Abstract:Image demosaicking and denoising are the two key steps for color image production pipeline. The classical processing sequence consists of applying denoising first, and then demosaicking. However, this sequence leads to oversmoothing and unpleasant checkerboard effect. Moreover, it is very difficult to change this order, because once the image is demosaicked, the statistical properties of the noise will be changed dramatically. This is extremely challenging for traditional denoising models that strongly rely on statistical assumptions. In this paper, we attempt to tackle this prickly problem. Indeed, here we invert the traditional CFA processing pipeline by first demosaicking and then denoising. In the first stage, we design a demosaicking algorithm that combines traditional methods and a convolutional neural network (CNN) to reconstruct a full color image ignoring the noise. To improve the performance in image demosaicking, we modify an Inception architecture for fusing R, G and B three channels information. This stage retains all known information that is the key point to obtain pleasurable final results. After demosaicking, we get a noisy full-color image and use another CNN to learn the demosaicking residual noise (including artifacts) of it, that allows to obtain a restored full color image. Our proposed algorithm completely avoids the checkerboard effect and retains more image detail. Furthermore, it can process very high-level noise while the performances of other CNN based methods for noise higher than 20 are rather limited. Experimental results show clearly that our method outperforms state-of-the-art methods both quantitatively as well as in terms of visual quality.

Joint Demosaicing and Denoising with Gradient Guidance in Quad Bayer CFA

Learning to Joint Remosaic and Denoise in Quad Bayer CFA via Universal Multi-scale Channel Attention Network.

Multi-frame Blind Deconvolution Using Sparse Priors

A None-blind Deblurring Algorithm for Noisy Images Via Distributed Gradient Prior

Quad Bayer Joint Demosaicing and Denoising Based on Dual Encoder Network with Joint Residual Learning

Joint Denoising and Demosaicking With Green Channel Prior for Real-World Burst Images

Joint Demosaicing and Denoising With Self Guidance

QCGFNet: A Lightweight Color Image Denoising Network

Joint image demosaicking and denoising with mutual guidance of color channels

Inheriting Bayer's Legacy-Joint Remosaicing and Denoising for Quad Bayer Image Sensor

Toward Blind Joint Demosaicing and Denoising of Raw Color Filter Array Data.

Learning Rich Information for Quad Bayer Remosaicing and Denoising.

Inheriting Bayer's Legacy: Joint Remosaicing and Denoising for Quad Bayer Image Sensor

Joint Demosaicking And Blind Deblurring Using Deep Convolutional Neural Network

Residual Learning for Effective Joint Demosaicing-Denoising

Learning spectral-spatial prior via 3ddncnn for hyperspectral image deconvolution

Mutual Guidance Distillation for Joint Demosaicking and Denoising of Raw Images

Joint Image Denoising with Gradient Direction and Edge-Preserving Regularization

An Efficient Transformer For Demosaicing Via Compressed Multi-Branch Attention Mechanism.

Joint demosaicing and denoising of noisy bayer images with ADMM

Joint Demosaicking and Denoising Benefits from a Two-stage Training Strategy