去除图像高强度乘性噪声的变分模型

Variational Model to Removal Strong Multiplicative Noise

本文提出了一种新的去除图像高强度乘性噪声的变分模型,该模型针对现有全变分方法在去除图像高强度乘性噪声时出现的边缘模糊、去噪效果不佳及"阶梯"效应等问题进行研究.然后导出了该模型对应的偏微分方程的初边值问题,分析了模型的去噪机理,并给出了相应的数值计算方法.数值实验结果表明,新模型不仅提高了图像去噪的质量,在视觉上更平滑自然,基本上消除了"阶梯"效应.此外,新模型在运行时间方面也具有较大的优势.

In this paper, a new variational model is proposed and studied to remove the strong multiplicative noise. The initial boundary value problem of the partial differential equation for the model is derived and discreted numerically. The model has succeeded in inhibiting multiplicative noise effectively, preserving the textures and fine details of images perfectly. It is concluded from the experiments that the new model not only can remove multiplicative noise effectively, but also is more superior to preventing the ＂step-casing effect＂ than the existing key models. Moreover, the new variational model is much superior than the existing key models in the run- ning time.