基于分数阶偏微分的数字图像滤波去噪仿真

Simulation of Digital Image Filtering and Denoising Based on Fractional Partial Differential

受多种因素影响,数字图像会呈现出不同程度的噪声图斑,导致图像质量、清晰度降低,加大了图像分析、检测、分割等难度,为此,提出基于偏微分方程的图像自适应滤波去噪算法。区分图像噪声点和非噪声点,检测图像的边缘特征,提高算法对边缘的保留能力;构建分数阶偏微分方程模型,通过模型完成对图像的自适应滤波去噪处理。选取含有不同程度的噪声图像对所提方法展开实验测试,结果表明,所提方法可以有效去除图像中的冗余噪声,使图像整体质量得到大幅度提升。

Due to various factors,digital images may exhibit varying degrees of noise spots,leading to a decrease in image quality and clarity,as well as an increase in the difficulty of image analysis,detection,and segmentation.Therefore,this paper put forward an adaptive filtering algorithm for image denoising based on partial dfferential equations.Firstly,we distinguished between noisy and non-noise points in images,detected edge features of image,thus improving the ability of algorithm to preserve edges.Secondly,we constructed a model of fractional partial differential equation.According to this model,we achieved the adaptive filtering for image denoising.Finally,we chose the image containing different noise for testing the proposed method.Simulation results show that the proposed method can effectively remove redundant noise in images and significantly improves the overall quality of images.