Assessing flood severity from georeferenced photos

Abstract

The use of georeferenced social media data in disaster and crisis management is increasing rapidly. Particularly in connection to flooding events, georeferenced images shared by citizens can provide situational awareness to emergency responders, as well as assistance to financial loss assessment, giving information that would otherwise be very hard to collect through conventional sensors or remote sensing products. Moreover, recent advances in computer vision and deep learning can perhaps support the automated analysis of these data. In this paper, focusing on ground-level images taken by humans during flooding events, we evaluate the use of deep convolutional neural networks for (i) discriminating images showing direct evidence of a flood, and (ii) estimating the severity of the flooding event. Considering distinct datasets (i.e., the European Flood 2013 dataset, and data from different editions of the MediaEval Multimedia Satellite Task), we specifically evaluated models based on the DenseNet and EfficientNet neural network architectures, concluding that these models for image classification can achieve a very high accuracy on both tasks. CCS CONCEPTS • Computing methodologies → Computer vision tasks; Neural networks; Visual content-based indexing and retrieval; Supervised learning.

Flood severity mapping from Volunteered Geographic Information by interpreting water level from images containing people: a case study of Hurricane Harvey

“…Automatic interpretation of the water level from crowdsourcing images has been tackled in only a few papers. Pereira et al (2019) classified water severity into three classes, namely no flood, below 1 m, and above 1 m. They used the DenseNet (Huang et al, 2017b) and EfficientNet (Tan and Le, 2019) neural network architectures, where only the global deep features of the whole images are considered. Chaudhary et al (2019) extended the Mask R-CNN model (He et al, 2017) In this paper, we propose a novel method in Section 3.3 that can automatically provide an estimation of water depth based on analyzing images of persons standing in the water.…”

“…Even though modern deep learning technologies can successfully interpret the relevance of the photos or texts to flooding events, the extraction of more detailed severity information from images has been presented in only a few papers (e.g. Chaudhary et al, 2019;Pereira et al, 2019). The extracted information has not yet been used for flood severity mapping.…”