Study of combined filter based on wavelet transform to denoise stripe images of electronic speckle shearography pattern interferometry

Stripe images of electronic speckle shearography pattern interferometry, in which stripe distribution are correlated with vertical micro distortion or micro vibration of objects, are severely disturbed by noises, and so denoising stripe images of electronic speckle shearography pattern interferometry is necessary to extract useful stripe distribution information. Denoising methods and flow for stripe images of electronic speckle shearography pattern interferometry are analyzed in this paper to get the stripe distribution correlated with vertical micro distortion or micro vibration of objects. The noises in the stripe images of electronic speckle shearography pattern interferometry are comprised of speckle noise and other random noises induced by environmental disturb and instrumental performance, so it's difficult to use familiar filters, such as mean-value filter, medium-value filter and adaptive filter, etc, to remove all noises in the stripe images. The combined filter composed of mean-value filter and wavelet filter is designed to denoise stripe images. The aim of mean-value filter is to remove random noises induced by environmental disturb and instrumental performance, and then the wavelet filter, in which the Meyer wavelet is adopted, is designed to remove speckle noise in the stripe images. The final stripe distribution images after denoising and binarization are listed to prove the denoising validity of combined filter based on wavelet transform.