The effect of the existing Total Variation (TV) method for image denoising is not ideal while removing noise,it is not well protection of the edge information of image and easy to appear the"staircase effect" and"blocky effect" when recovering the image. Ai-ming at this problem,a new method of image denoising based on TV of fourth-order differential is proposed in this paper. First,the ad-vantages and disadvantages of the traditional image denoising methods of TV and higher order differential equations are discussed respec-tively. Then,combined the denoising model of TV with fourth-order differential theory,a new adaptive ATV model with the edge indica-tor is obtained,and a rational differential mask in eight directions is drawn. The experimental results show that this method can reduce the noise well,and preserve edge features better,and reduce the influence of the "staircase effect",avoiding the "blocky effect" of the im-age.%针对现有全变差( TV)方法效果不太理想，在去除噪声的时候不能较好地保护图像的边缘信息，而且恢复图像易出现“阶梯效应”和“块效应”的问题，文中提出了一种基于四阶微分全变差的图像去噪模型。首先论述了传统全变差模型去噪方法及其高阶微分方程去噪方法各自的优缺点；然后将带有边缘指标的全变差模型与四阶微分理论相结合，得到了一个新的带有边缘指标的自适应全变差去噪模型，并引入差分方程去定义模型中的变量。实验结果表明，该模型能较好地抑制噪声、保留边缘特征和衰减图像的“阶梯效应”，并能较好地避免图像的“块效应”。
展开▼
