Enhancing Building Semantic Segmentation Accuracy with Super Resolution and Deep Learning: Investigating the Impact of Spatial Resolution on Various Datasets

TL;DR

This study investigates how spatial resolution affects building semantic segmentation accuracy using deep learning models, highlighting the importance of selecting cost-effective resolutions around 0.3m for remote sensing data.

Contribution

It provides a comprehensive analysis of the impact of spatial resolution on segmentation performance across multiple datasets using super-resolution and deep learning models.

Findings

Spatial resolution significantly influences segmentation accuracy.

Optimal cost-effective resolution identified around 0.3 meters.

Deep learning models perform better with higher spatial resolutions.

Abstract

The development of remote sensing and deep learning techniques has enabled building semantic segmentation with high accuracy and efficiency. Despite their success in different tasks, the discussions on the impact of spatial resolution on deep learning based building semantic segmentation are quite inadequate, which makes choosing a higher cost-effective data source a big challenge. To address the issue mentioned above, in this study, we create remote sensing images among three study areas into multiple spatial resolutions by super-resolution and down-sampling. After that, two representative deep learning architectures: UNet and FPN, are selected for model training and testing. The experimental results obtained from three cities with two deep learning models indicate that the spatial resolution greatly influences building segmentation results, and with a better cost-effectiveness around 0.3m, which we believe will be an important insight for data selection and preparation.