Sparse Representation of 3D Images for Piecewise Dimensionality Reduction with High Quality Reconstruction

Laura Rebollo-Neira, Daniel Whitehouse

Research output: Contribution to journalArticle

Abstract

Sparse representation of 3D images is considered within the context of data reduction. The goal is to produce high quality approximations of 3D images using fewer elementary components than the number of intensity points in the 3D array. This is achieved by means of a highly redundant dictionary and a dedicated pursuit strategy especially designed for low memory requirements. The benefit of the proposed framework is illustrated in the first instance by demonstrating the gain in dimensionality reduction obtained when approximating true color images as very thin 3D arrays, instead of performing an independent channel by channel approximation. The full power of the approach is further exemplified by producing high quality approximations of hyper-spectral images with a reduction of up to 371 times the number of data points in the representation.
Original languageEnglish
JournalArray
Early online date27 Jun 2019
DOIs
Publication statusE-pub ahead of print - 27 Jun 2019

Fingerprint

Glossaries
Data reduction
Color
Data storage equipment

Bibliographical note

Creative Commons Attribution 4.0 International (CC BY 4.0)

Cite this

@article{8ebbffc0aee14dc0916a2c13f46dca1f,
title = "Sparse Representation of 3D Images for Piecewise Dimensionality Reduction with High Quality Reconstruction",
abstract = "Sparse representation of 3D images is considered within the context of data reduction. The goal is to produce high quality approximations of 3D images using fewer elementary components than the number of intensity points in the 3D array. This is achieved by means of a highly redundant dictionary and a dedicated pursuit strategy especially designed for low memory requirements. The benefit of the proposed framework is illustrated in the first instance by demonstrating the gain in dimensionality reduction obtained when approximating true color images as very thin 3D arrays, instead of performing an independent channel by channel approximation. The full power of the approach is further exemplified by producing high quality approximations of hyper-spectral images with a reduction of up to 371 times the number of data points in the representation.",
author = "Laura Rebollo-Neira and Daniel Whitehouse",
note = "Creative Commons Attribution 4.0 International (CC BY 4.0)",
year = "2019",
month = "6",
day = "27",
doi = "10.1016/j.array.2019.100001",
language = "English",

}

Sparse Representation of 3D Images for Piecewise Dimensionality Reduction with High Quality Reconstruction. / Rebollo-Neira, Laura ; Whitehouse, Daniel .

In: Array , 27.06.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Sparse Representation of 3D Images for Piecewise Dimensionality Reduction with High Quality Reconstruction

AU - Rebollo-Neira, Laura

AU - Whitehouse, Daniel

N1 - Creative Commons Attribution 4.0 International (CC BY 4.0)

PY - 2019/6/27

Y1 - 2019/6/27

N2 - Sparse representation of 3D images is considered within the context of data reduction. The goal is to produce high quality approximations of 3D images using fewer elementary components than the number of intensity points in the 3D array. This is achieved by means of a highly redundant dictionary and a dedicated pursuit strategy especially designed for low memory requirements. The benefit of the proposed framework is illustrated in the first instance by demonstrating the gain in dimensionality reduction obtained when approximating true color images as very thin 3D arrays, instead of performing an independent channel by channel approximation. The full power of the approach is further exemplified by producing high quality approximations of hyper-spectral images with a reduction of up to 371 times the number of data points in the representation.

AB - Sparse representation of 3D images is considered within the context of data reduction. The goal is to produce high quality approximations of 3D images using fewer elementary components than the number of intensity points in the 3D array. This is achieved by means of a highly redundant dictionary and a dedicated pursuit strategy especially designed for low memory requirements. The benefit of the proposed framework is illustrated in the first instance by demonstrating the gain in dimensionality reduction obtained when approximating true color images as very thin 3D arrays, instead of performing an independent channel by channel approximation. The full power of the approach is further exemplified by producing high quality approximations of hyper-spectral images with a reduction of up to 371 times the number of data points in the representation.

UR - https://www.sciencedirect.com/science/article/pii/S2590005619300013

U2 - 10.1016/j.array.2019.100001

DO - 10.1016/j.array.2019.100001

M3 - Article

ER -