+高级检索
首页 > 过刊浏览>2021年第40卷第6期 >870-877. DOI:10.11972/j.issn.1001-9014.2021.06.023
PDF HTML阅读 XML下载 导出引用 引用提醒
高定位精度的毫米波全息图像三维目标检测
作者:
作者单位:

1.中国科学院上海微系统与信息技术研究所 中国科学院太赫兹固态技术重点实验室，上海 200050;2.中国科学院大学 材料与光电研究中心，北京 100049

中图分类号:

TP751

基金项目:

国家自然科学基金（61731021），中国科学院科技创新重点部署项目（KGFZD-135-18-028），上海市信息化发展专项资金（201901015），上海市科委科研计划项目（19511132400），硅基氮化镓射频及毫米波器件关键工艺技术开发及测试（20DZ1100702）


High localization accuracy 3D object detection in active millimeter wave holographic images
Author:
Affiliation:

1.Key Laboratory of Terahertz Solid State Technology， Shanghai Institute of Microsystem and Information Technology， Chinese Academy of Sciences， Shanghai 200050， China;2.Center of Materials Science and Optoelectronics Engineering， University of Chinese Academy of Sciences， Beijing 100049， China

Fund Project:

Supported by National Natural Science Foundation of China （61731021） ， the Key Research Program of the Chinese Academy of Sciences （KGFZD-135-18-028）， the Shanghai Municipal Commission of Economy and Informatization （201901015）， the Science and Technology Commission of Shanghai Municipality （19511132400），Development and Testing of Key Process Technologies for Silicon-based GaN RF and Millimeter Wave Devices（20DZ1100702）

  • 摘要
    • | |
  • 访问统计
    • |
  • 参考文献
    • |
  • 相似文献
    • |
  • 引证文献
    • | |
  • 文章评论
    摘要:

    投影角度不同导致目标的形状及尺寸变化限制了基于主动式毫米波（AMMW）全息图像投影视图的隐匿物品二维检测方法对小目标检测性能的提升，为此，提出了基于点云的隐匿物品三维检测方法。通过阈值化处理将AMMW全息图像转换为点云输入经空洞卷积及多分支结构改进的SECOND三维目标检测器，提取对目标的三维几何理解及其多尺度上下文信息以提高对小目标的检测能力。实验结果表明，较基于投影的二维检测方法，该方法平均召回率（AR）提升了3.33%，有效提升了定位精度；在交并比（IOU）阈值为0.5时的检出率提升了8.75%，虚警率降低了1.78%，平均精度（AP）提升了7.11%，不同IOU阈值下的平均AP提升了4.30%，有效提升了检测精度；检测速度为17.3 FPS，达实时水平。

    Abstract:

    The difference in projection angle leads to changes in the shape and size of objects， which limits the improvement of the detection performance of small objects by the two-dimensional （2D） concealed object detection method based on projected views of active millimeter wave （AMMW） holographic images. For this reason， a three-dimensional （3D） concealed object detection method based on point clouds was proposed for the first time. AMMW holographic images were converted into point clouds by thresholding， and then were input into the 3D object detector SECOND， which was improved by dilated convolution and multi-branch structure， to extract the 3D geometric understanding of the objects and their multi-scale context information to improve the detection ability for small objects. The experimental results showed that compared with the projection-based 2D detection methods， the average recall （AR） of this method was improved by 3.33%， which proved the effective improvement of localization accuracy. The detection rate and the average precision （AP） was relatively improved by 8.75% and 7.11%， and the false alarm was reduced by 1.78% at an intersection over union （IOU） threshold of 0.5. The average AP under different IOU thresholds was improved by 4.30%. The detection accuracy was effectively improved. The detection speed was 17.3 FPS， which reached the real-time level.

    参考文献
    相似文献
    引证文献
引用本文

李怀乾,杨明辉,吴亮.高定位精度的毫米波全息图像三维目标检测[J].红外与毫米波学报,2021,40(6):870~877]. LI Huai-Qian, YANG Ming-Hui, WU Liang. High localization accuracy 3D object detection in active millimeter wave holographic images[J]. J. Infrared Millim. Waves,2021,40(6):870~877.]

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:2021-01-28
  • 最后修改日期:2021-12-17
  • 录用日期:2021-06-01
  • 在线发布日期: 2021-11-29
文章二维码
您是第 位访问者
主管单位：中国科学院
主办单位：中国科学院上海技术物理研究所，中国光学学会
地址：上海市玉田路500号 电话：021-25051553
51La
红外与毫米波学报 ® 2025 版权所有