TY - GEN
T1 - Enhanced Remote Sensing Model Performance Through Self-Supervised Learning with Multi-Spectral Data
AU - Hakizimana, Marlyne
AU - Mavis, Emelia
AU - Chiu, Yuting
AU - Malof, Jordan
AU - Bradbury, Kyle
N1 - Publisher Copyright:
© 2024 IEEE.
DBLP License: DBLP's bibliographic metadata records provided through http://dblp.org/ are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.
PY - 2024/7/7
Y1 - 2024/7/7
N2 - Recently, self-supervised learning methods have shown remarkable performance rivaling supervised approaches, particularly in the realm of computer vision. This paper addresses a gap in current literature by focusing on the application of the SwAV (Swapping Assignments between Views) model to pre-train on an extensive dataset comprising one million unlabeled multi-spectral images from Sentinel-2 and Sentinel-1 (including SAR images), we investigate the impact of SSL techniques on multispectral and SAR data as compared to RGB data using crop delineation and land cover classification downstream tasks. Our results demonstrate superior performance exhibited by the 12-channel SwAV pre-trained model compared to RGB-only encodings, underscoring the benefits of SSL for enhancing computer vision for remote sensing applications. Additionally, our findings showcase the potential of SSL for smaller datasets and downstream applications.
AB - Recently, self-supervised learning methods have shown remarkable performance rivaling supervised approaches, particularly in the realm of computer vision. This paper addresses a gap in current literature by focusing on the application of the SwAV (Swapping Assignments between Views) model to pre-train on an extensive dataset comprising one million unlabeled multi-spectral images from Sentinel-2 and Sentinel-1 (including SAR images), we investigate the impact of SSL techniques on multispectral and SAR data as compared to RGB data using crop delineation and land cover classification downstream tasks. Our results demonstrate superior performance exhibited by the 12-channel SwAV pre-trained model compared to RGB-only encodings, underscoring the benefits of SSL for enhancing computer vision for remote sensing applications. Additionally, our findings showcase the potential of SSL for smaller datasets and downstream applications.
KW - machine learning
KW - multi-spectral satellite data
KW - remote sensing
KW - self-supervised learning
KW - SwAV
UR - http://www.scopus.com/inward/record.url?scp=85204874093&partnerID=8YFLogxK
U2 - 10.1109/igarss53475.2024.10640893
DO - 10.1109/igarss53475.2024.10640893
M3 - Conference contribution
AN - SCOPUS:85204874093
T3 - IGARSS 2024 - 2024 IEEE International Geoscience and Remote Sensing Symposium
SP - 2833
EP - 2836
BT - IGARSS 2024 - 2024 IEEE International Geoscience and Remote Sensing Symposium, Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2024 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2024
Y2 - 7 July 2024 through 12 July 2024
ER -