改进的小波收缩消噪技术及确定消噪门限的方法  

A Improved Wavelet Shrinkage Method and Determining of the Denoise Threshold

作　　者：邢占峰[1] 张凯[1] 吕扬生[1] 

机构地区：[1]天津大学精密仪器与光电子工程学院生物医学工程系,博士研究生300072

出　　处：《电子测量与仪器学报》2003年第3期37-41,共5页Journal of Electronic Measurement and Instrumentation

摘　　要：本文阐述了一种改进的小波收缩消噪技术 ,将在小波系数域进行的收缩改进到模值域上进行 ,并根据统计知识和李氏指数的理论推导出了噪声各个阶次小波变换模值上界之间的关系及对模值上界的估计方法。进而提出了各个阶次消噪阈值的自动确定技术 ,实现了无人工干预的小波消噪。This article introduced an improved wavelet shrinkage technique, in which the shrinkage was not processed in the domain of wavelet transform but processed in the domain of modulus of wavelet transform. In addition, we derived the relationship among various scales of wavelet transform superior bounds related to noise and provided a method used to estimate the superior bounds. Finally, method of automatically determining the denoise thresholds was given, and an unsupervised denoise algorithm was presented.

关 键 词：散斑噪声 小波收缩 噪声抑制 李氏指数 B型超声诊断仪 

分 类 号：R445.1[医药卫生—影像医学与核医学]