Feature-Augmented Deep Networks for Multiscale Building Segmentation in High-Resolution UAV and Satellite Imagery
This work addresses the problem of accurate building segmentation for remote sensing applications, but it is incremental as it builds on existing methods with feature augmentation and training optimizations.
The study tackled building segmentation from high-resolution RGB aerial and satellite imagery by developing a deep learning framework with feature-augmented inputs and optimized training strategies, achieving an overall accuracy of 96.5%, an F1-score of 0.86, and an IoU of 0.80, outperforming existing benchmarks.
Accurate building segmentation from high-resolution RGB imagery remains challenging due to spectral similarity with non-building features, shadows, and irregular building geometries. In this study, we present a comprehensive deep learning framework for multiscale building segmentation using RGB aerial and satellite imagery with spatial resolutions ranging from 0.4m to 2.7m. We curate a diverse, multi-sensor dataset and introduce feature-augmented inputs by deriving secondary representations including Principal Component Analysis (PCA), Visible Difference Vegetation Index (VDVI), Morphological Building Index (MBI), and Sobel edge filters from RGB channels. These features guide a Res-U-Net architecture in learning complex spatial patterns more effectively. We also propose training policies incorporating layer freezing, cyclical learning rates, and SuperConvergence to reduce training time and resource usage. Evaluated on a held-out WorldView-3 image, our model achieves an overall accuracy of 96.5%, an F1-score of 0.86, and an Intersection over Union (IoU) of 0.80, outperforming existing RGB-based benchmarks. This study demonstrates the effectiveness of combining multi-resolution imagery, feature augmentation, and optimized training strategies for robust building segmentation in remote sensing applications.