A Deep Transfer Learning Strategy for Identifying Photovoltaic Panels Mounted on Building Rooftops from Orthophotos

Marco A. M. Freitas Junior; Ricardo A. S. Fernandes

doi:10.22456/2175-2745.150976

Authors

Marco A. M. Freitas Junior Universidade de São Paulo (USP) https://orcid.org/0009-0003-6553-7298
Ricardo A. S. Fernandes Universidade de São Paulo (USP) https://orcid.org/0000-0003-2361-6505

DOI:

https://doi.org/10.22456/2175-2745.150976

Keywords:

photovoltaic panels, identification, orthophoto, convolutional neural network

Abstract

This work presents an analysis of orthophotos, which were used to train a predictive model based on a Convolutional Neural Network with the objective of identifying the presence of photovoltaic panels on building rooftops. To this end, the dataset named as Kortviseren, from Denmark, was used, which contains images of rooftops with and without photovoltaic panels. From this dataset, a convolutional neural network ResNet18 was employed, whose synaptic weights were adjusted through transfer learning. During the validation process, it was possible to observe a f1-score of approximately 98,7%. The results demonstrated the feasibility of the proposed approach for identifying photovoltaic panels mounted on building rooftops and, consequently, supporting power distribution utilities with the inspection of these consumer units that have the capacity to produce electricity.

Downloads

Download data is not yet available.

References

[1] MICRO e Minigeração Distribuída: Baterias Atrás do Medidor. https://www.epe.gov.br/sites-pt/publicacoes-dados-abertos/publicacoes/PublicacoesArquivos/publicacao-894/CadernoMMGDBaterias2035v2025.07.31.pdf, 2025.

[2] STRATEGIC PV Analysis and Outreach. https://iea-pvps.org/snapshot-reports/snapshot-2025, 2025. 26 p.

[3] ANUÁRIO Estatístico de Energia Elétrica. https://dashboard.epe.gov.br/apps/anuario-livro/, 2025.

[4] MOHAMMADI, M.; SHARIFI, A. Evaluation of convolutional neural networks for urban mapping using satellite images. Journal of the Indian Society of Remote Sensing, Springer, v. 49, n. 9, p. 2125–2131, 2021.

[5] WANG, C. et al. Tropical cyclone intensity estimation from geostationary satellite imagery using deep convolutional neural networks. IEEE Transactions on Geoscience and Remote Sensing, v. 60, p. 1–16, 2022.

[6] JENA, K. K. et al. Deep convolutional network based machine intelligence model for satellite cloud image classification. Big Data Mining and Analytics, v. 6, n. 1, p. 32–43, 2023.

[7] ILLARIONOVA, S. et al. Estimation of the canopy height model from multispectral satellite imagery with convolutional neural networks. IEEE Access, v. 10, p. 34116–34132, 2022.

[8] POLYMEROPOULOS, I. et al. Enhancing solar plant efficiency: A review of vision-based monitoring and fault detection techniques. Technologies, MDPI, v. 12, n. 10, p. 175, 2024.

[9] VOURKOS, E. G. et al. Improving photovoltaics hotspot and dirt accumulation detection with computer vision: A systematic yolo model comparison based on field aerial images. Sustainable Energy Technologies and Assessments, Elsevier, v. 78, p. 104355, 2025.

[10] WANG, H. et al. Novel cloud-edge collaborative detection technique for detecting defects in pv components, based on transfer learning. Energies, MDPI, v. 15, n. 21, p. 7924, 2022.

[11] HØJSTED, P. Kortviseren: a Danish map portal for topography, culture and history. 2017. ⟨https://kortviseren.dk/⟩. Front-end GIS client based on OpenLayers, displaying WMS services and historical maps.

[12] GU, J. et al. Recent advances in convolutional neural networks. Pattern Recognition, v. 77, p. 354–377, 2018. ISSN 0031-3203. Disponível em: ⟨https://www.sciencedirect.com/science/article/pii/S0031320317304120⟩.

[13] TAYE, M. M. Theoretical understanding of convolutional neural network: Concepts, architectures, applications, future directions. Computation, v. 11, n. 3, 2023. ISSN 2079-3197. Disponível em: ⟨https://www.mdpi.com/2079-3197/11/3/52⟩.

[14] SHARMA, P. et al. A survey on cancer detection via convolutional neural networks: Current challenges and future directions. Neural Networks, v. 169, p. 637–659, 2024. ISSN 0893-6080. Disponível em: ⟨https://www.sciencedirect.com/science/article/pii/S0893608023006287⟩.

[15] KRICHEN, M. Convolutional neural networks: A survey. Computers, v. 12, n. 8, 2023. ISSN 2073-431X. Disponível em: ⟨https://www.mdpi.com/2073-431X/12/8/151⟩.

[16] GALANIS, N.-I. et al. Convolutional neural networks: A roundup and benchmark of their pooling layer variants. Algorithms, v. 15, n. 11, 2022. ISSN 1999-4893. Disponível em: ⟨https://www.mdpi.com/1999-4893/15/11/391⟩.

[17] HU, Z.; ZHANG, J.; GE, Y. Handling vanishing gradient problem using artificial derivative. IEEE Access, v. 9, p. 22371–22377, 2021.

[18] HE, K. et al. Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). [S.l.: s.n.], 2016. p. 770–778.

[19] FENG, Q. et al. When do gans replicate? on the choice of dataset size. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). [S.l.: s.n.], 2021. p. 6701–6710.

[20] BARBEDO, J. G. A. Impact of dataset size and variety on the effectiveness of deep learning and transfer learning for plant disease classification. Computers and Electronics in Agriculture, v. 153, p. 46–53, 2018. ISSN 0168-1699. Disponível em: ⟨https://www.sciencedirect.com/science/article/pii/S0168169918304617⟩.

[21] DENG, J. et al. ImageNet: A Large-Scale Hierarchical Image Database. In: CVPR09. [S.l.: s.n.], 2009.

[22] KASMI, G. et al. A crowdsourced dataset of aerial images with annotated solar photovoltaic arrays and installation metadata. v. 10, n. 1, p. 59, 2023. Disponível em: ⟨https://doi.org/10.1038/s41597⟩.

[23] DESAI, C. Image classification using transfer learning and deep learning. International Journal of Engineering and Computer Science, Valley International, v. 10, n. 9, p. 25394–25398, 2021.

[24] LECUN, Y. A. et al. Efficient backprop. In: Neural Networks: Tricks of the Trade: Second Edition. Berlin, Heidelberg: Springer Berlin Heidelberg, 2012. p. 9–48. ISBN 978-3-642-35289-8. Disponível em: ⟨https://doi.org/10.1007/978-3-642-35289-8_3⟩.

[25] LI, L.; DOROSLOVACKI, M.; LOEW, M. H. Approximating the gradient of cross-entropy loss function. IEEE Access, v. 8, p. 111626–111635, 2020.

[26] KINGMA, D. P.; BA, J. Adam: A Method for Stochastic Optimization. 2017. Disponível em: ⟨https://arxiv.org/abs/1412.6980⟩.