On 25 March at 3:00 p.m., the seminar “Digital Innovation in Structural Monitoring and Modelling” will take place in room J302 at the Instituto Superior de Engenharia do Porto (ISEP).
This session, which will be held in English, will feature presentations by three visiting researchers and PhD students currently collaborating with iBuilt – Center for Innovation in Digital Construction, who will share the latest developments in their research in the fields of computer vision, railway monitoring, and 3D modelling.
Work 1
Measuring the invisible, a motion amplification algorithm for subpixel measurements of structural displacement
Abstract
Dominik Řičánek’s research focuses on the amplification and measurement of microscopic vibrations using video cameras (“motion amplification”), mainly through a phase-based optical flow analysis approach. His work ranges from implementing a camera-based heart-rate monitor to developing convolutional neural networks for amplified displacement measurement. Within this scope, the ongoing research aims to apply motion amplification to complement railway bridge structural health monitoring systems without the need for fixed targets, allowing camera systems mounted on UAVs (drones) to be used for highly precise structural displacement measurements.
Biography
Dominik Řičánek is a PhD student in Electrical Engineering at Brno University of Technology, in the Czech Republic. He is a member of the Machine Vision Group at the Faculty of Electrical Engineering and Communication, focusing on industrial applications of cameras and image processing using both classical algorithms and machine learning.
Work 2
Development of Methodologies for Damage Identification in Railway Bridges Based on Indirect Monitoring and Machine Learning
Abstract
The ageing of railway infrastructure and the increase in rolling loads require robust methods for the early detection of damage in bridges. The research evaluates the potential of indirect (“drive-by”) monitoring as an alternative to traditional visual inspections and the high costs associated with permanent structural health monitoring systems (SHM). The study proposes an automated methodology for damage detection using sensors installed on railway vehicles. The approach applies advanced digital signal processing techniques, such as wavelet transforms, to extract features from the vehicle-structure interaction, combined with unsupervised machine learning algorithms to recognise anomalies, providing a cost-effective and reliable tool for predictive maintenance.
Biography
Cássio S. C. de Bragança holds a degree in Mechanical Engineering and a master’s degree in Structural Engineering from the Federal University of Minas Gerais (UFMG, Brazil). He is currently a PhD student in Civil Engineering at the University of São Paulo (USP) and a visiting researcher at ISEP/iBuilt. With professional experience in acoustics and vibrations, his research focuses on structural health monitoring, digital signal processing, and damage detection in railway systems.
Work 3
Structural Element Segmentation in Scan-to-BIM Frameworks: Streamlining 3D Modelling for Construction Control
Abstract
The increasing complexity of urban environments demands advanced tools for accurate large-scale structural modelling. Although Digital Twins offer a promising paradigm, the transition from raw geospatial data to semantically enriched models still faces bottlenecks in automation. This research addresses these limitations by developing a framework that extends the Scan-to-BIM process from individual buildings to neighbourhood-scale environments. The methodology involves the automated semantic segmentation of structural elements (such as beams and columns), integrating high-resolution 3D data with topological reconstruction to ensure geometric and semantic consistency for optimising the lifecycle of built assets.
Biography
Patricia González Cabaleiro holds a degree in Mechanical Engineering and a master’s degree in Industrial Engineering from the University of Vigo (Spain). She is currently a PhD student and researcher at the Applied Geotechnologies Group (GeoTech) of CINTECX (University of Vigo), where her work is part of the BUILDHOOD project. As a visiting researcher at ISEP/iBuilt, she focuses on the Scan-to-BIM framework, particularly on the automated segmentation of point clouds to support sustainable monitoring and construction control.