Метод оценки качества супер-разрешения орторектифицированных снимков крон деревьев

Исследование посвящено разработке метода оценки качества супер-разрешения орторектифицированных снимков крон деревьев с использованием сегментирующей нейросети U-Net в качестве критика. Подход позволяет сопоставлять структуры, выделенные на HR- и SR-изображениях, и количественно оценивать корректность реконструкции текстур. Метод демонстрирует согласованность с традиционными метриками и эффективен при анализе сложных объектов лесного покрова.

Earth Resources Observation and Science (EROS) Center. (2020). Landsat 8-9 Operational Land Imager / Thermal Infrared Sensor Level-2, Collection 2 [dataset]. U. S. Geological Survey. https://doi.org/10.5066/P9OGBGM6
Goodfellow I. J., et al. Generative adversarial networks. Advances in neural information pro-cessing systems, 2014, vol. 27, pp. 1–9.
Kingma D. P. Adam: a method for stochastic optimization. 3rd International Conference for Learning Representations. San Diego, 2017.
National Ecological Observatory Network (NEON). High-resolution orthorectified camera imagery mosaic (DP3.30010.001). National Ecological Observatory Network (NEON). National Ecological Observatory Network (NEON), 2024.
Nguyen N. L., et al.  Self-supervised multi-image super-resolution for push-frame satellite images. IEEE / CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). 2021. pр. 1121–1131. https://doi.org/10.1109/CVPRW53098.2021.00123
Rabbi J., et al.  Small-object detection in remote sensing images with end-to-end edge-enhanced GAN and object detector network. Remote Sensing, 2020, vol. 12, pр. 1432.
Sharshov I., et al. Super-resolution of satellite images using Landsat Data. Applied Intelligence. Singapore: Springer Nature, 2025, pр. 83–93.
Shelhamer E., Long J., Darrell T. Fully convolutional networks for semantic segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, vol. 39, no. 4, pp. 640–651. https://doi.org/10.1109/TPAMI.2016.2572683
Wang X., et al. ESRGAN: Enhanced super-resolution generative adversarial networks.  Computer Science (including subseries lecture notes in Artificial Intelligence and lecture notes in Bioinformatics). 2019, vol. 11133 LNCS, pр. 63–79.
Wang Z., et al. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing. 2004, vol. 13, no. 4, pp. 600–612.
Wang Z., Chen J., Hoi S. C. H. Deep Learning for Image Super-Resolution. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, vol. 43, no. 10, pp. 3365–3387.