Improved Land Cover Classification of Aerial Imagery and Satellite Image Time Series using Diffusion-based Super-Resolution

Abstract. Accurate land cover classification requires both spatial details and temporal information of remote sensing data. While publicly available satellite image time series (SITS) offer short revisit times, they suffer from limited spatial resolution. In contrast, aerial imagery provides fine-grained spatial details, but its temporal coverage is limited. Thus, combining data from those sensors is of interest, because their properties are complementary w.r.t. the problem domain. However, the large gap in spatial resolution between these two sensors makes their integration challenging. Generating super-resolution-SITS (SR-SITS) before fusion can help to reduce this gap. In this work, we propose a new approach that integrates diffusion models for generating SR-SITS into a method for the joint pixel-wise classification of aerial and SITS data. Specifically, we employ a diffusion model to generate SRSITS at an intermediate resolution from the raw SITS and aerial imagery of the same observed area. The SR-SITS are temporally encoded and fused with the aerial features using a cross attention module to produce pixel-wise classification at the geometrical resolution of the aerial image. Experimental results on the existing FLAIR benchmark dataset indicate that our approach achieves state-of-the-art results, with a mean Intersection over Union score of 64.0% and an overall accuracy of 76.6%.