Deep learning approach for catchment detection in Asyut –Egypt
(2022). Deep learning approach for catchment detection in Asyut –Egypt. EKB Journal Management System, 51(1), 21-39. doi: 10.21608/aunj.2022.219648
. "Deep learning approach for catchment detection in Asyut –Egypt". EKB Journal Management System, 51, 1, 2022, 21-39. doi: 10.21608/aunj.2022.219648
(2022). 'Deep learning approach for catchment detection in Asyut –Egypt', EKB Journal Management System, 51(1), pp. 21-39. doi: 10.21608/aunj.2022.219648
Deep learning approach for catchment detection in Asyut –Egypt. EKB Journal Management System, 2022; 51(1): 21-39. doi: 10.21608/aunj.2022.219648

Deep learning approach for catchment detection in Asyut –Egypt

Assiut University Journal of Multidisciplinary Scientific Research
Volume 51, Issue 1, January 2022, Page 21-39  XML PDF (1.23 MB)
Document Type: Novel Research Articles
DOI: 10.21608/aunj.2022.219648
View on SCiNiTO View on SCiNiTO
Abstract
Water resources assessment is an essential element in the sustainable development and management of water resources. It provides a basis for many applications, such as maintenance of projects associated with irrigation and drainage. Catchment detection and identification is one of the water resources assessment fields, especially in dry areas. Few studies have attempted to detect catchments based on DEM, such as the level‐set method based on graph theory. In this work, a deep learning algorithm (DenseNet) was used to detect and locate catchments. Identifying Sink Features in the DEM is the first step. Then, using the level-set process to delineated topographic depressions in DEMs. Finally, Catchments are detected using DenseNet. As the DEM accuracy increase by removing uncertainty from DEM the catchment detection performance increase. Asyut Governorate, Egypt, is used as a study area.
Keywords
Elevation Model (DEM); deep learning; DenseNet; uncertainties; catchments; Geographic Information System(GIS)
Statistics
Article View: 163
PDF Download: 103