CAJ | 학술논문

分析了非下采样Contourlet变换,利用基于冗余结构的非线性提升方法实现了平移不变的不可分离小波变换,提出了一种改进的非下采样Contourlet变换理论(NSCT-I).NSCT-I具有更好的细节捕捉能力、平移不变性、良好的多分辨率、局部化和多方向性,能够获得更好的图像处理效果.在分析现有阈值函数和阈值选择方法的基础上,构建了适合NSCT-I变换方法的阈值函数,提出了一种基于NSCT-I的收缩阈值去噪算法.分析和仿真实验表明:该算法具有良好的去噪效果和保持细节的性能.
분석료비하채양Contourlet변환,이용기우용여결구적비선성제승방법실현료평이불변적불가분리소파변환,제출료일충개진적비하채양Contourlet변환이론(NSCT-I).NSCT-I구유경호적세절포착능력、평이불변성、량호적다분변솔、국부화화다방향성,능구획득경호적도상처리효과.재분석현유역치함수화역치선택방법적기출상,구건료괄합NSCT-I변환방법적역치함수,제출료일충기우NSCT-I적수축역치거조산법.분석화방진실험표명:해산법구유량호적거조효과화보지세절적성능.
An improved Nonsubsampled Contourlet Transform Theory(NSCT-I)was proposed based on the Nonseparable Wavelet Transform(NSWT).The shift invariant NSWT was realized by non-linear lifting method based on the redundancy structure.NSCT-I can produce better image processing effect because of its better detail capturing capability,shift invariance,good multi-resolution,localization,and multi-direction.After analyzing the current threshold functions and threshold selecting methods,a new threshold function suitable for NSCT-I was established.Furthermore,a novel shrinking threshold denoising algorithm was presented based on the threshold function.Experimental results show that the proposed algorithm has better denoising effect and detail-preserving capability.