针对散斑噪声很难用常用的滤波方法进行滤除,在已有等值线滤波方法的基础上,提出一种更优的自适应窗口滤波方法。由于等值线窗口是在条纹方向图的基础上得到的,深入研究条纹方向的求取方法,提出了一种更为可靠的条纹方向求取算法,在得到高精度条纹方向的同时,还能估计出条纹的密度。使用等值线窗口技术,保持了窗口形状的自适应能力,然后利用对条纹密度的估计,根据条纹的宽度来确定滤波窗口的大小,实现了滤波窗口大小的自适应。最后,根据条纹方向和滤波后的图像可以直接得到条纹密度变化较大的单幅散斑图的相位结果。

Electric speckle pattern interferometry (ESPI) is a modern measuring technology and has many applications. But the high speckle noise makes the signal-to-noise ratio (SNR) of the electric speckle pattern interferometry fringe pattern very low and so limits the applications of electric speckle pattern interferometry. It is hard for usual filters to filter off the high speckle noise. A novel filtering method is proposed based on contoured window filter. Because the contoured window is obtained from fringe orientations, so we proposed a new algorithm to get more precise orientations. This algorithm can calculate the precise orientations and estimate the fringe widths at the same time. Based on this algorithm, adaptive window filter is designed. This new filter can adapt its shape following the fringe orientations and adapt its size according to the density of the local fringe. Last the phase can be obtained from the filtered image whose density varies a lot.