This project leverages deep learning, specifically the UNet3+ architecture, to estimate high-resolution solar energy potential maps (ASM) using LiDAR-derived digital surface models (DSMs). The approach provides an efficient and scalable alternative to traditional physical models like ArcGIS Pro’s Area Solar Radiation (ASR) model, which is accurate but computationally intensive. The proposed method aims to enhance renewable energy planning by facilitating quicker and more precise identification of rooftops suitable for solar panel installation in urban areas. Study Area The central area of Amsterdam, Netherlands, was selected as the study area. Amsterdam, located in North Holland province in the western part of the country, is situated in the Amstel River delta. The city features low-lying topography with an average elevation of approximately 2 m above sea level, lacking significant elevation changes or steep slopes. Influenced by its proximity to the North Sea, Amsterdam experiences a maritime climate characterized by mild temperatures, high humidity, cool summers, and moderate winters. Despite its flat terrain, the orientation of buildings, streets, and waterways significantly impacts solar exposure and microclimatic conditions, making strategic placement of solar panels critical. This study uses the Digital Surface Model (DSM) derived from the third version of the Actueel Hoogtebestand Nederland (AHN3). The AHN3 DSM: Has a spatial resolution of 0.5 m. Was derived from LiDAR point cloud data without further modifications. Was collected by Publieke Dienstverlening Op de Kaart (PDOK) between 2014 and 2019.AHN3 DSM data is publicly available and offers high-quality raw raster data essential for the deep learning-based solar radiation prediction models. Dataset Input: LiDAR DSMs (0.5 m resolution) from AHN3 to capture urban topographical features. Labels: Annual Solar Energy Potential Maps (ASM) generated by ArcGIS Pro’s ASR model under clear-sky conditions. All data are in GeoTiff format. https://github.com/sinax9696/Enhancing-Solar-Energy-Potential-Mapping-for-Rooftop-Solar-Panel-Placement-Deep-Learning-Approach.git