Robust Remote Sensing Super-Resolution With Frequency Domain Decoupling for Multiscenarios
Sheng Wang, Boxun Han, Linzhe Yang, Chaoyue Zhao, Aokang Liang, Chunying Hu, Feng Yang, Fu Xu · IEEE Transactions on Geoscience and Remote Sensing · 2024
Super-Resolution for remote sensing has the potential for huge impact on diverse applications by reconstructing an accurate high-resolution image from its low-resolution counterpart. Despite a lot of attention, most recent methods concentrate on the overall modeling of the original inputs, treat various frequency information within the image equally, which can lead to distortion of image details or artefacts. To tackle this issue, we proposed a frequency-domain decoupling approach for image super-resolution. It uses nonparametric clustering to obtain the low-frequency contours, applies a linearized Transformer to model the long-range dependency of the mid-frequency details, processes the high-frequency textural with the attention mechanism, and employs Fourier transform to integrate different frequency information effectively. By processing the different frequencies independently, we achieve a more accurate enhancement of the original scene visual quality without adding noise. We conducted evaluations of our method across five real-world scenarios with distinct characteristics. The results show that our model has a significant improvement in peak signal-to-noise ratio of 2.4-5.5 dB over the state-of-the-art method.