二进小波在图像处理中的应用-计算数学专业论文.docxVIP

ABSTRACT THE APPLICATIONS OF DYADIC WAVELET IN THE IMAGE PROCESSING ABSTRACT This paper proposes the dyadic discrete time wavelet transform a` T rous algo- rithm based B Spline Function. although The a` T rous algorithm is similar with the Mallat algorithm, they also have different part. Unlike Mallat,The a` T rous algorithm doesn’t extract even sample while decomposing, and it wouldn’t insert zero sample while reconstruct. The decomposing and reconstruct filter coefficient changes will be done with the scale expansion and compression. The dyadic wavelet transform 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 the a` T rous algorithm to detect the edge of the image. The dyadic wavelet transforms at two adjacent scales are multiplied as a product function to magnify the edge structures and 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 a product function to magnify the edge structures and suppress the noise. It is shown that the scale multiplication at twice achieves better results than scales multiplica- tion in once, especially on the localization performance. According the signal and noise in the two different scale wavelet transform have different transformation fea- ture,this paper gives a brief introduction of SSNF algorithm. An improved SSNF algorithm implement in the image denosing. Experiments on natural images are compared with other denoising Algorithms, it is apparent that the improved SSNF algorithm is better. KEY WORDS: The Dyadic Discrete Time Wavelet The Scales Multipli- cation Edge Detection  a` T rous Algorithm SSNF Algorithm 目 录 中文摘要 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I 英文摘要 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II 第一章 绪 论 1 §1.1 图像处理相关研究方向 . . . . . . . . . . . . . . . .