基于提升小波多分辨分析的语音信号和图像去噪.docVIP

石家庄铁道大学四方学院毕业设计 基于提升小波多分辨分析的语音信号和图像去噪 Speech Signal and Image Denoising Analysis Based on Lifting Wavelet Transfrom 摘　要 基于提升小波多分辨分析的语音信号和图像去噪是小波理论中的一项重要的内容，它是解决去噪问题的一种非常有效手段，提升系统的小波变换是基于提升方案的小波变换，比传统的小波变换有很多的优越性。本课题提出基于提升系统的小波多分辨分析的语音信号和图像去噪主要目的是体现提升小波对语音信号和图像去噪比起其他算法的优势。 本课题采用了对比的方法，对cameraman图像和leleccum信号加入自定义噪声，噪声方差为0.5，利用haar小波函数对该图像和信号进行二级分解，分别用小波变换和其对应的提升小波进行图像和信号去噪，并采用空域滤波和频域低通滤波对其去噪，给出了仿真结果，从视觉直观图和峰值信噪比、均方误差等参数上进行了分析对比，比较其效果，体现了提升小波对语音信号和图像去噪比起其他算法的优劣。 仿真结果表明，提升小波和传统小波相比，其优点在于计算简单，编程容易，速度快，该算法去噪后的图像质量优于一般小波变换和传统方法。 关键词：小波变换　提升小波　图像消噪　信号去噪 Abstract Based on lifting wavelet multiresolution analysis of the speech signal and image denoising in wavelet theory is an important content, it is to solve denoising problem is a very effective means, lifting system based on lifting scheme wavelet transform is wavelet transform, than the traditional wavelet transform has many advantages. In this paper based on the lifting system of the multi resolution analysis of wavelet transform in speech signal and image denoising the main purpose is to reflect the lifting wavelet transform for speech signal and image denoising compared with other algorithms advantage. In this paper in order to reflect the lifting wavelet denoising effect, this paper adopts the comparative method, adding a custom noise, the noise variance is 0.5. The denoising by using two layers of decomposition and Haar wavelet basis function, respectively, using the wavelet transform and its corresponding lifting wavelet for image and signal denoising, the simulation results are given, from the visual chart and peak signal to noise ratio, mean square error, parameters are analyzed and compared, compare its effect, embodies the lifting wavelet transform for speech signal and image denoising compared to other algorithms. In this paper, the algorithm is realized by using Matlab software platform. The simulation results show that compared with traditional wavelet transform, lifting wavelet, has the advantages of sim