Enhancing Water Body Detection in Satellite Imagery Using U-Net Models

DOI

10.2991/978-94-6463-540-9_87How to use a DOI?

Keywords

Water Body Detection; Satellite Imagery; U-Net

Abstract

Precise and efficient detection of water bodies in satellite pictures is essential for diverse applications, like environmental surveillance, urban development, and disaster response. This study investigates the effectiveness of utilizing the U-shaped network (U-Net) models with input shapes of 128x128 and 256x256 to detect water bodies in satellite photos acquired from the Sentinel-2 Satellite. This research aims to address the dual challenge of recognizing global features in images while also capturing detailed characteristics, such as the boundaries of water bodies. It observes that both models achieve a commendable accuracy of approximately 0.8, accompanied by a modest loss of about 0.3. Notably, the model with a smaller input shape demonstrates a faster convergence during training but exhibits slightly diminished delineation of water body edges compared to its counterpart with a larger input shape. These findings contribute valuable insights into the optimization of water body detection algorithms, offering avenues for both broad-scale previews and fine-scale segmentation in satellite imagery analysis.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Jiongyi Li
PY  - 2024
DA  - 2024/10/16
TI  - Enhancing Water Body Detection in Satellite Imagery Using U-Net Models
BT  - Proceedings of the 2024 2nd International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2024)
PB  - Atlantis Press
SP  - 873
EP  - 881
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-540-9_87
DO  - 10.2991/978-94-6463-540-9_87
ID  - Li2024
ER  -