The Applications Of Dyadic Wavelet In The Image Processing

This paper proposes the dyadic discrete time wavelet transformàTrous algo-rithm based B Spline Function. although TheàTrous algorithm is similar withthe Mallat algorithm, they also have different part. Unlike Mallat,TheàTrousalgorithm doesn't extract even sample while decomposing, and it wouldn't insertzero sample while reconstruct. The decomposing and reconstruct filter coffcientchanges will be done with the scale expansion and compression. The dyadic wavelettransform doesn't occur compression phenomenon with the help of these above fea-tures,when it process signal or image. That is why we often choose theàTrousalgorithm to detect the edge of the image. The dyadic wavelet transforms at twoadjacent scales are multiplied as a product function to magnify the edge structuresand suppress the noise. Base on the multiplied scales algorithm, this paper in-troduce a idea that the multiplied scale at two adjacent scales are multiplied as aproduct function to magnify the edge structures and suppress the noise. It is shownthat the scale multiplication at twice achieves better results than scales multiplica-tion in once, especially on the localization performance. According the signal andnoise in the two different scale wavelet transform have di?erent transformation fea-ture,this paper gives a brief introduction of SSNF algorithm. An improved SSNFalgorithm implement in the image denosing. Experiments on natural images arecompared with other denoising Algorithms, it is apparent that the improved SSNFalgorithm is better.