EUVIMEDEUVIMED entdecken
Crossref

Spatiotemporal attention‐based adaptive deep learning for remote sensing in soil moisture monitoring

Sampath Gajula, M. Nagarajan, David Solomon Raju Yellampalli, G. Rosline Nesa Kumari

Soil Science Society of America Journal · 2026-11 · 90 (6)

DOI: 10.1002/saj2.70402 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Soil Science Society of America Journal, online erschienen am 20.11.2026. Verfasst von Sampath Gajula und 3 weiteren Autor:innen. Die Arbeit nennt 51 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Remote sensing–based soil moisture monitoring uses spaceborne and airborne sensors, but atmospheric effects and coarse spatial–temporal resolutions limit their ability to capture localized, rapid variations across heterogeneous agricultural landscapes. To resolve these limitations, this research develops an innovative deep learning architecture for dynamic soil moisture mapping, directly from the remote sensing images. In the beginning, remote sensing images are gathered from relevant databases and subsequently input to the developed spatiotemporal attention‐based adaptive graph convolutional neural network combined with a dense bidirectional recurrent neural network (STA‐AGCNN‐DBi‐RNN) to monitor soil moisture. The proposed STA‐AGCNN‐DBi‐RNN model effectively analyze both spatial and temporal dependencies within the input images. Additionally, the proposed model reliability is enhanced by fine‐tuning GCNN‐DBi‐RNN parameters using the improved random parameter–based orangutan optimization algorithm. Thus, the proposed model enables effective soil moisture monitoring, supporting crop management and sustainable water resource planning, with performance evaluated using standard measures. The proposed model achieved 98.42% accuracy and 98.36% precision, demonstrating superior monitoring performance and reliable outcomes for soil moisture monitoring, precision agriculture, and smart irrigation applications.

Zitieren

APA

Gajula, S., Nagarajan, M., Yellampalli, D. S. R., & Kumari, G. R. N. (2026). Spatiotemporal attention‐based adaptive deep learning for remote sensing in soil moisture monitoring. Soil Science Society of America Journal, 90(6). https://doi.org/10.1002/saj2.70402

Vancouver

Gajula S, Nagarajan M, Yellampalli DSR, Kumari GRN. Spatiotemporal attention‐based adaptive deep learning for remote sensing in soil moisture monitoring. Soil Science Society of America Journal. 2026;90(6). doi:10.1002/saj2.70402

BibTeX
@article{Gajula2026,
  author = {Gajula, Sampath and Nagarajan, M. and Yellampalli, David Solomon Raju and Kumari, G. Rosline Nesa},
  title = {Spatiotemporal attention‐based adaptive deep learning for remote sensing in soil moisture monitoring},
  journal = {Soil Science Society of America Journal},
  year = {2026},
  volume = {90},
  number = {6},
  doi = {10.1002/saj2.70402}
}
Kostenlos starten →