基于小波分析的自适应噪声识别

图像噪声类型直接影响去噪方法的去噪效果.因此,研究图像噪声类型的识别,对于数字图像去噪方法效果的提高具有重要意义.利用小波分解的高频系数,分析直方图和曲线拟合图的跳变出现概率特征以及黄金分割点处的窗口宽度特征,提出一种数字图象噪声类型自适应识别方法.针对图像噪声识别类型,采用相适用去噪方法提高图像去噪效果.通过大量实验表明,该方法是切实有效的.

Adaptive identifying of noise types based on wavelet analysis

Noise types of images directely impacted on the result for denoising.Therefor,the research to identify noise types of image was of great significance to the improvement of digital image denoising.A adaptive identification of noise types of digital image algorithm was mentioned by researching of the obvious features--probability of the emergence of jump points and window width of Golden Section points--extracted from the histogram and curve fitting and using high-frequency coefficients of wavelet decomposition.Aiming at the identification of noise types of image, applicable denoising was used to improve the effect of image denoising. The numerous experiments showed that this methodology used to identify the noise types was effective.