Study Of Electronic Speckle Pattern Interferometry And Its Application On Dynamic Thermal Deformation

The Electronic Speckle Pattern Interferometry (ESPI) is a non-destructive,whole-field optical measurement technique, which is applied widely in optical roughsurface displacement measurement, strain analysis, non-destructive technology (NDT)and vibration measurement. Electronic Speckle Pattern Interferometry with fringe orphase patterns of the form of output usually contains a large number of noise. In orderto obtain the precise phase distribution, denoising became the key step of ElectronicSpeckle Pattern Interferometry. In this paper, we research on the denoising of phaseand fringe patterns, and apply the results on dynamic ESPI thermal deformationmeasurement.We first analyze and discuss denoising methods in Fourier transform domain,and put forward the Localized Fourier Transform filtering method according to theadvantages and disadvantages of the former algorithms. Our method not onlyovercomes the faults of global Fourier transform’s difficult to choose cut-offfrequency, but also be simpler for its few parameters compare to Windowed Fouriertransform.Secondly, we discussed the empirical mode decomposition algorithm. In view ofthe characteristics of ESPI fringe patterns, we put forward the improvedbidimensional empirical mode decomposition algorithm. And our improvements areon the three aspects:(1) The selection of extreme points;(2) Applied the radial basisfunction interpolation method to the upper and lower envelope surface fitting;(3) Asthe median filter has the effects of edge preserving, we applied it in the improvedalgorithm. This method is a kind of multi-scale global data driven approach, and donot need all kinds of transform and cut-off frequency, which made it less parametersand easy to use.At last, We respectively compared the proposed two algorithms with kinds offormer methods by computer simulation or experiments ESPI patterns, and appliedthe improved bidimensional empirical mode decomposition algorithm on dynamicESPI thermal deformation measurement. The experimental results show that theperformance of the method. The improved Empirical Mode Decomposition algorithmis better than the original algorithm, especially for the wide fringe patterns. In the filtering of low quality fringe pattern which obtained by experiment, the improvedalgorithm is achieved good effect.