USO DE TÉCNICAS DE PROCESSAMENTO DE IMAGEM PARA INSPEÇÃO DE ESTRUTURAS DE TELHADOS DE EDIFICAÇÕES PARA FINS DE ASSISTÊNCIA TÉCNICA
DOI:
https://doi.org/10.46421/entac.v18i.1178Keywords:
Roof inspection, Unmanned aerial vehicle, building maintenance and operation, image processingAbstract
The roof is a complex and vital system because it protects the other internal elements of the building from the weather. However, from a technical maintenance viewpoint, this is a structure that is difficult to inspect. In this context, the Unmanned Aerial Vehicle (UAV) can contribute to roof inspections, providing a large amount of data in a short period, and allowing the visualization of pathologies under challenging areas. Also, the use of machine learning and computer vision can contribute to automating the recognition of these pathologies. The objective of this study was to evaluate the use of image processing techniques for roof inspection for technical maintenance purposes. Therefore, training and testing of the image processing algorithms from Microsoft's Custom Vision were performed for the types of nonconformities identified in the roofs, in a base of 1661 images collected with UAV from 61 roofs of buildings in use. The results indicated 72% identification of the non-conformities analyzed, with 11,5% error. Thus, the use of Custom Vision as an image processing tool to identify roof problems for technical maintanance purposes was validatedKeywords: ENTAC2020. Short Paper. Extended Abstract, Publication.
References
CONCEIÇÃO, J.; POÇA, B.; BRITO, J.; FLORES-COLEN, I.; CASTELO, A. Inspection, Diagnosis, and Rehabilitation System for Flat Roofs. Journal Of Performance Of Constructed Facilities, v. 31, n. 6, p. 04017100, 2017.
FANG, W.; DING, L.; ZHONG, B.; LOVE, P.; LUO, H. Automated detection of workers and heavy equipment on construction sites: A convolutional neural network approach. Advanced Engineering Informatics, v. 37, p. 139-149, 2018.
GARCEZ, N.; LOPES, N.; BRITO, J.; SILVESTRE, J. System of inspection, diagnosis and repair of external claddings of pitched roofs. Construction And Building Materials, v. 35, p. 1034-1044, 2012.
IRIZARRY, J.; COSTA, D.B.; KIM, S. Potential applications of Unmanned Aerial systems for construction management tasks. In: International Conference on Innovative Production and Construction, 2015, Perth. Proceedings... Perth: IPC, 2015.
KRAWCZYK, J; MAZUR, A.; SASIN, T.; STOKLOSA, A. Infrared building inspection with unmanned aerial vehicles. Transactions of the Institute of Aviation, v. 240, n. 3, p. 32–48, 2015.
KULKARNI, P.; LONDHE, S.; DEO, M. Artificial neural networks for construction management: a review. Journal of Soft Computing in Civil Engineering, v. 1, n. 2, p. 70-88, 2017.
NATH, N.; BEHZADAN, AMIR H.; PAAL, S. Deep learning for site safety: Real-time detection of personal protective equipment. Automation in Construction, v. 112, p. 103085, 2020.
SILVEIRA, B.; MELO, D.; COSTA, D. Using UAS for Roofs Structure Inspections at PostOccupational Residential Buildings. In: International ICCCBE and CIB W78 Joint Conference on Computing in Civil and Building Engineering 2020 - 18th ICCCBE + 37th CIB W78 2020 (em fase de publicação).
WALTER, A.; LOPES, J.; DE BRITO, J. Flat-roof waterproofing to prevent anomalies on its surface.
In: International Symposium on Building Pathology, Durability and Rehabilitation, 2, 519–528, 2003, Lisboa. Proceedings... Lisboa: CBPDR, 2003 WAZIRI, B.; BALA, K.; BUSTANI, S. Artificial neural networks in construction engineering and management. International Journal Of Architecture, Engineering And Construction, v. 6, n. 1, p. 50-60, 2017.
ZHANG, C. Development of a UAV-based remote sensing system for unpaved road condition assessment’. In: American Society for Photogrammetry and Remote Sensing Annual Conference, p. 820–826, 2008, Portland. Proceedings... Portland: ASPRS, 2008.
