{
  "paper_id": "0d92a5024f76a54e232ad31e33b99e28",
  "paper_title": "Human Detection Method by 3D-LiDAR With Low Calculation Costs",
  "paper_authors": [
    "Haruki Mochizuki",
    "Ryozo Kiyoharara"
  ],
  "publication_date": "",
  "publication_venue": "unknown",
  "abstract": "Robot technologies are 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. In Japan, for example, robots that serve food at restaurants move around in the flow of people. In terms of security, if it cannot be confirmed by robots that there is no problem, a security guard is required to go and check. In order to reduce such cases, improving the accuracy of confirmation is an absolute requirement. In addition, it is necessary to reduce the cost of introducing and operating the robot as much as possible. Technology for recognizing people using cameras mounted on robots and 3D-LiDAR (Light Detection And Raging) has already been established and is in use. However, no technology has been established for use in poorly lit areas or by low-cost 2D-LiDAR recognition. In this paper, we propose a new method for reducing calculation cost and evaluate our method.",
  "keywords": [
    "Human detection",
    "3D-LiDAR",
    "Security robots"
  ],
  "paper_institutions": [
    "Graduate School of Kanagawa Institute of Technology",
    "Kanagawa Institute of Technology"
  ],
  "output_language": "zh",
  "provider": "doubao",
  "model": "ark-code-latest",
  "prompt_template": "simple",
  "extracted_at": "2026-08-16T19:47:08.945394Z",
  "template_tag": "simple",
  "summary": "### 摘要\n本文针对安保机器人使用3D-LiDAR进行人体检测时计算成本过高的问题，提出一种利用部分点云数据模拟多个2D-LiDAR的人体识别方法，以减少数据量和机器学习评估的计算开销。相关研究包括使用图像处理、3D-LiDAR人体识别、将3D-LiDAR当作多个2D-LiDAR等，但均需较多计算资源。本方法基于日本老年人体型特征，选择特定高度范围的激光束（躯干和手臂80-110cm，腿部20-55cm），在距离约25m内确保至少三条激光束获取身体各部分点云。实现上采用VLP-16 3D-LiDAR，通过背景减除提取障碍物，将点云数据无线传输至服务器，使用PointNet进行人类与树木的二分类。训练数据包含正面和侧面两方向的人类点云（各10个）和树木点云（20个），测试数据在不同距离和角度（0°、45°、90°）下采集。在2-5m距离内各角度F值均为1.00；在5-25m距离内，正面在10m和20m为0.91，15m为1.00，侧面在10m为0.00，15m为0.75，20m为0.91，对角线在15m和20m为0.67。结果表明近距离识别准确，远距离稳定性不足，需增加训练数据量。未来将验证服装和姿态对识别的影响，并考虑真实场景中机器人巡逻的夜间等约束。该方法的核心理念是通过选择部分激光束，将3D点云降为2D级别数据，从而大幅降低计算成本。\n\n### 关键要点\n- 提出使用3D-LiDAR中部分激光束模拟多个2D-LiDAR，以减少点云数据量和计算成本。\n- 根据日本老年人平均身高135-175cm，选择躯干和手臂高度80-110cm、腿部20-55cm的激光束。\n- 在距离25m内需至少选择三条激光束以覆盖人体不同部分，距离超过37m则无法获取腿部数据。\n- 实验使用Velodyne VLP-16 3D-LiDAR，垂直视角30°，水平360°，转速5-20Hz。\n- 点云处理流程：机器人采集点云并做背景减除，将提取的障碍物点云发送至服务器，服务器用PointNet进行人与树分类。\n- 训练数据包含正面和侧面姿势的人体点云各10个及树木点云20个，每个训练150个epoch。\n- 测试数据在2-5m距离和0°、45°、90°角度下获取，每个组合16个样本，F值全部为1.00。\n- 在5-25m距离测试中，正面（0°）在10m和20m的F值为0.91，15m为1.00；侧面（90°）在10m为0.00，15m为0.75，20m为0.91；对角线（45°）在15m和20m为0.67。\n- 远距离识别不稳定的原因可能是旋转速度低导致点云数据量少，训练数据不足，以及测试数据量过小导致形状偏差。\n- 本方法在近距离（2-5m）能高精度识别不同朝向的人，验证了降低计算成本的可行性。\n- 未来需增加点云数据量，并研究服装（如夹克、裙子）和人体姿态对识别精度的影响。",
  "available_templates": [
    "deep_read",
    "simple"
  ]
}