arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2207.07321eess.IV
收藏

用于图像正则化的平均深度去噪器

Averaged Deep Denoisers for Image Regularization

Pravin Nair, Kunal N. Chaudhury

AI总结:

本文针对PnP和RED中CNN去噪器违反非扩张性导致发散的问题,提出通过展开小波去噪的分裂优化算法构造收缩性平均去噪器,首次给出训练收缩性去噪器的简单网络展开框架,并证明其正则化能力可与CNN去噪器匹配。

摘要:

即插即用(Plug-and-Play, PnP)和去噪正则化(Regularization-by-Denoising, RED)是近年来图像重建的范式,它们利用现代去噪器的能力进行图像正则化。特别是,它们已被证明能够借助CNN去噪器提供最先进的重建结果。由于正则化是以一种临时方式执行的,理解PnP和RED的收敛性一直是一个活跃的研究领域。最近的研究表明,如果去噪器是平均的或非扩张的,则可以保证迭代收敛。然而,将非扩张性与基于梯度的学习相结合具有挑战性,核心问题在于测试非扩张性是不可行的。通过数值示例,我们表明现有的CNN去噪器倾向于违反非扩张性质,这可能导致PnP或RED发散。事实上,训练非扩张去噪器的算法要么无法保证非扩张性,要么计算量很大。在这项工作中,我们通过展开应用于小波去噪的基于分裂的优化算法来构造收缩性和平均图像去噪器,并证明它们对PnP和RED的正则化能力可以与CNN去噪器相匹配。据我们所知,这是第一项提出使用网络展开训练收缩性去噪器的简单框架的工作。

英文摘要:

Plug-and-Play (PnP) and Regularization-by-Denoising (RED) are recent paradigms for image reconstruction that leverage the power of modern denoisers for image regularization. In particular, they have been shown to deliver state-of-the-art reconstructions with CNN denoisers. Since the regularization is performed in an ad-hoc manner, understanding the convergence of PnP and RED has been an active research area. It was shown in recent works that iterate convergence can be guaranteed if the denoiser is averaged or nonexpansive. However, integrating nonexpansivity with gradient-based learning is challenging, the core issue being that testing nonexpansivity is intractable. Using numerical examples, we show that existing CNN denoisers tend to violate the nonexpansive property, which can cause PnP or RED to diverge. In fact, algorithms for training nonexpansive denoisers either cannot guarantee nonexpansivity or are computationally intensive. In this work, we construct contractive and averaged image denoisers by unfolding splitting-based optimization algorithms applied to wavelet denoising and demonstrate that their regularization capacity for PnP and RED can be matched with CNN denoisers. To our knowledge, this is the first work to propose a simple framework for training contractive denoisers using network unfolding.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑