Solution-Driven Adaptive Total Variation Regularization

Solution-Driven Adaptive Total Variation Regularization
复制标题

DOI:
10.1007/978-3-319-18461-6_17
复制
发表时间:
2015-05
期刊:
--
影响因子:
--
通讯作者:
Frank Lenzen;J. Berger
Frank Lenzen;J. Berger
中科院分区:
其他
文献类型:
--
作者:
Frank Lenzen;J. Berger

文献摘要

被引文献

相似文献

我们认为解决方案驱动的自适应变量的总变分,其中的自适应性被引入作为一个不动点问题。我们提供的存在性理论,这样的不动点在连续域。对于图像去噪,去模糊和修复的应用,我们提供的实验表明,我们的方法在大多数情况下优于最先进的正则化方法。
We consider solution-driven adaptive variants of Total Variation, in which the adaptivity is introduced as a fixed point problem. We provide existence theory for such fixed points in the continuous domain. For the applications of image denoising, deblurring and inpainting, we provide experiments which demonstrate that our approach in most cases outperforms state-of-the-art regularization approaches.