摘要
提出了一种基于Split Bregman方法的快速曲率驱动(CDD)图像修补模型算法。由于CDD模型中曲率项的影响,数值求解高阶偏微分方程过程中需要大量迭代运算,修复速度缓慢。鉴于Split Bregman方法在L1正则化问题的成功应用,为提高算法计算速度,在CDD修补模型中引入Split Bregman方法。实验结果表明,与其他类似方法相比,新算法的实现速度显著加快,且视觉效果好。
A fast inpainting algorithm of curvature driven diffusions（CDD） model is proposed.It is known that the curvature term requires a large number of iterations in solving the high order partial differential equation in original CDD model.Due to the success of the Split Bregman method in L1 regularization problem,it is introduced into the CDD model to speed up the algorithm.The experiments show that the new algorithm is much faster and has better vision effects than other similar methods.
出处
《激光与光电子学进展》
CSCD
北大核心
2010年第8期64-68,共5页
Laser & Optoelectronics Progress
基金
国家自然科学基金(NSFC60872138)资助课题
