Research On Micro-vibration Measurement Technique Based On Eulerian Video Magnification

Micro-vibration behavior is ubiquitous in engineering technology and daily life,which often contains a wealth of useful information,and it is necessary to measure it.However,the amplitude of the micro-vibration is very weak,and it is difficult to be perceived by the naked eye and determine its position,thus causing difficulties in measurement.Traditional vibration measurement methods are not well-targeted for micro-vibration measurement,and may be subject to electromagnetic interference during measurement,or have a mass-load effects,or are expensive and inconvenient to operate.Therefore,it is still a huge challenge to study the micro-vibration measurement technique with strong adaptability,low cost and high precision.The purpose of the subject "Research on Micro-vibration Measurement Technique Based on Eulerian Video Magnification" is to study a micro-vibration measurement technique based on eulerian video magnification to avoid the above shortcomings existing in traditional vibration measurement methods while visualizing micro-vibration measurement.The specific research contents of this article are as follows:Firstly,based on the analysis of linear eulerian video magnification and phase based eulerian video magnification,the similarities and differences,advantages and disadvantages as well as the applicable scenes are compared,thus laying a theoretical foundation for the visualization part of the subsequent micro-vibration measurement.Secondly,a visual measurement algorithm of micro-vibration based on eulerian video magnification is proposed.In this algorithm,eulerian video magnification is used to strengthen the micro-vibration behavior,so as to extract the vibration area to be measured,and then the Gaussian pyramid is used for Multi-layer sampling filtering to achieve a large number of data reduction and spatial noise suppression.Then,the difference transform is used to suppress noise in the spectral domain,and the proposed automatic K-means clustering algorithm is used to improve the accuracy of micro-vibration measurement.Finally,the estimated spectrum of vibration region is obtained.Thirdly,the equipment construction of the micro-vibration visual measurement algorithm is realized,so that the micro-vibration measurement algorithm proposed in this paper can be applied to practice.In addition,the principle of camera calibration is studied,so as to apply it to avoid the degradation of the magnified video quality caused by lens distortion.Finally,three types of experiments are designed to verify the micro-vibration visualization measurement technique proposed in this paper,including cantilever beam experiment,pulse experiment,and loudspeaker experiment.The experimental results show that the eulerian video magnification algorithm can strengthen the micro-vibration behavior and help us determine the micro-vibration area.And the micro-vibration measurement results are in good agreement with the measurement results of the acceleration sensor,pulse oximeter and set truth value respectively.