Mapping stains on flat roofs using semantic segmentation based on deep learning
Source abstract
Abstract
Moisture stains indicate ongoing degradation processes and may reveal areas of the roof slab where water infiltration occurs, compromising the performance and durability of the building system. During inspections of roofing systems, an inspector's field of vision differs from that of drones during overflights. As a result, traditional inspections might not always detect the presence and severity of stains, making maintenance on flat roofs a complex task. In this context, this experimental study aims to analyze deep learning-based semantic segmentation with images obtained from drones to map and monitor damp patches during automated building inspections of flat roof systems. The research tested two convolutional neural networks for semantic segmentation: the Fully Convolutional Network (FCN) with a ResNet50 backbone and DeepLabV3 with a ResNet101 backbone, as well as a transformer-based deep artificial neural network called SegFormer with a MiT-B1 backbone. We evaluated three optimizers for each model—Adam, Adagrad, and SGD—along with learning rates of 1e-2, 1e-3, and 1e-4. The models were compared using four performance metrics. The FCN, optimized with Adagrad at a learning rate of 1e-2, showed the best results. The average metrics obtained in this case were as follows: precision: 79.69 %, recall: 67.81 %, F-score: 73.09 %, and Intersection over Union (IoU): 57.70 %.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Infrastructure Maintenance and Monitoring · 3D Surveying and Cultural Heritage · Remote Sensing and LiDAR Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (42)
Show source-provided references
1. Stains in facades rendering diagnosis and maintenance techniques classification10.1016/j.conbuildmat.2006.08.023 ↗
2. StainView: a fast and reliable method for mapping stains in facades using image classification in hsv and cielab colour space10.3390/rs15112895 ↗
3. Deep learning applied to equipment detection on flat roofs in images captured by UAV
4. Preventive maintenance optimisation of accessible flat roofs in healthcare centres using the Markov chain
5. Web platform for building roof maintenance inspection using UAS and artificial intelligence
6. Damage Detection and Level Classification of Roof Damage after Typhoon Faxai Based on Aerial Photos and Deep Learning10.3390/app12104912 ↗
7. Deep Learning for Detecting Building Defects Using Convolutional Neural Networks10.3390/s19163556 ↗
8. Emotional Maintenance: A Digital Model to Support Maintenance Decisions in Buildings Coatings
9. Software for roof defects recognition on aerial photographs
10. Concrete crack identification using a UAV incorporating hybrid image processing10.3390/s17092052 ↗
11. Deep learning-based structural health monitoring10.1016/j.autcon.2024.105328 ↗
12. Drone-assisted segmentation of tile peeling on building façades using a deep learning model
13. Advances and Challenges in Drone Detection and Classification Techniques: a state-of-the-art review10.3390/s24010125 ↗
14. Distributed Learning and Inference with Compressed Images10.1109/TIP.2021.3058545 ↗
15. Review the state-of-the-art technologies of semantic segmentation based on deep learning10.1016/j.neucom.2022.01.005 ↗
16. Automatic crack classification and segmentation on masonry surfaces using convolutional neural networks and transfer learning10.1016/j.autcon.2021.103606 ↗
17. Unsupervised Semantic defect with pose prior10.1111/mice.13001 ↗
18. A Brief Survey on Semantic Segmentation with Deep Learning10.1016/j.neucom.2019.11.118 ↗
19. A comparative study of LightGBM, XGBoost, and GEP models in shear strength management of SFRC-SBWS10.1016/j.istruc.2024.106009 ↗
20. A metric for assessing and optimizing data-driven prognostic algorithms for predictive maintenance10.1016/j.ress.2023.109723 ↗
21. Deep learning in the construction industry: A review of present status and future innovations
22. A systematic literature review on modern methods of construction in building: An integrated approach using machine learning
23. A comparative study of shear strength prediction models for SFRC deep beams without stirrups using Machine learning algorithms10.1016/j.istruc.2023.06.026 ↗
24. Slope displacement detection in construction: an automated management algorithm for disaster prevention10.1016/j.eswa.2023.121505 ↗
25. Machine learning in construction: from shallow to deep learning
26. Comparison of Four Convolutional Neural Network-Based Algorithms for Sports Image Classification
27. Fully convolutional networks for semantic segmentation
28. Rethinking Atrous Convolution for Semantic Image Segmentation
29. Attention Is All You Need
30. SegFormer: Simple and efficient design for semantic segmentation with transformers
31. Optimizing neural network algorithms for submerged membrane bioreactor: A comparative study of OVAT and RSM hyperparameter optimization techniques10.2166/wst.2024.099 ↗
32. DÀ-JIĀNG Innovations Science and Technology - DJI. DJI Fly. Available at: 〈https://www.dji.com/br/downloads/djiapp/dji-fly〉. Accessed on: Dezember 8, 2023.
33. ROBOFLOW (2023). Give your software the sense of sight. Available at: 〈https://roboflow.com/〉. Accessed on: Dezember 8, https://doi.org/10.1016/j.neucom.2022.01.005.10.1016/j.neucom.2022.01.005 ↗
34. X. Glorot, Y. Bengio. Understanding the difficulty of training deep feedforward neural networks. International Conference on Artificial Intelligence and Statistics, 13th AISTATS. Proceedings…, 9 of JMLR: W&CP 9. Italy, 2010. Available at: 〈https://proceedings.mlr.press/v9/glorot10a/glorot10a.pdf〉. Accessed on: May 15, 2024.
35. Image segmentation review: Theoretical background and recent advances10.1016/j.inffus.2024.102608 ↗
36. A Novel Approach to Optimizing Convolutional Neural Networks for Improved Digital Image Segmentation
37. Evaluation metrics and statistical tests for machine learning10.1038/s41598-024-56706-x ↗
38. A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, et al., An image is worth 16x16 words: Transformers for image recognition at scale. ICLR 2021 arXiv:2010.11929v2 [cs.CV], (2021), 01-21, https://doi.org/10.48550/arXiv.2010.11929.
39. Semantic segmentation using Vision Transformers: A survey10.1016/j.engappai.2023.106669 ↗
40. Predictive Modeling of Above-Ground Biomass in Brachiaria Pastures from Satellite and UAV Imagery Using Machine Learning Approaches10.3390/rs14225870 ↗
41. Effects of the Spatial Resolution of UAV Images on the Prediction and Transferability of Nitrogen Content Model for Winter Wheat10.3390/drones6100299 ↗
42. Automatic detection of coseismic landslides using a new transformer method10.3390/rs14122884 ↗