Flat roof classification and leaks detections by Deep Learning

Authors

  • David Zahradník CVUT FSv k155 https://orcid.org/0000-0002-5411-3882
  • Filip Roučka
  • Linda Karlovská České vysoké učení technické v Praze, Fakulta stavební, katedra Geomatiky

DOI:

https://doi.org/10.14311/CEJ.2023.04.0042

Keywords:

Thermography, remote sensing, CNN, U-NET, flat roof leak detection

Abstract

This paper presents an efficient and accurate method for detecting flat roof leaks using a combination of unmanned aerial vehicles (UAVs) and deep learning. The proposed method utilizes a DJI M300 drone equipped with RGB and thermal cameras to capture high-resolution images of the roof. These images are then processed to create orthomosaics and digital elevation models (DEMs). A deep learning model based on the U-NET architecture is then used to segment the roof into different classes, such as PVC foil, windows, and sidewalks. Finally, the damaged insulation is identified by analyzing the temperature distribution within the PVC foil segments. The proposed method has several advantages over traditional inspection methods. It is much faster and more efficient. A UAV can collect images of a large roof in a matter of minutes, while traditional methods can take several days or weeks. The orthomosaics and temperature maps generated by the UAV are much more detailed than the images that can be collected by a human inspector. Third, the UAV-based system is safer. The UAV can collect images of the roof without the need for a human inspector to climb onto the roof, which can be dangerous. The results of this study show that the proposed method is an effective and accurate way to detect flat roof leaks. The deep learning model was able to achieve an overall accuracy of 95% in segmenting the roof into different classes. The method was also able to identify damaged insulation with a high degree of accuracy.

Downloads

Download data is not yet available.

References

Antón García, D. y Amaro Mellado, J.L., 2021: Engineering graphics for thermal assessment: 3D thermal data visualisation based on infrared thermography, GIS and 3D point cloud processing software. Symmetry, 13 (2(335)) https://doi.org/10.3390/sym13020335

Jairo Acuña Paz y Miño, Nicolas Duport, Benoit Beckers, 2021:Pixel-by-pixel rectification of urban perspective thermography, Remote Sensing of Environment, Volume 266, 112689, https://doi.org/10.1016/j.rse.2021.112689

Adamopoulos E, Volinia M, Girotto M, Rinaudo F., 2020: Three-Dimensional Thermal Mapping from IRT Images for Rapid Architectural Heritage NDT, Buildings, 10(10):187. https://doi.org/10.3390/buildings10100187

Dlesk, A., Vach, K., and Holubec, P., 2018: USAGE OF PHOTOGRAMMETRIC PROCESSING OF THERMAL IMAGES FOR CIVIL ENGINEERS, Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-5, 99–103, https://doi.org/10.5194/isprs-archives-XLII-5-99-2018

Daffara C, Muradore R, Piccinelli N, Gaburro N, de Rubeis T, Ambrosini D., 2020: A Cost-Effective System for Aerial 3D Thermography of Buildings. Journal of Imaging. 6(8):76. https://doi.org/10.3390/jimaging6080076

Yishuo Huang, Paul Shih, Keng-Tsang Hsu, Chih-Hung Chiang, 2020: To identify the defects illustrated on building facades by employing infrared thermography under shadow,

Huang, Y., Shih, P., Hsu, K. T., & Chiang, C. (2020, April 1). To identify the defects illustrated on building facades by employing infrared thermography under shadow. NDT & E International. https://doi.org/10.1016/j.ndteint.2020.102240

Tarek Rakha, Yasser El Masri, Kaiwen Chen, Eleanna Panagoulia, Pieter De Wilde, 2021: Building envelope anomaly characterization and simulation using drone time-lapse thermography, Energy and Buildings, 111754, ISSN 0378-7788, https://doi.org/10.1016/j.enbuild.2021.111754

Yuan, W., & Hua, W. (2022). A Case Study of Vignetting Nonuniformity in UAV-Based Uncooled Thermal Cameras. Drones, 6(12), 394.

Norhan Bayomi, Shreshth Nagpal, Tarek Rakha, John E. Fernandez , 2021: Building envelope modeling calibration using aerial thermography, Energy and Buildings, Volume 233, 110648, ISSN 0378-7788, https://doi.org/10.1016/j.enbuild.2020.110648.

Dlesk A, Vach K, Pavelka K. 2021: Transformations in the Photogrammetric Co-Processing of Thermal Infrared Images and RGB Images, Sensors. 2021; 21(15):5061. https://doi.org/10.3390/s21155061

Motayyeb, S., Samadzedegan, F., Javan, F. D., & Hosseinpour, H. (2023). Fusion of UAV-based infrared and visible images for thermal leakage map generation of building facades. Heliyon, 9(3)

Stokowiec, K., & Sobura, S. (2022, September). Hand-held and UAV camera comparison in building thermal inspection process. In Journal of Physics: Conference Series (Vol. 2339, No. 1, p. 012017). IOP Publishing.

Kuo, C. F. J., Chen, S. H., & Huang, C. Y. (2023). Automatic detection, classification and localization of defects in large photovoltaic plants using unmanned aerial vehicles (UAV) based infrared (IR) and RGB imaging. Energy Conversion and Management, 276, 116495.

Meng, L., Zhou, J., Liu, S., Wang, Z., Zhang, X., Ding, L., ... & Wang, S. (2022). A robust registration method for UAV thermal infrared and visible images taken by dual-cameras. ISPRS Journal of Photogrammetry and Remote Sensing, 192, 189-214.

Osco, L. P., Marcato, J., Ramos, A. P. M., De Castro Jorge, L. A., Fatholahi, S. N., De Andrade Silva, J., Matsubara, E. T., Pistori, H., Gonçalves, W. N., & Li, J. (2021, October 1). A review on deep learning in UAV remote sensing. International Journal of Applied Earth Observation and Geoinformation. https://doi.org/10.1016/j.jag.2021.102456

Naushad, R., Kaur, T., & Ghaderpour, E. (2021, December 3). Deep Transfer Learning for Land Use and Land Cover Classification: A Comparative Study. Sensors. https://doi.org/10.3390/s21238083

Natesan, S., Armenakis, C., & Vepakomma, U. (2019, June 4). RESNET-BASED TREE SPECIES CLASSIFICATION USING UAV IMAGES. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. https://doi.org/10.5194/isprs-archives-xlii-2-w13-475-2019

Yi-Jun, L., Mohammadi, M. E., & Wood, R. L. (2020, June 22). Deep Learning Classification of 2D Orthomosaic Images and 3D Point Clouds for Post-Event Structural Damage Assessment. Drones. https://doi.org/10.3390/drones4020024

How U-net works? | ArcGIS API for Python. (n.d.). https://developers.arcgis.com/python/guide/how-unet-works/

R. (n.d.). GitHub - reachsumit/deep-unet-for-satellite-image-segmentation: Satellite Imagery Feature Detection with SpaceNet dataset using deep UNet. GitHub. https://github.com/reachsumit/deep-unet-for-satellite-image-segmentation

Reder, S., Mund, J., Albert, N. G., Waßermann, L., & Miranda, L. (2021, December 24). Detection of Windthrown Tree Stems on UAV-Orthomosaics Using U-Net Convolutional Networks. Remote Sensing. https://doi.org/10.3390/rs14010075

Kislov, D. E., & Korznikov, K. A. (2020, April 3). Automatic Windthrow Detection Using Very-High-Resolution Satellite Imagery and Deep Learning. Remote Sensing. https://doi.org/10.3390/rs12071145

Satari, R., Kazimi, B., & Sester, M. (2021, June 4). Extraction of linear structures from digital terrain models using deep learning. AGILE: GIScience Series. https://doi.org/10.5194/agile-giss-2-11-2021

A Deep Learning Model for Identifying Mountain Summits in Digital Elevation Model Data. (2018, September 1). IEEE Conference Publication | IEEE Xplore. https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8527481

Park, G., Lee, M., Jang, H. I., & Kim, C. (2021, May 1). Thermal anomaly detection in walls via CNN-based segmentation. Automation in Construction. https://doi.org/10.1016/j.autcon.2021.103627

Downloads

Published

2024-03-25

Issue

Section

Articles

How to Cite

Flat roof classification and leaks detections by Deep Learning. (2024). Stavební Obzor - Civil Engineering Journal, 32(4). https://doi.org/10.14311/CEJ.2023.04.0042
Received 2024-02-23
Accepted 2024-03-17
Published 2024-03-25