{
  "paper_id": "6511aa126e1a2f3c0cdcdef121725327",
  "paper_title": "Stationary Human Detection Method Using 2D LiDAR",
  "paper_authors": [
    "Haruki Mochizuki",
    "Ryozo Kiyohara"
  ],
  "publication_date": "",
  "publication_venue": "2024 IEEE 48th Annual Computers, Software, and Applications Conference (COMPSAC)",
  "abstract": "Since the COVID-19 pandemic, Japanese industry has been facing a labor shortage in various fields. In particular, the security sector is experiencing a serious labor shortage problem. University security operations are also being affected by the lack of security guards. On the other hand, robot technology is developing and it is becoming possible to perform not only simple tasks but also tasks that involve interaction with humans. In particular, robots are expected to be introduced in patrol security, which requires a large number of personnel, as much of the work involves confirming that there are no problems. A security guard should go there only if it cannot be confirmed that there is no problem. Therefore, the accuracy of confirmation and the cost of introducing and operation become issues. The technologies for recognizing human using cameras mounted on 3D-LiDAR (Light Detection And Raging) on robots has been established. However, how to use it in a dark place and recognition technology by the low-cost 2D-LiDAR have not been established. In this paper, we propose a human recognition method using 2D-LiDAR that extends the technologies of previous research, and present the results of our evaluation.\n",
  "keywords": [
    "Human recognition",
    "LiDAR",
    "Robot",
    "Security Guard"
  ],
  "paper_institutions": [
    "Graduate School of Kanagawa Institute of Technology",
    "Kanagawa Institute of Technology"
  ],
  "output_language": "zh",
  "provider": "doubao",
  "model": "deepseek-v4-flash",
  "prompt_template": "simple",
  "extracted_at": "2026-08-18T10:30:17.375054Z",
  "template_tag": "simple",
  "summary": "### 摘要\n\n本文针对日本安保行业（尤其是大学夜间巡逻）因疫情和劳动环境导致的保安人手不足问题，提出一种基于两个低成本2D-LiDAR的静止人体识别方法。已有基于摄像头和3D-LiDAR的人体识别技术虽较成熟，但摄像头在暗处效果差、3D-LiDAR点云数据量大且计算成本高，而低成本的2D-LiDAR在暗处的人体识别尚未确立。本方法在机器人上以95cm和40cm高度安装两个2D-LiDAR，分别对应日本老年人（135–175cm）的躯干/手臂和腿部特征；将2D-LiDAR获取的二维点云附加实际安装高度作为z坐标，构成三维点云，再使用MATLAB发布的PointNet分类工具对点云进行压缩、重构和特征提取，通过监督学习将目标分类为“人”或“树”。实验使用RPLiDAR A1M8（测量半径6m，实际采集1–5m），每个条件下采集150帧点云，并以树木点云作为负样本；训练时仅学习正面和侧面两个方向，利用人体点云在0°与180°等方向上的对称性来覆盖其他角度。结果表明：手臂位置（下垂、举起、胸前交叉、背后交叉）的F值均为1.00；在0°–180°各角度测试中F值为0.95–1.00；在1–5m距离测试中，正面、45°和90°方向的F值大多接近1，仅4m处90°方向F值降至0.82。这些结果支持“学习正面和侧面两个方向即可识别多种角度和距离的静止人体”的核心机制。未来需要验证非直立姿态（如倒地者）的识别精度，并考虑校园夜间巡逻、失智老人误入等真实场景以及5G通信、成本控制等约束。\n\n### 关键要点\n\n- 论文提出使用两个2D-LiDAR（分别安装在95cm和40cm高度）识别静止人体，以应对保安人手不足和夜间巡逻需求。\n- 选择95cm和40cm高度是为了覆盖135–175cm的日本老年人躯干、手臂和腿部特征，因为研究关注失智老人等非恶意人员。\n- 方法将2D-LiDAR获取的二维点云附加实际安装高度作为z坐标，转换为三维点云后输入PointNet进行压缩、重构和特征提取。\n- 使用MATLAB的“Classify point clouds using PointNet”进行监督学习，预先学习人体和树木点云，并将目标分类为人类或树木。\n- 实验采用RPLiDAR A1M8，其测量半径为6m，实际采集距离为1m到5m，每个条件准备150帧点云，树木点云作为错误数据。\n- 手臂位置实验显示，手臂下垂、举起、胸前交叉和背后交叉四种姿态的Recall、Precision和F值均为1.00，说明手臂位置不影响识别。\n- 角度实验训练正面和侧面两个方向后，在0°到180°每10°间隔的测试中F值为0.95–1.00，其中90°、120°、150°略低但仍接近1。\n- 距离实验使用2m处正面和侧面数据训练，在1–5m的正面、45°和90°测试中F值大多接近1，4m处90°方向F值为0.82。\n- 由于人体点云在0°与180°、45°与225°等相对方向形状相同，只需学习0°到180°的一半方向即可识别另一半方向。\n- 论文指出未来需要验证目标非直立状态（如倒地）下的识别精度，因为当前方法可能对非直立人体产生误识别。"
}