Flood Detection in Gaofen-3 SAR Images via Fully Convolutional Networks.

Wenchao Kang,Hongjian You,Yuming Xiang,Feng Wang,Ling Wan

doi:10.3390/s18092915

Abstract

Emergency flood monitoring and rescue need to first detect flood areas. This paper provides a fast and novel flood detection method and applies it to Gaofen-3 SAR images. The fully convolutional network (FCN), a variant of VGG16, is utilized for flood mapping in this paper. Considering the requirement of flood detection, we fine-tune the model to get higher accuracy results with shorter training time and fewer training samples. Compared with state-of-the-art methods, our proposed algorithm not only gives robust and accurate detection results but also significantly reduces the detection time.

Highlights

Floods frequently occur from June to September in south and northeast China, causing a great deal of economic losses and disaster-induced diseases
For the needs of SAR image flood detection, we fine-tuned the classical fully convolutional network (FCN) in three points
This paper introduces a fully convolutional network FCN16 based on the classical FCN for flood mapping

Summary

Introduction

Floods frequently occur from June to September in south and northeast China, causing a great deal of economic losses and disaster-induced diseases. Commission for Disaster Reduction, there were 43 large-scale heavy rainfall events in 2017, which made 10 provinces in China suffer from heavy flood disasters. With the rapid development of remote sensing sensors, many satellites have been launched, providing overall and continuous land cover information for disaster monitoring and damage assessment. Optical sensors, such as Landsat and SPOT, have been widely used for flood detection and inundation mapping. They are severely limited by meteorological conditions and cannot work at night due to their passive imaging characteristics.

Methods

Results

Conclusion