Towards Transformer-based Homogenization of Satellite Imagery for Landsat-8 and Sentinel-2
This is an incremental improvement for remote sensing applications, addressing incompatibility issues between satellite sensors to enhance temporal data coverage.
This work tackled the problem of combining Landsat-8 and Sentinel-2 satellite imagery to increase cloud-free data availability by reducing spectral and spatial differences, finding that a UNet model significantly outperformed a transformer-based approach in experiments.
Landsat-8 (NASA) and Sentinel-2 (ESA) are two prominent multi-spectral imaging satellite projects that provide publicly available data. The multi-spectral imaging sensors of the satellites capture images of the earth's surface in the visible and infrared region of the electromagnetic spectrum. Since the majority of the earth's surface is constantly covered with clouds, which are not transparent at these wavelengths, many images do not provide much information. To increase the temporal availability of cloud-free images of a certain area, one can combine the observations from multiple sources. However, the sensors of satellites might differ in their properties, making the images incompatible. This work provides a first glance at the possibility of using a transformer-based model to reduce the spectral and spatial differences between observations from both satellite projects. We compare the results to a model based on a fully convolutional UNet architecture. Somewhat surprisingly, we find that, while deep models outperform classical approaches, the UNet significantly outperforms the transformer in our experiments.