Spatial Super Resolution of Digital Surface Models Generated from Point Clouds by SRGAN Deep Neural Network
DOI:
https://doi.org/10.11137/1982-3908_2025_48_64837Keywords:
Generative Adversarial Network, Image Super Resolution, Machine LearningAbstract
Digital Surface Models (DSM) are responsible for providing altimetric information on a surface to be mapped. This work addresses
the viability of the utilization of deep learning algorithms coupled with Single Image Super Resolution (SISR) techniques in DSM
to obtain better spatial quality versions from lower resolution inputs. The development of a Generative Adversarial Network (GAN)-
based methodology enables this improvement of the initial spatial resolution of low resolution images. A dataset with different pairs of
DSM was created with the objective of allowing the study to be carried out, promoting the emergence of new research groups in the
area as well as enabling the comparison between the results obtained. It has been found that by increasing the number of iterations the performance of the generated model was improved and the quality of the generated image increased. Furthermore, the visual analysis of the generated image against the high and low resolution ones showed a great similarity between the first two.
Downloads
References
Chaplot, V., Darboux, F., Bourennane, H., Leguédóis, S., Silveirra, N. & Phacomphon, K. 2006, 'Accuracy of interpolation techniques for the derivation of digital elevation models in relation to landform types and data density', Geomorphology, vol. 77, no. 1, pp. 126-141.
Chaudhuri, S. 2001, Super-resolution imaging, Kluwer Academic Publishers.
Cruz, C.B.M., Barros, R.S., Cardoso, F.V., Reis, F.B., Rosário, L.S., Barbosa, S.S., Rabaco, L.M.L. & Lourenço, J.S.Q. 2011, 'Avaliação da exatidão planialtimétrica dos modelos digitais de superfície (MDS) e do terreno (MDT) obtidos através do LIDAR', Proceedings of XV Simpósio Brasileiro de Sensoriamento Remoto (SBSR), INPE, Curitiba, pp. 5463-5470
Dhakal, B. 2021, 'Diving into different GAN architectures', Towards Data Science, viewed 28 Jul. 2025, https://towardsdatascience.com/diving-into-different-gan-architectures-a96d05c03c5c.
Gaidhani, P. 2020, 'Super-resolution', Education Information UK, CVonline, viewed 28 Jul. 2025, https://homepages.inf.ed.ac.uk/rbf/CVonline/LOCAL_COPIES/GAIDHANI1/super.htm.
Goodfellow, I.J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A. & Bengio, Y. 2014, 'Generative adversarial nets', Advances in Neural Information Processing Systems, vol. 27, DOI:10.1145/3422622
Ledig, C., Theis, L., Husz, F., Caballero, J., Cunningham, A., Acosta, A., Aitken, A., Tejani, A., Totz, J., Wang, Z. & Shi, W. 2017, 'Photo-realistic single image super-resolution using a generative adversarial network', Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, DOI: 10.48550/arXiv.1609.04802
Lunardi, O.A., Penha, A.L.T. & Cerqueira, W. 2012, 'O Exército Brasileiro e os Padrões de Dados Geoespaciais para a INDE', Proceedings of IV Simpósio Brasileiro de Ciências Geodésicas e Tecnologias da Geoinformação.
Moreira, L.A. 2022, Aumento de resolução de superfícies com informações altimétricas com a rede neural profunda DEM-ESRGAN, PhD Thesis, Universidade Federal do Paraná, Curitiba, 271 p.
Paparoditis, N. & Polidori, L. 2002, 'DSM quality: internal and external validation', in Y., Egels & M., Kasser (eds), Digital Photogrammetry, Taylor & Francis, New York, pp. 225-240.
Simões, M.G. 1993, Modeladores digitais de terreno em sistemas de informação geográfica, Universidade Federal do Rio de Janeiro - UFRJ.
Sun, G., Chen, Y., Huang, J., Ma, Q. & Ge, Y. 2024, 'Digital surface model super-resolution by integrating high-resolution remote sensing imagery using generative adversarial networks', IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 17, pp. 10636-10647, DOI: 10.1109/JSTARS.2024.3399544
Wagner, W., Ullrich, A., Melzer, T., Briese, C. & Kraus, K. 2004, 'From single-pulse to full-waveform airborne laser scanners: potential and practical challenges', ISPRS Journal of Photogrammetry and Remote Sensing, vol. 35, no. Part B.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This journal is licensed under a Creative Commons — Attribution 4.0 International — CC BY 4.0, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
