{
  "paper_id": "1805176e41aeb6c73eece265e5d4e465",
  "paper_title": "Person Tracking Using Ankle-Level LiDAR Based on Enhanced DBSCAN and OPTICS",
  "paper_authors": [
    "Mahmudul Hasan",
    "Junichi Hanawa",
    "Riku Goto",
    "Hisato Fukuda",
    "Yoshinori Kuno",
    "Yoshinori Kobayashi"
  ],
  "publication_date": "2021",
  "publication_venue": "Institute of Electrical Engineers of Japan (IEEJ)",
  "abstract": "Along with the progress of deep learning techniques, people tracking using video cameras became easy and accurate. However, privacy and security issues are not enough to be concerned with vision-based monitoring. People may not be tolerated surveillance cameras installed everywhere in our daily life. A camera-based system may not work robustly in unusual situations such as smoke, fogs, or darkness. To cope with these problems, we propose a two-dimensional (2D) LiDAR-based people tracking technique based on clustering algorithms. A LiDAR sensor is a prominent approach for tracking people without disclosing their identity, even under challenging conditions. For tracking people, we propose modified density-based spatial clustering of applications with noise (DBSCAN) and ordering points to identify cluster structure (OPTICS) algorithms for clustering 2D LiDAR data. We have confirmed that our approach significantly improves the accuracy and robustness of people tracking through the experiments. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.",
  "keywords": [
    "person tracking",
    "DBSCAN",
    "OPTICS",
    "density-based clustering",
    "LiDAR"
  ],
  "paper_institutions": [
    "Saitama University",
    "Comilla University"
  ],
  "output_language": "zh",
  "provider": "doubao",
  "model": "ark-code-latest",
  "prompt_template": "simple",
  "extracted_at": "2026-08-16T19:46:37.611787Z",
  "template_tag": "simple",
  "summary": "### 摘要\n本文针对基于视觉的人体跟踪在隐私泄漏和恶劣环境（如烟雾、雾气、黑暗）下鲁棒性不足的问题，提出了一种基于二维LiDAR的踝部级人体跟踪方法。该方法利用HOKUYO UMT30LX LiDAR传感器在20cm高度获取踝部距离数据，生成运动历史图像，并采用改进的DBSCAN（EDBSCAN）和OPTICS（EOPTICS）密度聚类算法进行单步聚类以识别个体，避免了传统两步式踝部分组易混淆的缺点。实验中，作者自建了包含单人行走、单人跑步、仅踝部运动、组合行走和组合跑步五类场景的数据集，在3~9人、44~136帧随机样本上验证了算法。结果显示，EOPTICS在组合行走中达到96.95%的准确率，远超传统方法（DBSCAN 93.71%，OPTICS 93.81%），在组合跑步中达到77.94%的准确率，且算法时间复杂度仍保持O(n²)，内存需求为O(n)。此外，该方法成功解决了多人靠近时的数据过拟合和遮挡问题，并展示了在自然灾难救援、商业服务等实际场景中的应用潜力。核心机制是引入maxPts参数限制簇的最大点数，并在超出时自适应减小半径ε重新聚类，同时EOPTICS通过核心距离和可达距离发现子簇，提高了多密度数据下的聚类精度。\n\n### 关键要点\n- 提出一种使用二维LiDAR传感器在踝部高度进行人体跟踪的方法，以解决视频监控的隐私和恶劣环境适应性问题。\n- 改进DBSCAN算法，增加maxPts参数（如50点），限制簇的最大点数，防止多个人的踝部被错误合并为一个簇。\n- EDBSCAN在簇内点数超过maxPts时自动减小半径ε（如从900单位减小），仅对丢弃的点重新聚类，且minPts和maxPts保持不变。\n- OPTICS算法被增强为EOPTICS，同样引入maxPts参数，当数据点过多时，算法丢弃外层主簇，利用核心距离和可达距离计算内部子簇，以区分不同人的踝部。\n- 实验采用自建数据集，涵盖五种运动类别，传感器高度固定为20cm，扫描周期为25ms，数据经过背景减法预处理。\n- 对于组合步行（9人，131帧），EOPTICS正确识别127帧，准确率96.95%，高于EDBSCAN的93.89%。\n- 对于组合跑步（9人，136帧），EOPTICS正确识别106帧（77.94%），EDBSCAN识别101帧（74.26%），均优于传统DBSCAN（71.87%）和OPTICS（72.11%）。\n- 所提出的增强算法保持了与传统算法相同的O(n²)最坏情况时间复杂度，内存需求为O(n)，适合实时处理。\n- 系统利用Kalman滤波跟踪聚类重心，通过预测和修正方程估计目标轨迹，能处理短期遮挡和异常检测。\n- 与现有的三维点云聚类方法（如Euclidean clustering、Depth clustering、Run clustering等）相比，EOPTICS达到96.9%的最高精度，显著优于此前基于DBSCAN的93.7%。\n- 算法参数（minPts、maxPts、ε）需根据人群密度进行启发式调整，实验中将maxPts从50到300变化，自适应半径策略可减少误检。\n- 本文方法未来可应用于移动机器人实时人体检测、步态识别和护理设施等场景，计划结合DNN进一步学习改进。",
  "available_templates": [
    "deep_read",
    "simple"
  ]
}