A novel framework for automated water level estimation using CCTV imagery in Yongseong Agricultural Reservoir, South Korea

The study region is Yongseong Reservoir, located in Gyeongsangbuk-do, South Korea, a small agricultural reservoir primarily used for irrigation and is subject to pronounced hydrological seasonality. This study proposes a novel framework for estimating water levels in ungauged agricultural reservoirs...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of hydrology. Regional studies Jg. 61; S. 102631
Hauptverfasser: Kwon, Soon Ho, Lim, Suhyun, Lee, Seungyub
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.10.2025
Elsevier
Schlagworte:
Agricultural reservoirs
automation
cameras
CCTV imagery
hydrology
irrigation
Machine-learning regression
monitoring
prediction
Region of interest
regression analysis
South Korea
Water level estimation
South Korea
Water level estimation
Region of interest
Machine-learning regression
Agricultural reservoirs
CCTV imagery
ISSN:2214-5818, 2214-5818
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The study region is Yongseong Reservoir, located in Gyeongsangbuk-do, South Korea, a small agricultural reservoir primarily used for irrigation and is subject to pronounced hydrological seasonality. This study proposes a novel framework for estimating water levels in ungauged agricultural reservoirs using images from CCTVs originally installed for security purposes. The method integrates a U-Net-based water-body segmentation model with four machine learning regression algorithms (support vector regression, SVR; random forest, RF; extreme gradient boosting, XGB; and light gradient boosting machine, LGBM) to predict reservoir water levels from segmented water pixel counts. Importantly, we assess the potential of region of interest (ROI) filtering to enhance prediction accuracy, demonstrating that surveillance camera imagery can be effectively repurposed for hydrological monitoring in data-scarce environments. The results revealed that ROI filtering significantly improved prediction performance, increasing R² by 10–20 % and reducing root mean squared error by up to 0.197 (for RF). The RF model achieved the highest overall accuracy (R² = 0.964), while SVR performed best during no temporal variations. XGB and LGBM showed balanced residuals but slightly underestimated water levels during peak fluctuations. This study demonstrates the feasibility of image-based water-level estimation in ungauged agricultural reservoirs using security CCTVs. The results underscore the importance of spatial input refinement (ROI filtering) for reliable hydrological modeling. [Display omitted] •A framework for water levels estimation using security CCTV imagery was proposed.•U-Net-based water-body segmentation model was integrated with ML models.•ROI filtering demonstrated potential improvement in predictive model performance.•Although RF achieved the highest overall, SVR performed best during smaller variation.•XGB and LGBM slightly underestimated water levels during peak fluctuations.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:2214-5818
2214-5818
DOI:10.1016/j.ejrh.2025.102631