Scholarlink

<p>Millions of landslides are triggered annually by the destabilization of soil and rock bodies, causing significant economic losses and human casualties. Detecting slopes has become a critical focus. This paper proposes a systematic method for identifying, segmenting, and analyzing landslide areas in 3D point clouds by automatically determining DBSCAN parameters using the silhouette coefficient. The iterative maximum silhouette coefficient method determines the optimal DBSCAN parameters, addressing the limitation of manual parameter setting. Once the optimal parameters are identified, they are used to recognize and segment the landslide zone. Finally, the discontinuity surface of the landslide zone is extracted, and its stability is analyzed. Experimental results demonstrate that the proposed method effectively addresses the limitations of traditional manual slope exploration and the challenges of manually selecting DBSCAN parameters, making it suitable for detecting landslide areas of slopes in diverse conditions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Detection and automatic identification of landslide areas from the LiDAR point clouds using improved DBSCAN

  • 期刊论文

  • 发表日期：

    2025年7月22日

查看全文
  • Jiaqian Ma,
  • Bin Luo,
  • Yan Zhao,
  • Shanwei Li,
  • Mo Chen,
  • Jian Wei,
  • Shigang Wang,
  • Xinyu Zhang

摘要

Millions of landslides are triggered annually by the destabilization of soil and rock bodies, causing significant economic losses and human casualties. Detecting slopes has become a critical focus. This paper proposes a systematic method for identifying, segmenting, and analyzing landslide areas in 3D point clouds by automatically determining DBSCAN parameters using the silhouette coefficient. The iterative maximum silhouette coefficient method determines the optimal DBSCAN parameters, addressing the limitation of manual parameter setting. Once the optimal parameters are identified, they are used to recognize and segment the landslide zone. Finally, the discontinuity surface of the landslide zone is extracted, and its stability is analyzed. Experimental results demonstrate that the proposed method effectively addresses the limitations of traditional manual slope exploration and the challenges of manually selecting DBSCAN parameters, making it suitable for detecting landslide areas of slopes in diverse conditions.