Comparison of different methods of classification in subband coding of images

Rajan L. Joshi, Hamid Jafarkhani, James H. Kasner, Thomas R. Fischer, Nariman Farvardin, Michael W Marcellin, Roberto H. Bamberger

Research output: Contribution to journalArticle

117 Citations (Scopus)

Abstract

This paper investigates various classification techniques, applied to subband coding of images, as a way of exploiting the nonstationary nature of image subbands. The advantages of subband classification are characterized in a rate-distortion framework in terms of "classification gain" and overall "subband classification gain." Two algorithms, maximum classification gain and equal mean-normalized standard deviation classification, which allow unequal number of blocks in each class, are presented. The dependence between the classification maps from different subbands is exploited either directly while encoding the classification maps or indirectly by constraining the classification maps. The trade-off between the classification gain and the amount of side information is explored. Coding results for a subband image coder based on classification are presented. The simulation results demonstrate the value of classification in subband coding.

Original languageEnglish (US)
Pages (from-to)1473-1486
Number of pages14
JournalIEEE Transactions on Image Processing
Volume6
Issue number11
DOIs
StatePublished - 1997

Fingerprint

Coding
Side Information
Rate-distortion
Unequal
Standard deviation
Encoding
Trade-offs

Keywords

  • Arithmetic code
  • Classification
  • Image coding
  • Subband
  • Trellis-coded quantization
  • Wavelet

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Graphics and Computer-Aided Design
  • Software
  • Theoretical Computer Science
  • Computational Theory and Mathematics
  • Computer Vision and Pattern Recognition

Cite this

Joshi, R. L., Jafarkhani, H., Kasner, J. H., Fischer, T. R., Farvardin, N., Marcellin, M. W., & Bamberger, R. H. (1997). Comparison of different methods of classification in subband coding of images. IEEE Transactions on Image Processing, 6(11), 1473-1486. https://doi.org/10.1109/83.641409

Comparison of different methods of classification in subband coding of images. / Joshi, Rajan L.; Jafarkhani, Hamid; Kasner, James H.; Fischer, Thomas R.; Farvardin, Nariman; Marcellin, Michael W; Bamberger, Roberto H.

In: IEEE Transactions on Image Processing, Vol. 6, No. 11, 1997, p. 1473-1486.

Research output: Contribution to journalArticle

Joshi, RL, Jafarkhani, H, Kasner, JH, Fischer, TR, Farvardin, N, Marcellin, MW & Bamberger, RH 1997, 'Comparison of different methods of classification in subband coding of images', IEEE Transactions on Image Processing, vol. 6, no. 11, pp. 1473-1486. https://doi.org/10.1109/83.641409
Joshi, Rajan L. ; Jafarkhani, Hamid ; Kasner, James H. ; Fischer, Thomas R. ; Farvardin, Nariman ; Marcellin, Michael W ; Bamberger, Roberto H. / Comparison of different methods of classification in subband coding of images. In: IEEE Transactions on Image Processing. 1997 ; Vol. 6, No. 11. pp. 1473-1486.
@article{622ffe2eaacc4ff6a3185c56d9540cce,
title = "Comparison of different methods of classification in subband coding of images",
abstract = "This paper investigates various classification techniques, applied to subband coding of images, as a way of exploiting the nonstationary nature of image subbands. The advantages of subband classification are characterized in a rate-distortion framework in terms of {"}classification gain{"} and overall {"}subband classification gain.{"} Two algorithms, maximum classification gain and equal mean-normalized standard deviation classification, which allow unequal number of blocks in each class, are presented. The dependence between the classification maps from different subbands is exploited either directly while encoding the classification maps or indirectly by constraining the classification maps. The trade-off between the classification gain and the amount of side information is explored. Coding results for a subband image coder based on classification are presented. The simulation results demonstrate the value of classification in subband coding.",
keywords = "Arithmetic code, Classification, Image coding, Subband, Trellis-coded quantization, Wavelet",
author = "Joshi, {Rajan L.} and Hamid Jafarkhani and Kasner, {James H.} and Fischer, {Thomas R.} and Nariman Farvardin and Marcellin, {Michael W} and Bamberger, {Roberto H.}",
year = "1997",
doi = "10.1109/83.641409",
language = "English (US)",
volume = "6",
pages = "1473--1486",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "11",

}

TY - JOUR

T1 - Comparison of different methods of classification in subband coding of images

AU - Joshi, Rajan L.

AU - Jafarkhani, Hamid

AU - Kasner, James H.

AU - Fischer, Thomas R.

AU - Farvardin, Nariman

AU - Marcellin, Michael W

AU - Bamberger, Roberto H.

PY - 1997

Y1 - 1997

N2 - This paper investigates various classification techniques, applied to subband coding of images, as a way of exploiting the nonstationary nature of image subbands. The advantages of subband classification are characterized in a rate-distortion framework in terms of "classification gain" and overall "subband classification gain." Two algorithms, maximum classification gain and equal mean-normalized standard deviation classification, which allow unequal number of blocks in each class, are presented. The dependence between the classification maps from different subbands is exploited either directly while encoding the classification maps or indirectly by constraining the classification maps. The trade-off between the classification gain and the amount of side information is explored. Coding results for a subband image coder based on classification are presented. The simulation results demonstrate the value of classification in subband coding.

AB - This paper investigates various classification techniques, applied to subband coding of images, as a way of exploiting the nonstationary nature of image subbands. The advantages of subband classification are characterized in a rate-distortion framework in terms of "classification gain" and overall "subband classification gain." Two algorithms, maximum classification gain and equal mean-normalized standard deviation classification, which allow unequal number of blocks in each class, are presented. The dependence between the classification maps from different subbands is exploited either directly while encoding the classification maps or indirectly by constraining the classification maps. The trade-off between the classification gain and the amount of side information is explored. Coding results for a subband image coder based on classification are presented. The simulation results demonstrate the value of classification in subband coding.

KW - Arithmetic code

KW - Classification

KW - Image coding

KW - Subband

KW - Trellis-coded quantization

KW - Wavelet

UR - http://www.scopus.com/inward/record.url?scp=0031275763&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031275763&partnerID=8YFLogxK

U2 - 10.1109/83.641409

DO - 10.1109/83.641409

M3 - Article

VL - 6

SP - 1473

EP - 1486

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 11

ER -