The Effect of the Color Filter Array Layout Choice on State-of-the-Art Demosaicing

Abstract

Interpolation from a Color Filter Array (CFA) is the most common method for obtaining full color image data. Its success relies on the smart combination of a CFA and a demosaicing algorithm. Demosaicing on the one hand has been extensively studied. Algorithmic development in the past 20 years ranges from simple linear interpolation to modern neural-network-based (NN) approaches that encode the prior knowledge of millions of training images to fill in missing data in an inconspicious way. CFA design, on the other hand, is less well studied, although still recognized to strongly impact demosaicing performance. This is because demosaicing algorithms are typically limited to one particular CFA pattern, impeding straightforward CFA comparison. This is starting to change with newer classes of demosaicing that may be considered generic or CFA-agnostic. In this study, by comparing performance of two state-of-the-art generic algorithms, we evaluate the potential of modern CFA-demosaicing. We test the hypothesis that, with the increasing power of NN-based demosaicing, the influence of optimal CFA design on system performance decreases. This hypothesis is supported with the experimental results. Such a finding would herald the possibility of relaxing CFA requirements, providing more freedom in the CFA design choice and producing high-quality cameras.

Adaptive Image Demosaicing Algorithm Based On K-Nearest Neighbor for Improved Visual Quality

“…(2) Similarly, the SNR is also a measure that estimates the quality of the reconstructed image with respect to the original image. The Formulas for evaluation are given by equation (1)(2)(3)(4)(5)(6)(7)(8)(9). Where s=65535, for the 16-bit image.…”

“…An article proposed an identified strategy that re-processes the analyzed image with eigen algorithms and builds a set of identifying features for the algorithm. [6] The demosaicing algorithms have the most efficient and quality results for color image acquisition, especially in noisy images. Researchers' challenges in the demosaicing algorithm are false-color artifacts, edge blur in images, and zippering.…”

Sensor Signal and Information Processing II

“…General demosaicing would be defined as the reconstruction of a multi-dimensional color signal from an inherently single-dimensional array of sensors. In [ 12 ], Stojkovic et al have tested the influence of Bayer-like CFA patterns with the same sampling ratio as the Bayer CFA, considering them to be sufficient to show that the difference in quality performance between different CFA designs decreases with the increasing power of demosaicing algorithms.…”

Efficient Training Procedures for Multi-Spectral Demosaicing

“…The selection of algorithms for comparison was limited to methods with publicly available code that used the uniform pattern, so as to ensure the experiment could be controlled. However, we expect that with powerful deep learning algorithms, the reliance on the pattern layout will not be strong [40], if the patterns have the same spectral sampling density.…”