Urban Waterlogging Disaster Level Assessment Method Based on YOLOv8
Linglin Zhang, Xiangsheng Feng, Lei Xiong, Haoxuan Ren, Zhenkun Wang · 2024
In recent years, extreme weather incidents such as heavy rainfall and frequent typhoons have led to the intensification of urban waterlogging disasters, which has constituted a major challenge to urban safety. In order to cope with this challenge, accurate and efficient disaster assessment and early warning technology have become particularly important, and it can provide solid support for disaster prevention and mitigation in cities. Therefore, this article uses the advanced framework of YOLOV8-SEG to develop a set of urban waterlogging disaster evaluation systems. The system uses accurate instance division technology to quickly identify and extract key elements in the affected area, including the scope of water, affected people, and important infrastructure. In addition, this article uses YSPOSE human posture analysis to achieve a quantitative assessment of disaster levels by capturing key points of the skeleton, providing a scientific basis for disaster warning, emergency response and post -disaster reconstruction.