掠飞弹药对地面装甲目标的提取与识别,充分考虑高旋掠飞弹药的稳态扫描过程,提出了一种用于弹载单元激光雷达一维距离像的目标分割与识别方法。首先通过基于梯度的自动生长阈值分割算法对一维点云数据进行预分割,并获取初始聚类数与聚类中心;接着基于改进的kmeans算法对分割结果进行调整,实现对装甲目标的优化分割;最后将分割出的目标与真实装甲目标进行特征匹配从而识别装甲目标。实验结果表明:此方法能在不同的高度、视场角下有效地对装甲目标进行分割与识别,为高旋掠飞弹药目标识别研究提供技术支持。

    Abstract:

    In order to complete the extraction and recognition of the ground armored target by the hedgehopping munition with high rotational speed which carries the single-line laser radar,fully considering the steady scanning process of the hedgehopping munition with high rotational speed,a target extraction method for the one-dimensional range profile of missile-borne single-line laser radar is proposed.Firstly,a threshold self-growth segmentation algorithm based on gradient is proposed to obtain the initial cluster number and cluster center.Then based on the improved kmeans algorithm,the segmentation results are adjusted and optimized to achieve the correct segmentation of armored targets.Finally,based on the real feature matching of segmented target and armored target,the armored target is identified.The experimental result shows that this method can effectively segment and identify the target at different height and field of view angle,and provide technical support for the target identification of the hedgehopping munition with high rotational speed.

    参考文献
    相似文献
    引证文献
引用本文

蒋罕寒,郭锐,武军安,赵旭.高旋掠飞弹载激光雷达对地面装甲目标的分割与识别[J].激光与红外,2021,51(2):166~170
JIANG Han-han, GUO Rui, WU Jun-an, ZHAO Xu. Segmentation and recognition of ground armor target for the hedgehopping munition with high rotational speed carrying laser radar[J]. LASER & INFRARED,2021,51(2):166~170

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2021-03-04
  • 出版日期: