LMHLD (Large-scale Multi-source High-resolution Landslide Dataset)
收藏资源简介:
LMHLD是一个大规模多源高分辨率滑坡数据集,由中国地质大学(武汉)未来技术学院构建。该数据集收集了全球七个研究区域的遥感图像,包括中国汶川、巴西里约热内卢、尼泊尔戈尔卡、中国九寨沟、中国台湾、日本北海道和意大利艾米利亚-罗马涅,涵盖了不同触发条件下的多种类型滑坡。数据集包含25365个不同大小的斑块,以适应不同尺度的滑坡检测需求,为基于深度学习的滑坡检测提供了丰富的训练样本。
LMHLD is a large-scale multi-source high-resolution landslide dataset constructed by the School of Future Technology, China University of Geosciences (Wuhan). This dataset collects remote sensing images from seven global study areas, including Wenchuan in China, Rio de Janeiro in Brazil, Gorkha in Nepal, Jiuzhaigou in China, Taiwan of China, Hokkaido in Japan, and Emilia-Romagna in Italy, covering multiple types of landslides under different triggering conditions. It contains 25,365 patches with varying sizes to accommodate the demands of landslide detection at different scales, thus providing abundant training samples for deep learning-based landslide detection.

- 1LMHLD: A Large-scale Multi-source High-resolution Landslide Dataset for Landslide Detection based on Deep Learning中国地质大学(武汉)未来技术学院 · 2025年



