论文部分内容阅读
对于地形遥测数据中引入的脉冲噪声,传统的滤波算法往往需要选择滤波门限,自适应能力不强,或者不能在滤除噪声的同时有效地保护信号数据,或者运算复杂。针对以上问题,文中提出了一种快速的脉冲噪声自适应滤除算法,该算法基于数理统计思想,对局部数据(滤波窗口)进行均值和方差估计,并根据估计结果自动选择检噪门限,进而实现噪声检测和平滑。实验结果显示,在脉冲噪声密度小于5%时,该算法的滤波信噪比增益远高于常用滤波算法和其他同类算法;对平稳变化信号的处理效果较好。
For impulsive noise introduced in topographic telemetry data, the traditional filtering algorithms often need to select the filtering threshold, the adaptive ability is not strong, or can not effectively protect the signal data while filtering the noise, or the operation is complicated. In view of the above problems, a fast impulsive noise adaptive filtering algorithm is proposed in this paper. Based on the idea of mathematical statistics, this algorithm estimates the mean and variance of local data (filter window), and automatically selects the threshold for detection according to the estimation result Achieve noise detection and smoothing. Experimental results show that when the impulse noise density is less than 5%, the filtering signal-to-noise ratio gain of this algorithm is far higher than that of the common filtering algorithm and other similar algorithms; and the effect of smooth changing signal processing is better.