Skip to main navigation Skip to search Skip to main content

Farmland Change Monitoring Using Multi-Temporal SAR Data with Deep Learning

  • Chaowei Jiang
  • , Chao Wang*
  • , Fan Wu
  • , Lu Xu
  • , Nan Chen
  • , Tianyang Li
  • , Yixian Tang
  • *Corresponding author for this work
  • CAS - Aerospace Information Research Institute
  • International Research Center of Big Data for Sustainable Development Goals
  • University of Chinese Academy of Sciences

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Accurate farmland change monitoring is crucial for achieving global 'Zero Hunger' under Sustainable Development Goal 2 (SDG2), proposed by the United Nations (UN). We propose a novel deep learning framework combining multi-temporal synthetic aperture radar (SAR) data with change detection techniques. SAR's all-weather imaging capabilities overcome optical data limitations. The Unpaired Image Augmentation (UIA) boosts training diversity and significantly improves detection metrics, achieving up to 95.29 % Accuracy. The application in Jinan, China, highlights robust performance and scalability across multiple frames. The approach cost-effectively identifies subtle farmland conversions in near-real time, aiding sustainable agriculture and informed policymaking.

Original languageEnglish
Title of host publicationIGARSS 2025 - 2025 IEEE International Geoscience and Remote Sensing Symposium, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages304-307
Number of pages4
ISBN (Electronic)9798331508104
DOIs
StatePublished - 2025
Externally publishedYes
Event2025 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2025 - Brisbane, Australia
Duration: 3 Aug 20258 Aug 2025

Publication series

NameInternational Geoscience and Remote Sensing Symposium (IGARSS)
ISSN (Print)2153-6996
ISSN (Electronic)2153-7003

Conference

Conference2025 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2025
Country/TerritoryAustralia
CityBrisbane
Period3/08/258/08/25

Keywords

  • Change Detection
  • Deep Learning
  • Farmland
  • Synthetic Aperture Radar

Fingerprint

Dive into the research topics of 'Farmland Change Monitoring Using Multi-Temporal SAR Data with Deep Learning'. Together they form a unique fingerprint.

Cite this