Satellite imagery super-resolution using GANs and aerial images

Detection-Oriented Super-Resolution for Cost-Efficient Aerial Monitoring

From Aerial to Satellite: Can Super-Resolution Enable Label-Free Model Transfer?

Abstract. Satellite imagery enables large-scale remote sensing applications by providing frequent and large-scale coverage. However, its limited spatial resolution often restricts the use of satellite images in tasks that require detailed, fine-scale information. In contrast, aerial images offer a much higher spatial resolution, allowing the extraction of fine-grained features, but typically cover smaller, more localized areas. In this work, we investigate whether super-resolution (SR) methods can bridge the gap between aerial and high-resolution satellite imagery, enabling a label-free model transfer, meaning without fine-tuning our model with additional manual annotations. The idea is to enhance the spatial resolution of high-resolution satellite images, allowing models trained on aerial data to be directly applied to satellite images. Towards this goal, a state-of-the-art SR algorithm is used to upscale three high-resolution satellite images, matching the resolution of the aerial training data. Then, a segmentation network trained on an aerial image dataset is applied to segment roads and parking areas in the super-resolved satellite images. The approach is evaluated on an annotated dataset and compared to the results in the original satellite images. Additionally, we investigate its performance on a low-resolution aerial image. Our results demonstrate that SR facilitates the utilization of models trained on aerial image datasets for large-scale satellite applications without requiring new labels.

From Super-Resolution to Superior Land Cover Detection: Cross-Channel Attention Network for Aerial Image

Abstract. Low-resolution imagery is a major constraint for remote sensing tasks (e.g., urban land cover detection) where accurate classification of buildings, roads, vegetation, and small objects is required. Deep learning-based segmentation models are highly sensitive to image quality, resulting in degraded performance on low-resolution inputs. Super-resolution (SR) techniques offer a promising solution by enhancing image fidelity to support downstream tasks. This work applied MAPSRNet, a Multi-Attention Pyramid SR Network to aerial images used for multi-class land cover detection. Evaluated on the ISPRS Potsdam dataset, MAPSRNet achieves state-of-the-art SR performance with PSNR of 32.92 dB and SSIM of 0.87, outperforming existing methods such as SRCNN (31.54 dB, 0.83) and DRRN (31.03 dB, 0.82) while maintaining competitive inference speed. Beyond image quality, MAPSRNet significantly improves multi-class land cover segmentation when integrated with a ConvNeXtV2-based U-Net, achieving an overall accuracy of 80.60%, mean IoU of 62.54%, and FwIoU of 68.34%, surpassing not only low-resolution inputs (Overall Accuracy: 65.28%, mIoU: 40.20%, FwIoU: 50.12%) but also high-resolution(HR) ones (Overall Accuracy: 80.50%, mIoU: 62.40%, FwIoU: 68.01%), especially in certain classes such as impervious surface and clutter. These results demonstrate that perceptual and structural fidelity, rather than pixel-level similarity, can drive superior performance in urban land cover segmentation. MAPSRNet offers a practical solution for scenarios where HR imagery is limited or unavailable, highlighting its potential for large-scale remote sensing applications.

Comparative Study of Edge Losses for Remote Sensing Image Super-Resolution

Abstract. Image super-resolution (SR) techniques have achieved significant performance improvements with the advancement of deep learning. Accordingly, deep learning-based SR methods have become the mainstream approach in SR research and are widely applied across various fields, including remote sensing. However, most state-of-the-art SR studies are primarily driven by computer vision research and tend to focus on generating visually realistic images rather than preserving structural fidelity with respect to the input images. In remote sensing applications, maintaining structural fidelity is particularly important because SR outputs are often used in downstream analytical tasks such as object detection.In this study, we investigate the use of edge loss to enhance the structural fidelity of SR images for remote sensing imagery. The effectiveness of edge loss was evaluated using multiple benchmark datasets on both convolutional neural network (CNN)- and generative adversarial network (GAN)-based SR models. Several representative SR network architectures and GAN training frameworks were employed to assess the impact of integrating edge loss into the training objective. The experimental results demonstrate that incorporating edge loss improves both the structural fidelity and perceptual quality of SR images. Among the evaluated edge operators, the Prewitt-based edge loss showed the most consistent improvements compared with the Sobel- and Laplacian-based edge losses. These results indicate that edge loss is an effective and easily implementable strategy for improving SR reconstruction quality in remote sensing imagery. Furthermore, it can be combined with other edge-aware techniques to further enhance perceptual quality.

Deep Learning-Based Compression Artifact Removal and Super Resolution for Aerial Imagery

High-resolution aerial imagery plays a significant role in many fields such as urban planning, environmental monitoring, weather prediction, disaster management, change detection and map generation. However, acquiring high-resolution data is often limited by sensor capacities and cost constraints. Moreover, in aerial remote sensing platforms, raw images are compressed during downlink transmission to reduce bandwidth requirements, energy consumption and storage capacity. Lossy image compression algorithms including JPEG degrade image quality, causing artifacts such as blurring, blocking and ringing. In this work, we propose a two-stage framework that employs FBCNN for artifact removal and ESRGAN for super-resolution reconstruction in aerial imagery. We construct our test set based on the SODA-A dataset. Experimental results show that FBCNN achieves high PSNR, SSIM and PSNR-B values across different JPEG quality factors, with PSNR ranging from 28.63 dB to 35.09 dB, SSIM from 0.769 to 0.930, and PSNR-B from 28.40 dB to 34.23 dB. Building upon the outputs of FBCNN, ESRGAN further enhances perceptual quality while maintaining strong quantitative performance, achieving PSNR values from 25.30 dB to 27.99 dB and SSIM values from 0.610 to 0.696 across different quality factors.

Noise-Aware Data Augmentation for Robust Road Detection in Small Satellite Imagery

Abstract. Road extraction from small satellite imagery is challenging because raw images often suffer from low signal-to-noise ratio (SNR), high radiometric variability, and reduced sharpness. In this work, we investigate whether noise and blur data augmentation during pretraining can improve robustness in such conditions. We use a two-stage transfer-learning framework in which a U-Net with a ResNet-50 encoder is pretrained on PlanetScope RGB imagery and fine-tuned on NEMO-HD imagery. During pretraining, we evaluate Gaussian, ISO-like, and Perlin noise, as well as Gaussian and motion blur, each at three severity levels. On the internal held-out test split, augmentation effects were modest, with the best strict IoU improving from 26.7% for the geometric-only baseline to 27.2%. However, evaluation on external full-scene NEMO-HD images showed clearer benefits. Augmentation-based models consistently improved road detection in raw imagery, mainly by increasing completeness and recall, while there was little or no systematic benefit in stacked imagery. No clear trend was observed across augmentation severity levels, indicating that performance depended more on scene conditions than on perturbation strength. The results show that augmentation is most useful for single-acquisition small satellite imagery, where it improves robustness to lower image quality.