JIHAD EL SANA

High-resolution satellite imagery is vital for applications like urban planning and environmental monitoring, but its high cost limits accessibility. We present a deep learning framework that enhances low-cost, low-resolution remote sensing images using super-resolution techniques based on CNNs, GANs. Our method incorporates spectral attention and physics-informed losses to preserve spatial and spectral integrity. Evaluations on dataset such SpaceNet 2 show significant gains in image quality and GeoAnalytics performance. This approach offers a cost-effective alternative, expanding access to high-quality geospatial data.