Optimal rate allocation for joint compression and classification in JPEG2000

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

1 Citation (Scopus)

Abstract

We present a framework for optimal rate allocation to image subbands to minimize the distortion in the joint compression and classification of JPEG2000-compressed images. The distortion due to compression is defined as a weighted linear combination of the mean-square error (MSE) and the loss in the Bhattacharyya distance (BD) between the class-conditional distributions of the classes. Lossy compression with JPEG2000 is accomplished via deadzone uniform quantization of wavelet subbands. Neglecting the effect of the deadzone, expressions are derived for the distortion in the case of two classes with generalized Gaussian distributions (GGDs), based on the high-rate analysis of Poor. In this regime, the distortion function takes the form of a weighted MSE (WMSE) function, which can be minimized using reverse water-filling. We present experimental results based on synthetic data to evaluate the efficacy of the proposed rate allocation scheme. The results indicate that by varying the weight factor balancing the MSE and the Bhattacharyya distance, we can control the trade-off between these two terms in the distortion function.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsA.G. Tescher
Pages260-267
Number of pages8
Volume5558
EditionPART 1
DOIs
StatePublished - 2004
EventApplications of Digital Image Processing XXVII - Denver, CO, United States
Duration: Aug 2 2004Aug 6 2004

Other

OtherApplications of Digital Image Processing XXVII
CountryUnited States
CityDenver, CO
Period8/2/048/6/04

Fingerprint

Mean square error
weight (mass)
error functions
Gaussian distribution
normal density functions
water
Water

Keywords

  • Generalized gaussian distribution
  • Joint compression and classification
  • JPEG2000
  • Rate allocation

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Tabesh, A., Marcellin, M. W., & Neifeld, M. A. (2004). Optimal rate allocation for joint compression and classification in JPEG2000. In A. G. Tescher (Ed.), Proceedings of SPIE - The International Society for Optical Engineering (PART 1 ed., Vol. 5558, pp. 260-267). [32] https://doi.org/10.1117/12.564824

Optimal rate allocation for joint compression and classification in JPEG2000. / Tabesh, Ali; Marcellin, Michael W; Neifeld, Mark A.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / A.G. Tescher. Vol. 5558 PART 1. ed. 2004. p. 260-267 32.

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

Tabesh, A, Marcellin, MW & Neifeld, MA 2004, Optimal rate allocation for joint compression and classification in JPEG2000. in AG Tescher (ed.), Proceedings of SPIE - The International Society for Optical Engineering. PART 1 edn, vol. 5558, 32, pp. 260-267, Applications of Digital Image Processing XXVII, Denver, CO, United States, 8/2/04. https://doi.org/10.1117/12.564824
Tabesh A, Marcellin MW, Neifeld MA. Optimal rate allocation for joint compression and classification in JPEG2000. In Tescher AG, editor, Proceedings of SPIE - The International Society for Optical Engineering. PART 1 ed. Vol. 5558. 2004. p. 260-267. 32 https://doi.org/10.1117/12.564824
Tabesh, Ali ; Marcellin, Michael W ; Neifeld, Mark A. / Optimal rate allocation for joint compression and classification in JPEG2000. Proceedings of SPIE - The International Society for Optical Engineering. editor / A.G. Tescher. Vol. 5558 PART 1. ed. 2004. pp. 260-267
@inproceedings{8f7fc651bae44c4e87717c646e737d24,
title = "Optimal rate allocation for joint compression and classification in JPEG2000",
abstract = "We present a framework for optimal rate allocation to image subbands to minimize the distortion in the joint compression and classification of JPEG2000-compressed images. The distortion due to compression is defined as a weighted linear combination of the mean-square error (MSE) and the loss in the Bhattacharyya distance (BD) between the class-conditional distributions of the classes. Lossy compression with JPEG2000 is accomplished via deadzone uniform quantization of wavelet subbands. Neglecting the effect of the deadzone, expressions are derived for the distortion in the case of two classes with generalized Gaussian distributions (GGDs), based on the high-rate analysis of Poor. In this regime, the distortion function takes the form of a weighted MSE (WMSE) function, which can be minimized using reverse water-filling. We present experimental results based on synthetic data to evaluate the efficacy of the proposed rate allocation scheme. The results indicate that by varying the weight factor balancing the MSE and the Bhattacharyya distance, we can control the trade-off between these two terms in the distortion function.",
keywords = "Generalized gaussian distribution, Joint compression and classification, JPEG2000, Rate allocation",
author = "Ali Tabesh and Marcellin, {Michael W} and Neifeld, {Mark A}",
year = "2004",
doi = "10.1117/12.564824",
language = "English (US)",
volume = "5558",
pages = "260--267",
editor = "A.G. Tescher",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",
edition = "PART 1",

}

TY - GEN

T1 - Optimal rate allocation for joint compression and classification in JPEG2000

AU - Tabesh, Ali

AU - Marcellin, Michael W

AU - Neifeld, Mark A

PY - 2004

Y1 - 2004

N2 - We present a framework for optimal rate allocation to image subbands to minimize the distortion in the joint compression and classification of JPEG2000-compressed images. The distortion due to compression is defined as a weighted linear combination of the mean-square error (MSE) and the loss in the Bhattacharyya distance (BD) between the class-conditional distributions of the classes. Lossy compression with JPEG2000 is accomplished via deadzone uniform quantization of wavelet subbands. Neglecting the effect of the deadzone, expressions are derived for the distortion in the case of two classes with generalized Gaussian distributions (GGDs), based on the high-rate analysis of Poor. In this regime, the distortion function takes the form of a weighted MSE (WMSE) function, which can be minimized using reverse water-filling. We present experimental results based on synthetic data to evaluate the efficacy of the proposed rate allocation scheme. The results indicate that by varying the weight factor balancing the MSE and the Bhattacharyya distance, we can control the trade-off between these two terms in the distortion function.

AB - We present a framework for optimal rate allocation to image subbands to minimize the distortion in the joint compression and classification of JPEG2000-compressed images. The distortion due to compression is defined as a weighted linear combination of the mean-square error (MSE) and the loss in the Bhattacharyya distance (BD) between the class-conditional distributions of the classes. Lossy compression with JPEG2000 is accomplished via deadzone uniform quantization of wavelet subbands. Neglecting the effect of the deadzone, expressions are derived for the distortion in the case of two classes with generalized Gaussian distributions (GGDs), based on the high-rate analysis of Poor. In this regime, the distortion function takes the form of a weighted MSE (WMSE) function, which can be minimized using reverse water-filling. We present experimental results based on synthetic data to evaluate the efficacy of the proposed rate allocation scheme. The results indicate that by varying the weight factor balancing the MSE and the Bhattacharyya distance, we can control the trade-off between these two terms in the distortion function.

KW - Generalized gaussian distribution

KW - Joint compression and classification

KW - JPEG2000

KW - Rate allocation

UR - http://www.scopus.com/inward/record.url?scp=14944353737&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=14944353737&partnerID=8YFLogxK

U2 - 10.1117/12.564824

DO - 10.1117/12.564824

M3 - Conference contribution

AN - SCOPUS:14944353737

VL - 5558

SP - 260

EP - 267

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Tescher, A.G.

ER -