arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1203.4035cs.MM
收藏

一维语音信号和二维图像的小波多尺度定量分析

Quantitative Multiscale Analysis using Different Wavelets in 1D Voice Signal and 2D Image

Niraj Shakhakarmi

摘要:

多尺度分析代表对信号的多分辨率审查,以改善其信号质量。利用DCT、FFT和不同的小波(如Haar、Deubachies、Morlet、Cauchy、Shannon、Biforthogonal、Symmlet和Coiflet)对一维语音信号和二维图像进行多分辨率分析,并部署基于级联滤波器组的分解和重构。对所选小波进行了定量分析,研究了一维语音信号和二维图像在多尺度第四级的信号质量、均方误差、熵和峰峰值信噪比。此外,在多尺度分析中,2D图像压缩性能在DB-4中为93.00%,在bior-4.4中为93.68%,在Sym-4中为93.18%,在Coif-2中为92.20%。

英文摘要:

Mutiscale analysis represents multiresolution scrutiny of a signal to improve its signal quality. Multiresolution analysis of 1D voice signal and 2D image is conducted using DCT, FFT and different wavelets such as Haar, Deubachies, Morlet, Cauchy, Shannon, Biorthogonal, Symmlet and Coiflet deploying the cascaded filter banks based decomposition and reconstruction. The outstanding quantitative analysis of the specified wavelets is done to investigate the signal quality, mean square error, entropy and peak-to-peak SNR at multiscale stage-4 for both 1D voice signal and 2D image. In addition, the 2D image compression performance is significantly found 93.00% in DB-4, 93.68% in bior-4.4, 93.18% in Sym-4 and 92.20% in Coif-2 during the multiscale analysis.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑