文档介绍：摘要
信息时代,科技飞速发展,信息资源中的信号应用日益广泛,信号的结构越来越复杂,为了更加清楚地分析和研究实际工程中信号的有用信息,对信号进行消噪处理是至关重要的。信号消噪后,在语音识别方面,可以提取有效的语音信号;在图像处理方面,可以观察到清晰的图像等等,总之,在实际的工程应用中,信号消噪具有重要意义。
本文对基于小波变换的信号去噪方法进行了深入的研究分析,详细介绍了傅里叶变换和几种经典的小波变换去噪方法。结合相关的理论分析和Matlab实验结果,讨论了在一维与二维空间的去噪方法:分析了在一维空间阈值去噪过程中的小波的选取、阈值形式的选择以及阈值选择等因素对去噪效果的影响,介绍了模极大值去噪和小波阈值去噪等方法;探讨了在二维图像中,传统去噪与小波去噪的实现方法与比较。
通过本文的研究,可以得出结论:小波去噪比传统去噪效果更佳;采用不同的阈值选取形式所得去噪效果不同;选择软阈值的去噪效果比硬阈值的去噪效果要好。
关键词小波变换,信号去噪,Matlab
Abstract
At the information age, technology is rapidly developing ,The use of signal in the information resources es increasing, the signal structure is e plex, in order to analysis and research the useful information of signal more clearly in the actual engineering , the signal de-noising processing is critical .After the signal is be denoised, in the area of speech recognition, you can extract a voice signal more effective , in the area of image processing, you can be observed clear images, and so, in short, the signal de-noising is of great significance in practical engineering applications.
The subject deeply analysis the signal de-noising method based on wavelet transform, provide the knowledge of details in the Fourier transform and several classic wavelet transform denoising method. Combined with the theoretical analysis and Matlab experimental results it discussed denoising method in one-dimensional and two-dimensional space :it analysis the wavelet choice during the threshold denoising in one-dimensional space , the selection of threshold forms and threshold selection , it introduce the modulus maxima denoising method and wavelet threshold denoising method; it also discusses the method and parative in two-dimensional image between the traditional denoising and the wavelet denoising.
Through this study, it can be concluded : wavelet denoising is better than the traditional denoising ; selecting the different thresholds form will obtain different denoising e; selecting soft threshold denoising is better than the hard threshold