摘要
Curvelet去噪彻底,易产生"阶梯"效应,四阶偏微分方程LLT模型去噪能很好地保护图像细节特征,不会产生阶梯效应。提出了一个模型将Curvelet去噪和四阶偏微分方程中LLT模型去噪的优点通过权函数λ结合起来,实验表明,新的模型能很好地发挥二者的优点,不但很好地保护图像细节特征,消除阶梯效应,去噪彻底且能很好地保持边缘,去噪率和视觉效果优于Curvelet和LLT方法。
Curvelet is able to remove noise thoroughly,but the staircase effect is easily brought about.Smooth regions are protected well and the staircase is not produced by four-order partial differential equation LLT model.A new model based on Curvelet and fourth-order PDE is proposed by combining Curvelet and LLT model with weight function 2.The experiment results have demonstrated that the new model can preserve the advantages of Curvelet and LLT, at the same time, the de-noising rate and visual effect are superior to both of them.
出处
《计算机工程与应用》
CSCD
北大核心
2010年第25期168-170,177,共4页
Computer Engineering and Applications
基金
湖南省自然科学基金No.08JJ3131~~
