Lossless image compression using reversible integer wavelet transforms and convolutional neural networks

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In this work we introduce a lossless compression framework which incorporates convolutional neural networks (CNN) for wavelet subband prediction. A CNN is trained to predict detail coefficients from corresponding approximation coefficients, prediction error is then coded in place of wavelet coefficients. At decompression an identical CNN is used to reproduce the prediction and combine with the decoded residuals for perfect reconstruction of wavelet subbands.

Original languageEnglish (US)
Title of host publicationProceedings - DCC 2018
Subtitle of host publication2018 Data Compression Conference
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages1
Volume2018-March
ISBN (Electronic)9781538648834
DOIs
StatePublished - Jul 19 2018
Event2018 Data Compression Conference, DCC 2018 - Snowbird, United States
Duration: Mar 27 2018Mar 30 2018

Other

Other2018 Data Compression Conference, DCC 2018
CountryUnited States
CitySnowbird
Period3/27/183/30/18

Fingerprint

Image compression
Wavelet transforms
Neural networks

Keywords

  • cnn
  • compression
  • convolutional neural network
  • deep learning
  • wavelets

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Ahanonu, E., Marcellin, M. W., & Bilgin, A. (2018). Lossless image compression using reversible integer wavelet transforms and convolutional neural networks. In Proceedings - DCC 2018: 2018 Data Compression Conference (Vol. 2018-March). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/DCC.2018.00048

Lossless image compression using reversible integer wavelet transforms and convolutional neural networks. / Ahanonu, Eze; Marcellin, Michael W; Bilgin, Ali.

Proceedings - DCC 2018: 2018 Data Compression Conference. Vol. 2018-March Institute of Electrical and Electronics Engineers Inc., 2018.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ahanonu, E, Marcellin, MW & Bilgin, A 2018, Lossless image compression using reversible integer wavelet transforms and convolutional neural networks. in Proceedings - DCC 2018: 2018 Data Compression Conference. vol. 2018-March, Institute of Electrical and Electronics Engineers Inc., 2018 Data Compression Conference, DCC 2018, Snowbird, United States, 3/27/18. https://doi.org/10.1109/DCC.2018.00048
Ahanonu E, Marcellin MW, Bilgin A. Lossless image compression using reversible integer wavelet transforms and convolutional neural networks. In Proceedings - DCC 2018: 2018 Data Compression Conference. Vol. 2018-March. Institute of Electrical and Electronics Engineers Inc. 2018 https://doi.org/10.1109/DCC.2018.00048
Ahanonu, Eze ; Marcellin, Michael W ; Bilgin, Ali. / Lossless image compression using reversible integer wavelet transforms and convolutional neural networks. Proceedings - DCC 2018: 2018 Data Compression Conference. Vol. 2018-March Institute of Electrical and Electronics Engineers Inc., 2018.
@inproceedings{cb9b053e825045ebb7c93c8e622ca858,
title = "Lossless image compression using reversible integer wavelet transforms and convolutional neural networks",
abstract = "In this work we introduce a lossless compression framework which incorporates convolutional neural networks (CNN) for wavelet subband prediction. A CNN is trained to predict detail coefficients from corresponding approximation coefficients, prediction error is then coded in place of wavelet coefficients. At decompression an identical CNN is used to reproduce the prediction and combine with the decoded residuals for perfect reconstruction of wavelet subbands.",
keywords = "cnn, compression, convolutional neural network, deep learning, wavelets",
author = "Eze Ahanonu and Marcellin, {Michael W} and Ali Bilgin",
year = "2018",
month = "7",
day = "19",
doi = "10.1109/DCC.2018.00048",
language = "English (US)",
volume = "2018-March",
booktitle = "Proceedings - DCC 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Lossless image compression using reversible integer wavelet transforms and convolutional neural networks

AU - Ahanonu, Eze

AU - Marcellin, Michael W

AU - Bilgin, Ali

PY - 2018/7/19

Y1 - 2018/7/19

N2 - In this work we introduce a lossless compression framework which incorporates convolutional neural networks (CNN) for wavelet subband prediction. A CNN is trained to predict detail coefficients from corresponding approximation coefficients, prediction error is then coded in place of wavelet coefficients. At decompression an identical CNN is used to reproduce the prediction and combine with the decoded residuals for perfect reconstruction of wavelet subbands.

AB - In this work we introduce a lossless compression framework which incorporates convolutional neural networks (CNN) for wavelet subband prediction. A CNN is trained to predict detail coefficients from corresponding approximation coefficients, prediction error is then coded in place of wavelet coefficients. At decompression an identical CNN is used to reproduce the prediction and combine with the decoded residuals for perfect reconstruction of wavelet subbands.

KW - cnn

KW - compression

KW - convolutional neural network

KW - deep learning

KW - wavelets

UR - http://www.scopus.com/inward/record.url?scp=85050968133&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050968133&partnerID=8YFLogxK

U2 - 10.1109/DCC.2018.00048

DO - 10.1109/DCC.2018.00048

M3 - Conference contribution

VL - 2018-March

BT - Proceedings - DCC 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -