节点文献
基于模糊异质扩散的图像去噪方法
Image Denoising Using Fuzzy Theory Based Anisotropic Diffusion
【摘要】 将模糊思想与Perona-Malik方法相结合,提出一种新的基于模糊理论的图像异质扩散滤波方法。把异质扩散方程中的边界停止函数看作像素梯度对于图像连续平滑区域的模糊隶属度函数,利用模糊熵的概念来确定图像中分片连续区域间的边界阈值σ。本文的模糊方法和其他经典方法的异质扩散比较实验说明,该方法在保持图像的细节方面更具有优越性。
【Abstract】 A new image anisotropic diffusion technique by combining the fuzzy theory with the Perona-Malik method is proposed.The "edge-stopping" function used in anisotropic diffusion is modeled with the fuzzy membership between the image gradient and the corresponding smooth regions.The related edge threshold σ is determined by the fuzzy entropy.Experiments by the proposed method are compared with that of other classical methods.And the feasibility and the efficiency are well validated.
【关键词】 异质扩散;
偏微分方程;
图像去噪;
模糊隶属度;
模糊熵;
【Key words】 anisotropic diffusion; PDE; image denoise; fuzzy membership; fuzzy entropy;
【Key words】 anisotropic diffusion; PDE; image denoise; fuzzy membership; fuzzy entropy;
【基金】 国家“九七三”基础研究(2003CB716102)资助项目
- 【文献出处】 数据采集与处理 ,Journal of Data Acquisition & Processing , 编辑部邮箱 ,2007年01期
- 【分类号】TN911.73
- 【被引频次】5
- 【下载频次】213