EUVIMEDEUVIMED entdecken
Crossref

Enhanced Motion Magnification for Accurate Sub-Pixel Displacement Estimation Using the Non-Decimated Dual-Tree Complex Wavelet Transform

Bo Wang, Chunyu Liu, Xinwei Shi, Guangyuan Weng, Xiyu Zhu, Xiao Tan

Journal of Testing and Evaluation · 01.01.2027

DOI: 10.1520/jte20250431 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Testing and Evaluation, online erschienen am 07.07.2026. Verfasst von Bo Wang und 5 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The motion magnification algorithm facilitates visualization and quantification of sub-pixel displacements associated with subtle motions that are otherwise imperceptible to the naked eye. This study presents a video decomposition and image pyramid construction method based on the non-decimated dual-tree complex wavelet transform (NDDTCWT), specifically designed to amplify minute motions and compute optical flow. The paper reviews limitations inherent in conventional motion magnification algorithms and advocates the NDDTCWT as a solution, owing to its edge-preserving properties and shift-invariance. A comparative analysis reveals that the NDDTCWT maintains reconstruction accuracy near machine precision, whereas the traditional complex steerable pyramid accumulates significant errors with increasing layers. Leveraging this advantage, a motion magnification algorithm based on the NDDTCWT is developed. The effectiveness of the proposed algorithm has been validated using both simulated and experimental videos, demonstrating its superior capabilities in motion magnification visualization and optical flow computation. The proposed method demonstrates robust performance on videos captured using low-cost imaging devices, maintaining stable motion magnification results under practical acquisition conditions. In terms of displacement measurement, the proposed method achieves displacement measurement accuracy comparable to state-of-the-art optical flow approaches such as RAFT under sub-pixel random vibration scenarios, and introduces fewer visual artifacts in motion magnification compared with conventional methods.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Wang, B., Liu, C., Shi, X., Weng, G., Zhu, X., & Tan, X. (2027). Enhanced Motion Magnification for Accurate Sub-Pixel Displacement Estimation Using the Non-Decimated Dual-Tree Complex Wavelet Transform. Journal of Testing and Evaluation. https://doi.org/10.1520/jte20250431

Vancouver

Wang B, Liu C, Shi X, Weng G, Zhu X, Tan X. Enhanced Motion Magnification for Accurate Sub-Pixel Displacement Estimation Using the Non-Decimated Dual-Tree Complex Wavelet Transform. Journal of Testing and Evaluation. 2027. doi:10.1520/jte20250431

BibTeX
@article{Wang2027,
  author = {Wang, Bo and Liu, Chunyu and Shi, Xinwei and Weng, Guangyuan and Zhu, Xiyu and Tan, Xiao},
  title = {Enhanced Motion Magnification for Accurate Sub-Pixel Displacement Estimation Using the Non-Decimated Dual-Tree Complex Wavelet Transform},
  journal = {Journal of Testing and Evaluation},
  year = {2027},
  doi = {10.1520/jte20250431}
}
Kostenlos starten →