Presentation + Paper
12 March 2024 Remote motion sensing system utilizing enhanced video motion magnification
Author Affiliations +
Proceedings Volume 12899, MOEMS and Miniaturized Systems XXIII; 1289906 (2024) https://doi.org/10.1117/12.3000440
Event: SPIE OPTO, 2024, San Francisco, California, United States
Abstract
Optical imaging-based remote sensing of the mechanical motion of an object is presented utilizing phase-based motion magnification (PMM) technology that amplifies subtle movement of the object invisible to the naked eye. In case that the oscillatory motion of an object is very small compared to the pixel size of optical image, the resulting motion measurement is inaccurate due to the low signal-to-noise ratio (SNR). To overcome this limitation, various vibration measurement method using phase information of the optical image is known to be more robust to noise and lighting conditions as well, however the existing PMM technology has several inherent issues to be addressed such as phase ambiguity. In this context, this paper focuses on the improvement of motion measurement accuracy by overcoming the phase ambiguity, in turn, limitation of magnification range of the remote motion sensing in existing PMM technique. The proposed PMM utilizes Gabor-like complex steerable pyramid (CSP) to expand the magnification range while retaining the local characteristics of CSP. Specifically, phase unwrapping is employed to resolve the discontinuity of the phase, and the novel envelope shift method is newly applied on the top of phase unwrapping to extend the magnification range. We verified the proposed method through the experiment of the vibrating structure. The proposed method reduces the ringing artifact and blurring of the magnified motion of the target structure. The accuracy of the proposed method was compared to the physical accelerometer measurements and previous methods. The experimental verification showed that the proposed method has accuracy of more than 90% with reduced motion artifact, while the original PMM has an accuracy of about 67% to 87% depending on the magnification factor, which demonstrate accuracy and robustness of the proposed method.
Conference Presentation
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Jae-Duck Lee, Kang-Jae Park, and Yong-Hwa Park "Remote motion sensing system utilizing enhanced video motion magnification", Proc. SPIE 12899, MOEMS and Miniaturized Systems XXIII, 1289906 (12 March 2024); https://doi.org/10.1117/12.3000440
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Video

Motion measurement

Detection and tracking algorithms

Motion estimation

Vibrometry

Phase unwrapping

Image processing

Back to Top