Comprehensive Semantic Segmentation on High Resolution UAV Imagery for Natural Disaster Damage Assessment
This work addresses the problem of natural disaster damage assessment for emergency responders and researchers, but is incremental as it applies existing methods to new data.
The authors introduced a large-scale dataset of approximately 2000 high-resolution aerial images from Hurricane Michael for semantic segmentation, and evaluated state-of-the-art deep neural network models on it to assess their performance in recognizing disaster scenarios.
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.