Adaptive total variation model for image inpainting and denoising

作者：Lu Zhaolin^*; Qian Jiansheng; Li Leida

来源：ICIC Express Letters, Part B: Applications , 2011, 2(5): 1219-1224.

• De-noising
• De-noising algorithm
• Diffusion functions
• Image Inpainting
• Inpainting
• Numerical scheme
• Speed of convergence
• Staircase effect
• Total variation
• TV model
• Visual qualities

Image inpainting and denoising algorithms based on gradient dependent regularizers, such as total variation (TV) model, often suffer the staircase effect. In order to overcome such a drawback, a novel adaptive total variation (ATV) model is proposed. By modifying the diffusion function and related parameters, the ATV model not only preserves the edges well but also eliminates the staircase effect. Experimental results show that visual quality of the restored image and the speed of convergence of the numerical scheme are both much improved in comparison with other algorithms.

• 出版日期2011
• 单位中国矿业大学（北京）