Leveraging Self-Supervised Features for Efficient Flooded Region Identification in UAV Aerial Images

TL;DR

This paper demonstrates that self-supervised features from DINOv2, trained on natural images, can effectively identify flooded regions in UAV aerial images, reducing the need for manual annotation and improving segmentation accuracy.

Contribution

It introduces two encoder-decoder segmentation methods that incorporate DINOv2 features, showing their effectiveness for flood detection in UAV images without extensive labeled data.

Findings

DINOv2 features improve segmentation accuracy.

Self-supervised features generalize across image domains.

Reduced manual annotation needed for aerial image segmentation.

Abstract

Identifying regions affected by disasters is a vital step in effectively managing and planning relief and rescue efforts. Unlike the traditional approaches of manually assessing post-disaster damage, analyzing images of Unmanned Aerial Vehicles (UAVs) offers an objective and reliable way to assess the damage. In the past, segmentation techniques have been adopted to identify post-flood damage in UAV aerial images. However, most of these supervised learning approaches rely on manually annotated datasets. Indeed, annotating images is a time-consuming and error-prone task that requires domain expertise. This work focuses on leveraging self-supervised features to accurately identify flooded regions in UAV aerial images. This work proposes two encoder-decoder-based segmentation approaches, which integrate the visual features learned from DINOv2 with the traditional encoder backbone. This study investigates the generalization of self-supervised features for UAV aerial images. Specifically, we evaluate the effectiveness of features from the DINOv2 model, trained on non-aerial images, for segmenting aerial images, noting the distinct perspectives between the two image types. Our results demonstrate that DINOv2's self-supervised pretraining on natural images generates transferable, general-purpose visual features that streamline the development of aerial segmentation workflows. By leveraging these features as a foundation, we significantly reduce reliance on labor-intensive manual annotation processes, enabling high-accuracy segmentation with limited labeled aerial data.