Mapping Agricultural Greenhouses in Algeria: A Deep Learning Approach

Agricultural Greenhouses (AGs) are increasingly crucial in global food security, particularly in arid and semi-arid regions like the Maghreb, where they offer sustainable solutions to water scarcity and climate change challenges. Despite their growing importance, the precise extent and distribution of AGs in these regions remain poorly mapped, largely due to the limitations of traditional survey methods and the diverse nature of AG infrastructures. This study introduces a novel, deep learning-based approach to map AGs in the Maghreb, focusing on a 20-year case study in Biskra, Algeria (2000-2024). Employing remote sensing with Landsat imagery and the U-Net deep learning architecture, this research aims to provide a detailed, temporal mapping of AGs, offering insights into their dynamics and informing sustainable agricultural policies in the region.