Rapid measurement of video-amplified microvibration signals based on stagewise weak orthogonal sparse representation

Video vibration measurement technology has attracted much attention for its ability to make high-resolution measurement of tiny vibration in a non-contact state. Traditional measurement methods rely on pyramid amplification techniques to amplify microvibration signals for easy processing, however, this method is computationally large and cumbersome, resulting in a longer processing time. To solve this problem, this paper proposes a rapid measurement of video-amplified micro-vibration signals based on stagewise weak orthogonal sparse representation, based on the pyramid amplification algorithm and the characteristics of sparse representation and division motion estimation. The algorithm first obtains moving pixels by pre-processing, then decomposes the moving pixels into subbands through sparse multi-scale decomposition and transforms them into vectors, and then obtains sparse subbands by combining with sparse representation, and then passes through time domain filtering and phase amplification to obtain the reconstructed video. Finally, the phase change is extracted by using the division motion estimation. Finally, the phase change was extracted by dividing motion estimation to obtain the spectrogram. Meanwhile, the algorithm is verified through simulation experiment and specific experiment, the results show that the algorithm can achieve high-precision, short-time video amplification and rapid measurement of micro-vibration signal.