Solar energy production has grown significantly in recent years in the European Union (EU), accounting for 12\% of the total in 2022. The growth can be attributed to the increasing adoption of solar photovoltaic (PV) panels, which have become cost-effective and efficient means of energy production, supported by government policies and incentives. The maturity of solar technologies has also led to a decrease in the cost of solar energy, making it more competitive with other energy sources. 

As a result, there is a growing need for efficient methods for detecting and mapping the locations of PV panels. An automated detection can in fact save time and resources compared to manual inspection. Moreover, the resulting information can also be used by governments, environmental agencies and other companies to track the adoption of renewable sources or to optimize the energy distribution across the grid.

However, building such an automated system presents several challenges, including the availability of high-resolution aerial imagery, high-quality, manually-verified labels and annotations, and consequently effective models.  

We address these challenges by constructing a dataset of PV panels using very-high resolution (VHR) aerial imagery, specifically focusing on the region of Piedmont in Italy. The dataset comprises 105 large-scale images, providing more than 9,000 accurate and detailed manual annotations, including a set of attributes such as the PV panel category.

On this dataset, we perform a comprehensive benchmark evaluation of different deep learning techniques, comparing semantic and instance segmentation approaches and making ad-hoc modifications to address specific the issues of this task, such as the wide range of scales of the installations and the sparsity of the annotations.