Comprehensive Semantic Segmentation on High Resolution UAV Imagery for Natural Disaster Damage Assessment

TL;DR

This paper introduces a large-scale high-resolution UAV imagery dataset for hurricane damage assessment and evaluates current deep learning models for semantic segmentation in disaster scenarios.

Contribution

It provides a new extensive dataset and benchmarks state-of-the-art models for semantic segmentation in natural disaster imagery.

Findings

Deep neural networks achieve varying accuracy on the dataset.

Challenges identified for improving semantic segmentation in disaster scenarios.

Future research directions discussed.

Abstract

In this paper, we present a large-scale hurricane Michael dataset for visual perception in disaster scenarios, and analyze state-of-the-art deep neural network models for semantic segmentation. The dataset consists of around 2000 high-resolution aerial images, with annotated ground-truth data for semantic segmentation. We discuss the challenges of the dataset and train the state-of-the-art methods on this dataset to evaluate how well these methods can recognize the disaster situations. Finally, we discuss challenges for future research.