光谱测量数据降噪处理滤波器设计  

Noise reducing filter design for the spectra data processing

作　　者：程海林[1] 孙杰[1] 马冉冉[1] 苑泊舟[1] 贺芳[1] 马斌[1] 

机构地区：[1]天津理工大学计算机与通信工程学院,天津300384

出　　处：《天津理工大学学报》2012年第1期45-50,共6页Journal of Tianjin University of Technology

基　　金：天津市产品质量技术监督检测局科技进步项目(09-06)

摘　　要：为准确分析光谱数据,有效降低噪声干扰是十分重要的.本文使用小波分析和中值滤波技术,提出了组合滤波器设计方案.使用模拟光谱数据进行了处理实验并将该方法用于实测光谱数据处理.实验表明,此方法不仅能有效地降低光谱中的脉冲噪声和高斯白噪声,而且较好地保留了光谱中的细节信息.Reducing the noise interference effectively is very important in the accurate analysis of the spectral data.On the base of the wavelet analysis and the median filtering technology a combined filter design scheme is described in this paper.The simulation spectrum data and the measured spectrum data captured with spectrum instrument are used to check the validity of this method.Experiment results show that not only the pulse noise and white Gaussian noise can be effectively reduced,but also it remains the details information in the spectrum after the processing finished with the method of this paper.

关 键 词：光谱 降噪 组合滤波器 小波分析 中值滤波 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]