Optimal Combination of Image Denoisers
Optimal Combination of Image Denoisers
复制标题
DOI：
10.1109/tip.2019.2903321
复制
发表时间：
2019-08-01
影响因子：
10.6
通讯作者：
Chan, Stanley H.
中科院分区：
文献类型：
--
作者：
Choi, Joon Hee;Elgendy, Omar A.;Chan, Stanley H.
给定一组图像去噪器，每个都具有不同的去噪能力，是否存在可证明的最佳方法来组合这些去噪器以产生整体更好的结果？这个问题的答案是为复杂场景设计弱估计器集合的基础。本文提出了一种利用深度神经网络和凸优化的最优组合方案。所提出的框架，称为共识神经网络（CsNet），在图像去噪中引入了三个新概念：1)一个可证明的优化过程，通过凸优化来组合去噪的输出；2)一种深度神经网络，用于估计去噪图像的均方误差（MSE），而不需要ground truth；3)利用深度神经网络提高图像对比度，恢复图像丢失的细节。实验结果表明，CsNet对确定性去噪器和神经网络去噪器的去噪性能都有显著提高。
Given a set of image denoisers, each having a different denoising capability, is there a provably optimal way of combining these denoisers to produce an overall better result? An answer to this question is fundamental to designing an ensemble of weak estimators for complex scenes. In this paper, we present an optimal combination scheme by leveraging the deep neural networks and the convex optimization. The proposed framework, called the Consensus Neural Network (CsNet), introduces three new concepts in image denoising: 1) a provably optimal procedure to combine the denoised outputs via convex optimization; 2) a deep neural network to estimate the mean squared error (MSE) of denoised images without needing the ground truths; and 3) an image boasting procedure using a deep neural network to improve the contrast and to recover the lost details of the combined images. Experimental results show that CsNet can consistently improve the denoising performance for both deterministic and neural network denoisers.