Building Segmentation on Satellite Images and Performance of Post-Processing Methods

TL;DR

This paper explores building segmentation in satellite images, evaluating models trained in China and tested in Chicago, highlighting the challenges and initial promising results in generalizing across regions.

Contribution

The study trains models in China and assesses their performance on the Chicago dataset, providing initial experimental insights into cross-region building segmentation.

Findings

Models trained in China show promising results when applied to Chicago.

Post-processing improves segmentation accuracy.

State-of-the-art performance has not yet been achieved.

Abstract

Researchers are doing intensive work on satellite images due to the information it contains with the development of computer vision algorithms and the ease of accessibility to satellite images. Building segmentation of satellite images can be used for many potential applications such as city, agricultural, and communication network planning. However, since no dataset exists for every region, the model trained in a region must gain generality. In this study, we trained several models in China and post-processing work was done on the best model selected among them. These models are evaluated in the Chicago region of the INRIA dataset. As can be seen from the results, although state-of-art results in this area have not been achieved, the results are promising. We aim to present our initial experimental results of a building segmentation from satellite images in this study.