机构地区: 哈尔滨理工大学计算机科学与技术学院,黑龙江哈尔滨150080
出 处: 《计算机工程与科学》 2017年第8期1520-1524,共5页
摘 要: 针对全变分(TV)模型在去除图像噪声时容易产生阶梯效应的缺点,将二阶总广义变分(TGV)作为正则项应用于全变分模型中可以有效地去除阶梯效应,并且还能够更好地保持图像边缘纹理结构;利用非局部均值滤波算法的思想来构造非局部微分算子,将非局部微分算子应用于总广义变分模型中,综合提出了一种基于非局部总广义变分的图像去噪新模型。新模型充分利用了图像的全局信息进行去噪。实验结果显示了该模型的有效性和优越性。 The total variation model can remove noise effectively, however, it also brings in staircase effect. To overcome this shortco regularization term in the new d effect, which model ming, we use the second order total generalized variation (TGV) as the enoising model. The TGV model can not only eliminate the staircase also preserve structures are constructed based on , and the new method ma perimental results demonstrate such as edges and textures better. The nonlocal differential operators idea of the nonlocal means filtering algorithm are applied to the TGV good use of the global information of the image to remove noise. Exvalidity and superiority of the proposed method.