Precision Micro-Vibration Measurement Using 1D Complex Morlet Wavelet Phase Magnification for Line-Scan Imaging

Conventional vision-based vibration assessment encounters considerable difficulties in detecting high-frequency, micro-amplitude structural dynamics due to the inadequate sampling rates of area-scan cameras and the computational inefficiency of 2D video magnification algorithms. This research provides a high-speed measuring framework that combines ultra-high-speed line-scan imaging with a one-dimensional Complex Morlet Wavelet Phase-based Video Magnification (CMW-PVM) method to address these difficulties. This approach operates only on the localized phase of 1 D spatial signals, avoiding the topological discrepancies and computing complexities associated with conventional 2D pyramids, which often lead to inaccurate results in high-speed imaging applications. The theoretical study delineates a definitive upper limit for the magnification factor to avert phase wrapping. Simulation outcomes demonstrate that the suggested approach preserves remarkable structural fidelity (FSIM > 0.85) and elevated PSNR in the presence of substantial noise (SNR =10 dB), considerably broadening the linear magnification range (more than 30). Experimental validations with a Laser Doppler Vibrometer (LDV) verify its metrological precision, attaining a <2% relative error across 10 ∼ 30 Hz baseline. Crucially, the framework successfully resolved microscopic displacements (≈ 20 μ m) at 100 Hz. Supplementary stress-tests under severe illumination degradation further proved the algorithm's exceptional environmental robustness, strictly preserving the frequency peak and maintaining a micrometer-level absolute accuracy (error ≈ 3 μm). This study presents a highly efficient and resilient framework for non-contact structural health monitoring in complex real-world conditions.