PSI - Issue 77

Available online at www.sciencedirect.com Available online at www.sciencedirect.com Available online at www.sciencedirect.com

ScienceDirect

Procedia Structural Integrity 77 (2026) 584–592 Structural Integrity Procedia 00 (2026) 000–000 Structural Integrity Procedia 00 (2026) 000–000

www.elsevier.com / locate / procedia www.elsevier.com / locate / procedia

© 2026 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of ICSI organizers © 2026 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http: // creativecommons.org / licenses / by-nc-nd / 4.0 / ) Peer-review under responsibility of ICSI organizers. Keywords: Neuromorphic; Vibration Analysis; Non-contact Testing Abstract The analysis of vibration is a widely used method for structure or equipment health monitoring, since structural faults may alter its vibration behaviour. Usually performed using contact methods, mainly accelerometers, certain restraints such as structural fragility, mass loading, or accessibility limitations may require alternative approaches. In this study, preliminary tests were con ducted to evaluate event-based (also known as neuromorphic) imaging, comparing it with conventional frame camera approaches. A 120 mm fan served as the specimen, tested while balanced and unbalanced by a small mass. Baseline rotation speed measure ments using a tachometer were 32.5 rotations per second (rps) while balanced and 31.8 rps while unbalanced. The unbalanced fan was tested with an accelerometer, showing a peak near 31 Hz along the motor axis. Hardware included a Prophesee EVK4 neuromorphic camera, a Basler a2A1920-160umBAS frame camera, a Prophesee 8 mm lens, a macro-lens assembly (4.0 × mag nification) and printed targets (striped bands, grid and a printed gradient resembling a speckle pattern). The final configurations were: neuromorphic camera with standard lens and unbalanced fan; accelerometer with unbalanced fan; neuromorphic camera with macro lens and balanced fan; and frame camera with macro lens and balanced fan. Event data showed dominant frequencies at 31.6 Hz (Neuromorphic camera with standard lens and unbalanced fan) and 32.3 Hz (Neuromorphic camera with macro lens and balanced fan), aligning with the 31.0 Hz accelerometer peak and the tachometer baselines, respectively. For the frame camera with macro lens, image tracking and digital image correlation (DIC) both revealed a peak around 33.0 Hz; however, in the DIC analy sis, it was not clearly dominant. Overall, neuromorphic imaging provided a simple setup that returned the dominant frequency in good agreement with frame-based and accelerometer results. Frame tracking o ff ered richer results, but required external triggering and synchronization. This work is a result of Agenda “ATE – Alianc¸a para a Transic¸a˜o Energe´tica”, nr C644914747-00000023, investment project nr 56, financed by the Recovery and Resilience Plan (PRR) and by European Union – NextGeneration EU. © 2026 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http: // creativecommons.org / licenses / by-nc-nd / 4.0 / ) Peer-review under responsibility of ICSI organizers. Keywords: Neuromorphic; Vibration Analysis; Non-contact Testing International Conference on Structural Integrity Non-contact vibration measurement using event-based neuromorphic imaging Francisco Afonso a, ∗ , Pedro J. S. C. P. Sousa a,b , Francisco Sousa b , Susana Dias a , Paulo J. Tavares a , Pedro M. G. J. Moreira a , Cassiano Linhares c , Helena Lopes c a INEGI, Institute of Science and Innovation in Mechanical Engineering and Industrial Engineering, Rua Dr. Roberto Frias 400, Porto 4200-465, Portugal b Departamento de Engenharia Mecaˆnica, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias s / n, Porto 4200-465, Portugal c EFACEC, Parque Empresarial Arroteia Poente, Porto 4466-952, Portugal Abstract The analysis of vibration is a widely used method for structure or equipment health monitoring, since structural faults may alter its vibration behaviour. Usually performed using contact methods, mainly accelerometers, certain restraints such as structural fragility, mass loading, or accessibility limitations may require alternative approaches. In this study, preliminary tests were con ducted to evaluate event-based (also known as neuromorphic) imaging, comparing it with conventional frame camera approaches. A 120 mm fan served as the specimen, tested while balanced and unbalanced by a small mass. Baseline rotation speed measure ments using a tachometer were 32.5 rotations per second (rps) while balanced and 31.8 rps while unbalanced. The unbalanced fan was tested with an accelerometer, showing a peak near 31 Hz along the motor axis. Hardware included a Prophesee EVK4 neuromorphic camera, a Basler a2A1920-160umBAS frame camera, a Prophesee 8 mm lens, a macro-lens assembly (4.0 × mag nification) and printed targets (striped bands, grid and a printed gradient resembling a speckle pattern). The final configurations were: neuromorphic camera with standard lens and unbalanced fan; accelerometer with unbalanced fan; neuromorphic camera with macro lens and balanced fan; and frame camera with macro lens and balanced fan. Event data showed dominant frequencies at 31.6 Hz (Neuromorphic camera with standard lens and unbalanced fan) and 32.3 Hz (Neuromorphic camera with macro lens and balanced fan), aligning with the 31.0 Hz accelerometer peak and the tachometer baselines, respectively. For the frame camera with macro lens, image tracking and digital image correlation (DIC) both revealed a peak around 33.0 Hz; however, in the DIC analy sis, it was not clearly dominant. Overall, neuromorphic imaging provided a simple setup that returned the dominant frequency in good agreement with frame-based and accelerometer results. Frame tracking o ff ered richer results, but required external triggering and synchronization. This work is a result of Agenda “ATE – Alianc¸a para a Transic¸a˜o Energe´tica”, nr C644914747-00000023, investment project nr 56, financed by the Recovery and Resilience Plan (PRR) and by European Union – NextGeneration EU. International Conference on Structural Integrity Non-contact vibration measurement using event-based neuromorphic imaging Francisco Afonso a, ∗ , Pedro J. S. C. P. Sousa a,b , Francisco Sousa b , Susana Dias a , Paulo J. Tavares a , Pedro M. G. J. Moreira a , Cassiano Linhares c , Helena Lopes c a INEGI, Institute of Science and Innovation in Mechanical Engineering and Industrial Engineering, Rua Dr. Roberto Frias 400, Porto 4200-465, Portugal b Departamento de Engenharia Mecaˆnica, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias s / n, Porto 4200-465, Portugal c EFACEC, Parque Empresarial Arroteia Poente, Porto 4466-952, Portugal

∗ Corresponding author E-mail address: fafonso@inegi.up.pt ∗ Corresponding author E-mail address: fafonso@inegi.up.pt

2452-3216 © 2026 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of ICSI organizers 10.1016/j.prostr.2026.01.074 2210-7843 © 2026 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http: // creativecommons.org / licenses / by-nc-nd / 4.0 / ) Peer-review under responsibility of ICSI organizers. 2210-7843 © 2026 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http: // creativecommons.org / licenses / by-nc-nd / 4.0 / ) Peer-review under responsibility of ICSI organizers.