新通用阈值滤波方法在涡轮机匣检测中的应用  

Application of a new general de-noising method for wavelet threshold filtering to turbine casing detection

作　　者：杜乐普[1,2] 金印[1] 石宏[1] 张帅[1,3] 

机构地区：[1]沈阳航空航天大学航空航天工程学部(院) [2]空军石家庄飞行学院第一训练旅 [3]空军93318部队机务大队

出　　处：《沈阳航空航天大学学报》2015年第2期7-10,共4页Journal of Shenyang Aerospace University

基　　金：航空基金资助项目(项目编号:2008ZG54024)

摘　　要：小波域阈值滤波方法因其实现最简单,计算量最小而得到广泛应用。多年来出现了大量的阈值确定方法。其中Donoho提出的通用阈值法是应用最广的阈值获取方法之一,但由于其阈值选取单一,通用性较差。结合小波域滤波方法中的又一主要分支:空域相关滤波算法中的优秀思想,从最基本的Lipschitz指数α的特性出发,对Donoho提出的通用阈值法进行改进提升,使得新通用阈值法能对低信噪比、高污染信号产生非常优异的滤波效果。The wavelet threshold filtering method is widely used because of its simplest implementation and the minimum amount of calculation. Among the methods of threshold filtering,the general thresholding method proposed by Donoho is one of the most widely used method. But its single thresholding selection prevents its generality. Combined with the thoughts of the spatial correlation filters,the paper proposes a new method to improve Donoho′s general thresholding method,based on the most basic characteristics of the Lipschitz index of alpha. The new-general de-noising method is excellent in filtering the high reject signals of low signal-to-noise ratio.

关 键 词：小波 降噪 阈值函数 尺度间相关性 

分 类 号：TN911.4[电子电信—通信与信息系统]