Scholarlink

<p>Landslides are an important category of geo-hazards which endanger millions of people annually, especially during rainy season. Conventional approaches to monitoring landslides are progressively being trumped by modern methods such as remote sensing and deep learning models. This technology not only supports for landslide identification but also classification. Automatic classification for landslide types has not attended dur to various challenges related to obtaining large, rich data and the overfitting. This study focuses on new process to detect and classify different landslide types using high-resolution remote-sensing images and deep learning models. This study classifies five major types of landslides: (1)&#xa0;Debris flow, (2)&#xa0;Transitional landslides, (3)&#xa0;Rotational landslides, (4)&#xa0;Shallow landslides, and (5) Human-induced landslides. Two architectures including U‑Net and BiSeNet were applied in classifying five types of landslides using WorldView‑2 images with various optimization techniques. These models demonstrate high accuracy for differentiating landslides from non-landslides as well as for classifying landslides into different categories where average accuracy is higher than 95%, loss function values are about 0.02 and other evaluation metrics, like Precision, Recall and F1-score to classify transitional landslides are over 85%. The BiSeNet with Adam optimizer and input size of 256 × 256 were chosen for classify landslides in the Da river basin of Vietnam. The outcomes can enhance landslide identification and categorization to common areas with difficult hilly-spatial situations such as monsoonal and humid subtropical climate zones.</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

Deep Learning for Landslide Classification On Worldview-2 Images in the Monsoonal and Humid Subtropical Climate of the Da River Basin, Vietnam

  • 期刊论文

  • 发表日期：

    2026年1月16日

查看全文
  • Hieu Nguyen,
  • Kinh Bac Dang,
  • Minh Hieu Nguyen,
  • Van Bao Dang,
  • Van Liem Ngo,
  • Trung Hieu Do,
  • Nguyen Vu Dang

摘要

Landslides are an important category of geo-hazards which endanger millions of people annually, especially during rainy season. Conventional approaches to monitoring landslides are progressively being trumped by modern methods such as remote sensing and deep learning models. This technology not only supports for landslide identification but also classification. Automatic classification for landslide types has not attended dur to various challenges related to obtaining large, rich data and the overfitting. This study focuses on new process to detect and classify different landslide types using high-resolution remote-sensing images and deep learning models. This study classifies five major types of landslides: (1) Debris flow, (2) Transitional landslides, (3) Rotational landslides, (4) Shallow landslides, and (5) Human-induced landslides. Two architectures including U‑Net and BiSeNet were applied in classifying five types of landslides using WorldView‑2 images with various optimization techniques. These models demonstrate high accuracy for differentiating landslides from non-landslides as well as for classifying landslides into different categories where average accuracy is higher than 95%, loss function values are about 0.02 and other evaluation metrics, like Precision, Recall and F1-score to classify transitional landslides are over 85%. The BiSeNet with Adam optimizer and input size of 256 × 256 were chosen for classify landslides in the Da river basin of Vietnam. The outcomes can enhance landslide identification and categorization to common areas with difficult hilly-spatial situations such as monsoonal and humid subtropical climate zones.