Quality Evaluation on the Image Noise Reduction by Daubechies Wavelet Function

• Main Authors: Sheng-yi Chou, 周聖益
• Other Authors: Kao-Fei Chiang
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/09862888993011935538

碩士 === 立德大學 === 資訊傳播研究所 === 98 === Wavelet Transform approach has the abilities of multi-resolution analysis for both the time and the frequency domains, in the noise reduction has better effect. However, in the wavelet transform the wavelet function have a wide range of resolution capabilities of different image. How to choose the appropriate wavelet letter number for image analysis, is a very important issue. Using different wavelet functions will produce different image quality results. The purpose of this study was using Donoho proposed wavelet coefficient thresholding method to reduce the image noise from different kind of Daubechies wavelet functions. We are looking for the best method of the above Daubechies wavelet functions which the image quality is the performance index. Image quality assessment results show the above wavelet coefficients thresholding mothod can obtain a best resolution at image recovery. After all, we can get a filtered image reduce noise to obtain improved performance. This thesis discovered the threshold method reduce the image noise has the best image quality when the noise is low, otherwise image quality is worse. This study discovered the noise reduction effect of the image quality level and number of the wavelet will increasing the decomposition level.