高阶非局部变分模型

Higher-order nonlocal variational model

为了更好地描述图像的非局部结构并抑制噪声,通过定义新的权函数,提出了一个高阶非局部变分模型.理论分析和实验结果表明,相比于非局部平均去噪算子和非局部变分模型,提出的高阶非局部变分模型在去噪结果上有更高的信噪比和更好的视觉效果.因此,新模型保持了非局部变分模型的优点,同时能够更好地保留图像特征.

To describe the nonlocal structure of images better and remove noise,this paper proposes a higher-order nonlocal variational model by defining a new weighted function.Theoretical analysis and numerical experiments are presented to demonstrate the performance of the proposed model,which is better than the original in terms of both PSNR and visual perception.As a result,the new model keeps the advantages of the nonlocal variational model,while it can preserve image features better.