Abstract

Color demosaicking is used to reconstruct full color images from incomplete color filter array samples captured by cameras with a single sensor array. In reconstructing natural-looking images, one key challenge is to model and respect the statistics of natural images. This paper presents a novel modeling strategy and an efficient color demosaicking algorithm. The approach starts with joint modeling of the color images, which supports simultaneous representation of inter-channel correlation and structural information in an image. The inter-channel correlation is explored by measuring the channel difference signals in the gradient domain, while the structural information is explored by nonlocal low-rank regularization. An efficient algorithm is then proposed to solve the joint formulation, by dividing the minimization problem into two sub-problems and solving them iteratively. The effectiveness of the proposed approach is demonstrated with extensive experiments on both noiseless and noisy datasets, with comparison with existing state-of-the-arts color demosaicking methods.

Original languageEnglish (US)
Pages (from-to)264-279
Number of pages16
JournalSignal Processing: Image Communication
Volume39
DOIs
StatePublished - Nov 1 2015

Fingerprint

Color
Sensor arrays
Cameras
Statistics
Experiments

Keywords

  • Color demosaicking
  • Inter-channel correlation
  • Joint modeling
  • Nonlocal self-similarity

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Color image demosaicking using inter-channel correlation and nonlocal self-similarity. / Chang, Kan; Ding, Pak Lun Kevin; Li, Baoxin.

In: Signal Processing: Image Communication, Vol. 39, 01.11.2015, p. 264-279.

Research output: Contribution to journalArticle

@article{2a9455bb42ae49349a38278171050258,
title = "Color image demosaicking using inter-channel correlation and nonlocal self-similarity",
abstract = "Color demosaicking is used to reconstruct full color images from incomplete color filter array samples captured by cameras with a single sensor array. In reconstructing natural-looking images, one key challenge is to model and respect the statistics of natural images. This paper presents a novel modeling strategy and an efficient color demosaicking algorithm. The approach starts with joint modeling of the color images, which supports simultaneous representation of inter-channel correlation and structural information in an image. The inter-channel correlation is explored by measuring the channel difference signals in the gradient domain, while the structural information is explored by nonlocal low-rank regularization. An efficient algorithm is then proposed to solve the joint formulation, by dividing the minimization problem into two sub-problems and solving them iteratively. The effectiveness of the proposed approach is demonstrated with extensive experiments on both noiseless and noisy datasets, with comparison with existing state-of-the-arts color demosaicking methods.",
keywords = "Color demosaicking, Inter-channel correlation, Joint modeling, Nonlocal self-similarity",
author = "Kan Chang and Ding, {Pak Lun Kevin} and Baoxin Li",
year = "2015",
month = "11",
day = "1",
doi = "10.1016/j.image.2015.10.003",
language = "English (US)",
volume = "39",
pages = "264--279",
journal = "Signal Processing: Image Communication",
issn = "0923-5965",
publisher = "Elsevier",

}

TY - JOUR

T1 - Color image demosaicking using inter-channel correlation and nonlocal self-similarity

AU - Chang, Kan

AU - Ding, Pak Lun Kevin

AU - Li, Baoxin

PY - 2015/11/1

Y1 - 2015/11/1

N2 - Color demosaicking is used to reconstruct full color images from incomplete color filter array samples captured by cameras with a single sensor array. In reconstructing natural-looking images, one key challenge is to model and respect the statistics of natural images. This paper presents a novel modeling strategy and an efficient color demosaicking algorithm. The approach starts with joint modeling of the color images, which supports simultaneous representation of inter-channel correlation and structural information in an image. The inter-channel correlation is explored by measuring the channel difference signals in the gradient domain, while the structural information is explored by nonlocal low-rank regularization. An efficient algorithm is then proposed to solve the joint formulation, by dividing the minimization problem into two sub-problems and solving them iteratively. The effectiveness of the proposed approach is demonstrated with extensive experiments on both noiseless and noisy datasets, with comparison with existing state-of-the-arts color demosaicking methods.

AB - Color demosaicking is used to reconstruct full color images from incomplete color filter array samples captured by cameras with a single sensor array. In reconstructing natural-looking images, one key challenge is to model and respect the statistics of natural images. This paper presents a novel modeling strategy and an efficient color demosaicking algorithm. The approach starts with joint modeling of the color images, which supports simultaneous representation of inter-channel correlation and structural information in an image. The inter-channel correlation is explored by measuring the channel difference signals in the gradient domain, while the structural information is explored by nonlocal low-rank regularization. An efficient algorithm is then proposed to solve the joint formulation, by dividing the minimization problem into two sub-problems and solving them iteratively. The effectiveness of the proposed approach is demonstrated with extensive experiments on both noiseless and noisy datasets, with comparison with existing state-of-the-arts color demosaicking methods.

KW - Color demosaicking

KW - Inter-channel correlation

KW - Joint modeling

KW - Nonlocal self-similarity

UR - http://www.scopus.com/inward/record.url?scp=84947928782&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947928782&partnerID=8YFLogxK

U2 - 10.1016/j.image.2015.10.003

DO - 10.1016/j.image.2015.10.003

M3 - Article

AN - SCOPUS:84947928782

VL - 39

SP - 264

EP - 279

JO - Signal Processing: Image Communication

JF - Signal Processing: Image Communication

SN - 0923-5965

ER -