Automated detection of center-pivot irrigation systems from remote sensing imagery using deep learning

Authors: BAZRAFKAN, ALIASGHAR - North Dakota State University; Kim, James; PROULX, ROB - North Dakota State University; LIN, ZHULU - North Dakota State University

Submitted to: Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/30/2025
Publication Date: 7/3/2025
Citation: Bazrafkan, A., Kim, J.Y., Proulx, R., Lin, Z. 2025. Automated detection of center-pivot irrigation systems from remote sensing imagery using deep learning. Remote Sensing. https://doi.org/10.3390/rs17132276.
DOI: https://doi.org/10.3390/rs17132276

Interpretive Summary: Satellites offer imagery of a large area in a fixed time interval from daily to biweekly and thus are useful for agriculture applications in regular assessment of natural resources and crop fields. Groundwater is the primary source of crop production through irrigation, especially in arid and semi-arid areas. Sustaining groundwater supply is essential for long-term food and water security and requires timely monitoring and sustainable management. This research evaluated AI models for detecting center-pivot irrigation systems from various satellite image datasets. The results showed that a trained learning model combined with selected satellite imagery achieved the highest detection accuracy, outperforming other deep learning models and datasets. The study provides a unique contribution to the fields of remote sensing, irrigation, and groundwater management by integrating advanced deep learning techniques into a Geographic Information System-based workflow.

Technical Abstract: Effective detection of center-pivot irrigation systems is crucial in understanding agricultural activity and managing groundwater resources for sustainable uses, especially in semi-arid regions such as North Dakota, where irrigation primarily depends on groundwater resources. In this study, we have adopted YOLOv11 to detect the center-pivot irrigation systems using multiple remote sensing datasets, including Landsat 8, Sentinel-2, and NAIP (National Agriculture Imagery Program). We developed an ArcGIS custom tool to facilitate data preparation and large-scale model execution for YOLOv11, which was not included in the ArcGIS Pro deep learning package. YOLOv11 was compared against other popular deep learning model architectures such as U-Net, Faster R-CNN, and Mask R-CNN. YOLOv11 using Landsat 8 panchromatic data achieved the highest detection accuracy (precision: 0.98, recall: 0.91, and F1 score: 0.94) among all tested datasets and models. Spatial autocorrelation and hotspot analysis revealed systematic prediction errors, suggesting a need to adjust training data regionally. Our research demonstrates the potential of deep learning in combination with GIS-based workflows for large-scale irrigation system analysis, adopting precision agricultural technologies for sustainable water resource management.