Learning-based video motion magnification approach for vibration-based damage detection

Lado-Roigé, Ricard; Font-Moré, Josep; Pérez, Marco A.

doi:10.1016/j.measurement.2022.112218

Cited by 15 publications

(6 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…VMM is computationally intensive; therefore, at the moment only short videos are suitable for processing without specialized workstations. However, this limitation could be overcome by using VMM based on machine learning methods that are currently under development [ 18 ]. Moreover, it seems rational to repeat measurements during an infusion test, during which significant changes in ICP occur over a short period of time.…”

Section: Discussionmentioning

confidence: 99%

Applying video motion magnification to reveal spontaneous tympanic membrane displacement as an indirect measure of intracranial pressure in patients with brain pathologies

et al. 2023

View full text Add to dashboard Cite

Background The observation of tympanic membrane displacement (TMD) opens up the possibility of indirect intracranial pressure (ICP) estimation. In this study, we applied a phase-based video motion magnification (VMM) algorithm to reveal spontaneous pulse TMD waveforms (spTMD) and compare them with invasively measured ICP in patients with intracranial pathologies. Methods Nine adults (six traumatic brain injury and three aneurysmal subarachnoid haemorrhage; median age 44 (29–53) years admitted to the intensive care unit of Wroclaw Medical University between October 2021 and October 2022 with implanted ICP sensors were included in this retrospective study. Video recordings of the tympanic membrane were performed using a portable otoscope with a video camera and analysed by a custom-written VMM algorithm. ICP was monitored using intraparenchymal sensors and arterial blood pressure (ABP) was measured in the radial arterial lines. ICP, ABP, and spTMD videos were captured simultaneously. The pulse amplitudes of ICP (Amp_ICP), ABP (Amp_ABP) and spTMD (Amp_spTMD) were estimated using fast Fourier transform within the heart rate (HR)–related frequency range. Results Amp_spTMD was significantly correlated with mean ICP (rS = 0.73; p = 0.025) and with Amp_ICP (rS = 0.88; p = 0.002). Age was not a significant moderator of this association. There were no significant relationships between Amp_spTMD and either mean ABP, HR, or Amp_ABP. Conclusions The study suggests that Amp_spTMD increases with the increase in mean ICP and Amp_ICP. Estimation of Amp_spTMD using the VMM algorithm has the potential to allow for non-invasive detection of the risk of elevated ICP; however, further investigation in a larger group of patients is required.

show abstract

Section: Discussionmentioning

confidence: 99%

Applying video motion magnification to reveal spontaneous tympanic membrane displacement as an indirect measure of intracranial pressure in patients with brain pathologies

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Nevertheless, the better edge stability offered by STB-VMM enables the authors to obtain better frequency readings from the magnified video. Such application is interesting on technical fields where vibration needs to be monitored, such as in structural health monitoring [22]. Figure 6 shows the cropped upper right corner of the building [51] and the slice used for frequency measuring.…”

Section: Qualitative Comparisonmentioning

confidence: 99%

“…The transformer-based model offers less blurry frame reconstruction, better noise tolerance, and fewer artifacts than prior-art. These improvements can benefit applications that obtain numerical data from magnified video such as Structural Health Monitoring (SHM) applications [22].…”

Section: Introductionmentioning

confidence: 99%

STB-VMM: Swin Transformer Based Video Motion Magnification

Lado-Roigé¹,

Pérez²

2023

Preprint

View full text Add to dashboard Cite

The goal of video motion magnification techniques is to magnify small motions in a video to reveal previously invisible or unseen movement. Its uses extend from bio-medical applications and deep fake detection to structural modal analysis and predictive maintenance. However, discerning small motion from noise is a complex task, especially when attempting to magnify very subtle often sub-pixel movement. As a result, motion magnification techniques generally suffer from noisy and blurry outputs. This work presents a new state-of-the-art model based on the Swin Transformer, which offers better tolerance to noisy inputs as well as higher-quality outputs that exhibit less noise, blurriness and artifacts than prior-art. Improvements in output image quality will enable more precise measurements for any application reliant on magnified video sequences, and may enable further development of video motion magnification techniques in new technical fields.

show abstract

“…This work is based on the method developed by Lado-Roigé et al (2022) for vibration-based damage detection and on the Swin Transformer Based Video Motion Magnification (STB-VMM) method (Lado-Roigé & Pérez, 2023), which improves on the previous motion magnification backend (Oh et al, 2018) in terms of image quality.…”

Section: Related Workmentioning

confidence: 99%

ViMag: A Visual Vibration Analysis Toolbox

Lado-Roigé,

Pérez

2023

JOSS

Self Cite

View full text Add to dashboard Cite

Recent developments in computer vision have brought about a new set of techniques called Video Motion Magnification, which are capable of identifying and magnifying eye-imperceptible movements in video data. These techniques have proved effective in applications such as producing visual representations of an object's operating deflection shapes or recovering sound from a room behind soundproof glass. Our research explores the new possibilities of motion magnification applied to Structural Health Monitoring (SHM) and vibration testing, harnessing the latest advances in deep-learning to achieve state-of-the-art results.

show abstract

Learning-based video motion magnification approach for vibration-based damage detection

Cited by 15 publications

References 37 publications

Applying video motion magnification to reveal spontaneous tympanic membrane displacement as an indirect measure of intracranial pressure in patients with brain pathologies

Applying video motion magnification to reveal spontaneous tympanic membrane displacement as an indirect measure of intracranial pressure in patients with brain pathologies

STB-VMM: Swin Transformer Based Video Motion Magnification

ViMag: A Visual Vibration Analysis Toolbox

Contact Info

Product

Resources

About