Automatic Landslide Mapping with Earth Observation Data

利用地球观测数据自动绘制滑坡地图

• 批准号: 2884587
• 金额: --
• 依托单位: University of Nottingham
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2884587
• 关键词: Automatic; Landslide; Mapping; Earth; Observation

This PhD project aims to develop a new approach to rapidly and automatically mapping landslides the novel combination of data from both satellite and social media, with the ambition to routinely use the approach during disaster response activities. Working with researchers at the University of Nottingham and British Geological Survey, the research project will seek to use Artificial Intelligence to extract information on landslide events from news and social media outlets, which will then be used to develop a Deep-Learning method for automatically mapping landslides based on their spectral and topographic characteristics. The successful applicant will then undertake field work in the UK to validate the landslide mapping results and to better understanding possible triggering factors.

该博士学位项目旨在开发一种新的方法，以快速自动地绘制滑坡，从卫星和社交媒体中进行新的数据组合，并在灾难响应活动期间定期使用该方法。该研究项目与诺丁汉大学和英国地质调查局的研究人员合作，将寻求使用人工智能从新闻和社交媒体媒体中提取有关滑坡事件的信息，然后将其用于开发一种根据其光谱和地形特征自动绘制陆板的深度学习方法。然后，成功的申请人将在英国进行现场工作，以验证滑坡映射结果，并更好地理解可能的触发因素。