Denoising Techniques Of Morphology And Morphological Wavelet And Their Applications In Displacement Measurement

The techniques of morphological analysis and morphological wavelet analysis are the core digital image processing technique,with theoretical researches being deepening and applications being broadening,there are some issues should to be solved.Such as the difficult problem of structuring element decomposition in digital space,it makes morphological analysis only can use limited structuring element to describe the limited geometrical information of an image.The poor performance of the morphological filter makes the morphological filter only can reduce the low-density noise.Since the difficult problem of adaptive lifting for morphological wavelet,it is difficult to improve the theories and applications of the morphological wavelet analysis technique.Therefore,the research on the techniques of morphological analysis and morphological wavelet analysis can take an important place in the fields of both science and industry.Morphological analysis and morphological wavelet analysis possess the ability to describe the geometrical information of an image,and are widely used in the special geometrical characteristic of noisy image denoising.Digital speckle metrology is a full-field,non-contact and high accuracy measurement technology based on the digital images.However,the speckle interference fringe image always has the speckle noise,it will affect the accuracy and precision of the measurement results.Therefore,the interference fringe image denoising has been the important research issue in the speckle interference displacement measurement.The speckle noisy interference fringe image has the fringe characteristics of periodicity,directivity,and bright-and-dark.According to the morphological characteristics of a noisy image,the research on image denoising by using the techniques of morphological analysis and morphological wavelet analysis has very important meaning and wide application in the speckle interference displacement measurement.Based on reviewing the state of the research of the morphological analysis and morphological wavelet analysis,the scope of this dissertation work is focused on the morphological structuring element decomposition,the denoising methods of the impulse noisy image and speckle noisy interference fringe image,the lifting methods and applications of the morphological wavelet,etc.The main research contents and achievements of this thesis are as follows:1.In the digital space,morphological structure element decomposition becomes distortion,and is a difficult problem.The decomposition theorem and corollary for the convex structuring element is hence presented,and the necessary and sufficient conditions of the decomposition theorem have been proved.2.The methods for the impulse noisy image denoising based on the adaptive morphological filter,improved quantum inspired morphological filter,switching morphology-mean and switching median-mean filters are hence presented,respectively.Moreover,the step-by-step switching and double switching adaptive filtering models are presented.The proposed filtering models can be used to construct some effective filtering methods for the impulse noisy image.Simulations show that the proposed filtering methods are effective.3.For the application of the morphological analysis technique in the speckle interference displacement measurement,the denoising model for the Gaussian noisy interference fringe image is presented based on its binary image.The guiding morphological denoising method for the speckle noisy interference fringe image is presented.The experiments show that the proposed method is effective and achievable to extract the displacement information from the intensive interference fringe image.By comparison the theoretical results and the results of the speckle interference displacement measurement,the measurement results are accuracy.4.The variant morphological wavelet transform,the maximum and minimum lifting variant morphological wavelet transforms are presented by using the variant morphology.Moreover,the filtering method for the impulse noisy image based on the variant morphological wavelet transform,and the filtering method for the speckle noisy interference fringe image based on the morphological wavelet transform are presented.The proposed filtering method is achievable to extract the displacement information from the intensive interference fringe image.