DEVELOPERS: Mr. Dimitrios Loverdos and Dr. Vasilis Sarhosis (University of Leeds, United Kingdom)
A vital aspect when modelling masonry structures is the accuracy in which their geometry is transferred in the numerical model . So far, the geometry of masonry infrastructure is captured with traditional techniques (e.g. visual inspection and manual surveying methods), which are labour intensive and error prone. In the last ten years, advances in remote surveying methods, such as laser scanning and photogrammetry, have started to drastically change the building industry, since such techniques are able to capture rapidly and remotely exact digital records of objects and features in a very short time with excellent accuracy .
Researchers at the University of Leeds were able to use current developments in remote surveying methods and couple them with algorithms developed based on Artificial Intelligence to fully automate the “scan to structural modelling” procedure for the efficient and accurate structural analysis of our ageing masonry infrastructure stock [3, 4]. According to the method, first, images captured from smartphones or DSLR cameras are uploaded into our “Image2DEM” software. Using computer vision and Artificial Intelligence (AI) techniques, we are then able to detect masonry units (e.g. bricks, blocks) and cracks automatically. The geometry of the masonry structure generated, can then be extracted in the form of simplified lines (x, y coordinates) in a DXF format. Finally, DXF files can be used for the UDEC/3DEC geometric model construction with the automatic conversion of geometry into crack commands using our CAD2UDEC software.
This transition from the physical to the digital environment has the potential to gain a better understanding of the “as is” condition of our existing masonry infrastructure and revolutionize the way structural analysis is performed in industry.
Using efficient and accurate estimation of the “as is” structural condition of ageing masonry infrastructure, we are able to provide detailed and accurate data that will better inform maintenance programmes and asset management decisions.
- Kassotakis N., Sarhosis V., Peppa M.V., Mills J. (2021). Quantifying the effect of geometric uncertainty on the structural behaviour of arches developed from direct measurement and Structure-from-Motion (SfM) photogrammetry. Engineering Structures. 230, 111710, https://doi.org/10.1016/j.engstruct.2020.111710
- Kassotakis N., Sarhosis V. (2021). Employing non-contact sensing techniques for improving efficiency and automation in numerical modelling of existing masonry structures: A critical literature review. Structures. 32, 1777-1797. https://doi.org/10.1016/j.istruc.2021.03.111
- Loverdos D., Sarhosis V., Adamopoulos E. Drougkas A. (2021). An innovative image processing-based framework for the numerical modelling of cracked masonry structures. Automation in Construction. 125, 103633. https://doi.org/10.1016/j.autcon.2021.103633
- Dais D., Bal IE., Smyrou E., Sarhosis V. (2021). Automatic crack classification and segmentation on masonry surfaces using convolutional neural networks and transfer learning. Automation in Construction. 125, 103606. https://doi.org/10.1016/j.autcon.2021.103606
- Codes, data and networks relevant to the study can be found in the GitHub repository: https://github.com/dimitrisdais/crack_detection_CNN_masonry