Noise reduction in electronic speckle pattern interferometry fringes by fourth-order Partial Differential Equations

摘要

Noise reduction is one of the largest problems and biggest difficulties involved in electronic speckle pattern interferometry (ESPI). Although the second-order PDEs denoising method is a useful tool of noise reduction for the ESPI fringe patterns, its main drawback is that the second-order PDE model does not remove impulse noise, a 3×3 mean window filter is generally needed to improve the fringes. For overcome this main drawback, in this paper we apply the fourth-order PDE denoising model to the computer-simulated and experimentally obtained ESPI fringe, respectively. In both tests, the fourth-order PDE denoising model clearly outperforms the second-order PDE denoising model. Experimental results have confirmed that the fourth-order PDE denoising model is capable of removing noise in ESPI fringe images effectively.