Random Color Filter Arrays are Better than Regular Ones

Amba, Prakhar; Dias, Jérôme; Alleysson, David

doi:10.2352/issn.2169-2629.2017.32.294

Cited by 7 publications

(13 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides these mainstream CFA design frameworks, there are a number of other CFA designs such as [23], [26], [27], [30]. On the hardware side, Biay-Cheng et al [36], [37] took into account that color filter fabrication technology lags the image sensor technology in terms of miniaturization.…”

Section: Related Workmentioning

confidence: 99%

Color Filter Arrays for Quanta Image Sensors

Elgendy¹,

Chan

2020

IEEE Trans. Comput. Imaging

View full text Add to dashboard Cite

Quanta image sensor (QIS) is envisioned to be the next generation image sensor after CCD and CMOS. In this paper, we discuss how to design color filter arrays for QIS and other small pixels. Designing color filter arrays for small pixels is challenging because maximizing the light efficiency while suppressing aliasing and crosstalk are conflicting tasks. We present an optimization-based framework which unifies several mainstream color filter array design methodologies. Our method offers greater generality and flexibility. Compared to existing methods, the new framework can simultaneously handle luminance sensitivity, chrominance sensitivity, cross-talk, antialiasing, manufacturability and orthogonality. Extensive experimental comparisons demonstrate the effectiveness of the framework. Spatial CFA Design: By suppressing the Moiré artifacts and crosstalk while keeping the demosaicing algorithm simple, Lukac and Plataniotis [20] proposed a CFA and compared it with other CFAs using a universal demosaicking method. However, their work did not provide a mathematical framework to analyze the CFA optimality. AbstractThis supplementary report provides the following additional information of the main article.

show abstract

Section: Related Workmentioning

confidence: 99%

Color Filter Arrays for Quanta Image Sensors

Elgendy¹,

Chan

2020

IEEE Trans. Comput. Imaging

View full text Add to dashboard Cite

show abstract

“…We used a neighborhood of size 10 as it gives the best trade off between performance and computational complexity. 23 sPSNR and sSSIM 73 are used to compare the reconstructed image to the full resolution image. Results of PSNR and SSIM for the benchmark are reported in Table I, while results for the N-LMMSE are reported in Table II.…”

Section: Quantitative Analysis In Simulationmentioning

confidence: 99%

“…Among the others, most require heavy processing or iteration, which breaks the interesting potential of SFA for real-time robotic applications. One very interesting result recently published is the introduction of neighborhood in the LMMSE (N-LMMSE) formulation by Amba et al 23 LMMSE has been developed for CFA [63][64][65][66] with great success. It has been proven to be a very good general solution in the evaluations 23,39 on both CFA and SFA.…”

Section: Introductionmentioning

confidence: 99%

“…One very interesting result recently published is the introduction of neighborhood in the LMMSE (N-LMMSE) formulation by Amba et al 23 LMMSE has been developed for CFA [63][64][65][66] with great success. It has been proven to be a very good general solution in the evaluations 23,39 on both CFA and SFA. This formulation provides a potentially very good candidate for real-time applications since, after training, it could be embedded into the camera hardware and perform real time without losing the generality required by the different layouts present in the market.…”

Section: Introductionmentioning

confidence: 99%

“…However, scientific communications related to this problem are still published regularly. 19 Spectral sensitivities 20 and spatial layout [21][22][23] influence the demosaicing greatly. Other factors, such as optical aberrations, 24 also contribute.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

N-LMMSE Demosaicing for Spectral Filter Arrays

Amba¹,

Thomas²,

Alleysson³

2017

jist

Self Cite

View full text Add to dashboard Cite

Abstract. Spectral filter array (SFA) technology requires development on demosaicing. The authors extend the linear minimum mean square error with neighborhood method to the spectral dimension. They demonstrate that the method is fast and general on Raw SFA images that span the visible and near infra-red part of the electromagnetic range. The method is quantitatively evaluated in simulation first, then the authors evaluate it on real data by the use of non-reference image quality metrics applied on each band. Resulting images show a much better reconstruction of text and high frequencies at the expense of a zipping effect, compared to the benchmark binary-tree method.

show abstract

CNN Color Demosaicking Generalizes for any CFA

Bourrier¹,

Mermillod²,

Reyboz³

et al. 2022

cic

View full text Add to dashboard Cite

A convolutional neural network is trained in auto/heteroassociative mode for reconstructing RGB components from a randomly mosaicked color image. The trained network was shown to perform equally well when images are sampled periodically or with a different random mosaic. Therefore, this model is able to generalize on every type of color filter array. We attribute this property of universal demosaicking to the network learning the statistical structure of color images independently of the mosaic pattern arrangement.

show abstract

Random Color Filter Arrays are Better than Regular Ones

Cited by 7 publications

References 0 publications

Color Filter Arrays for Quanta Image Sensors

Color Filter Arrays for Quanta Image Sensors

N-LMMSE Demosaicing for Spectral Filter Arrays

CNN Color Demosaicking Generalizes for any CFA

Contact Info

Product

Resources

About