A Gaussian Mixture With Salt And Pepper Noise Denoising Based Research

Digital image processing has become an important method of aquiring complexinformation in an information age，and it has extended the vision of human beings insome extent．At present，digital image processing has been applied comprehensivelyin lots of fields and acquired enormous social and economic benefits. Therefore,therelative research and application of the digital image processing makes a significantsense for theory and practice．In the field of digital image processing, image denoising is one of the mostimportant and basic research topic at all times. The mixed noise composed byGaussian noise and Salt and pepper noise is a typical noise model in digital image.Traditional methods used to resolve this kind of noise are often less satisfactorythan expect, which appears that some image details may be lost while doing imagedenoising．Most natural images containing gauss noise and salt and pepper noise at thesame time, the use of simple traditional filtering algorithms can not get idealfiltering effect. In order to solve the problem of mixed noise of image filtering,respectively to give priority to with Gaussian noise and mixed noise image is givenpriority to with salt and pepper noise mixed noise of image, Proposeddual-threshold filtering algorithmThese two kinds of algorithm is revised in alpha average filtering algorithmimprovements made on the basis of two aspects: first, in the image in theneighborhood for different grey value of pixels are normalized weights, using theseweights and their corresponding grey value to filter the output. Secondly, theweights can be designed to fine-tune with correction factor, to obtain the ideal filtereffect. Experiments show that the treatment effect is superior to the traditionalfiltering algorithm and the adjusted alpha average filtering algorithm. In this thesis, experiments have been done in the PC of software MATLAB andexperiment results indicate that the fingerpri nt verification algorithm is moreeffective, feasible, and it achieves the desired results.