| 基于P（x）-Laplace流的向量值图像恢复模型 |
| |
| 引用本文： | 董刚.基于P（x）-Laplace流的向量值图像恢复模型[J].人天科学研究,2010(2):168-170. |
| |
| 作者姓名： | 董刚 |
| |
| 作者单位： | 东北农业大学成栋学院,黑龙江哈尔滨150030 |
| |
| 摘 要： | 提出一种基于变指数型扩散方程的向量值图像（例如RGB图像或多谱图像）恢复模型。在新的模型中,我们利用光滑后向量值图像的几何流形特征来判断图像的真假边缘,利用P（x）-Laplace型扩散系数来控制扩散量和扩散方向。由于扩散系数具有良好的自适应判断边缘能力,新的模型在图像恢复的同时良好地保持了图像边缘。通过数值实验,对比以往的全变差模型,所提模型具有更好的图像恢复效果,明显减少了＂块状＂效应。
|
| 关 键 词： | 变指数扩散 偏微分方程 向量值图像 图像恢复 |
Vector Valued Image Restoration Methods Based on P（x）-Laplace Flow |
| |
| Abstract: | In this paper, new vector value image restoration algorithms are proposed ,which is based on variable exponential diffusion sys- tem.In the new method, on one hand, geometric features for vector value image is considered to detective the true edge, on the other hand, we utilize the P （x）-Laplace mixed diffusion coefficients to control the velocity and direction of the diffusion.The diffusion coefficients in the new model are self-adaptive for the detective of edges ,so the new method performs very well for image denoising and edge protecting. |
| |
| Keywords: | Variable Exponential Diffusion Partial Differential Equation Vector Value Image Image Restoration |
| 本文献已被 维普 等数据库收录！ |
|
