The Research Of Digital Image Processing System Based On ESPI

Electronic Speckle Pattern Interferometry is a whole-field, nondestructivemeasurement technique, which is well-known for deformative and nondestructivemeasurement of coarse surface. It has advantages of high accuracy and fastresponse. In the research of ESPI, It is very important to process the measuredimage data. The image processing highly related to the final measurement result.Traditional speckle fringe patterns are inherently noisy and low visibility, so it isdifficult to extract the deformation and displacement information accurately. Forthe extraction of the effect precision information from the image, it demands thespecial image processing method for ESPI image.This paper is based on the research of formation principle of ESPI and thedigital image processing algorithm. A digital image processing system base onESPI was independent designed. The dissertation mainly includes:(1)On the basis of the sufficient research of mechanism of ESPI, we designedand set up the in-plane and out-plane displacement ESPI system; and we alsodesigned and manufactured the PZT phase shifter. In order to guarantee itsprecision, we did calibration experiment.(2)The digital image processing algorithm and its computer program wasstudied, such as image preprocessing, image smoothing, and phase map calculating,phase map unpacking algorithm.(3)By using the Microsoft Visual C++and Matrox Imaging Library which isprovided by Matrox, we designed the digital image process software. This softwarehas function of displaying real time images of speckle pattern interferometry fringe.It can also control the PZT phase shifter automatically which based on temporalphase shift technology. This software is capable of the task from original imagecollection, phase map calculating, image smoothing and phase map unpacking tothe final result’s3D display. It has the on-line and off-line process abilities.(4)By using the ESPI system and its software, the out-plane deformationmeasurement experiment and the defect detecting experiment were performed. Theresult demonstrates the system and software’s reliability.