Adaptive transform coding as constrained vector quantization

Cynthia Archer, Todd K. Leen

Research output: Chapter in Book/Report/Conference proceedingChapter

1 Citation (Scopus)

Abstract

We investigate the application of local Principal Component Analysis (PCA) to transform coding for fixed-rate image compression. Local PCA transform coding adapts to differences in correlations between signal components by partitioning the signal space into regions and compressing signal vectors in each region with a separate local transform coder. Previous researchers optimize the signal space partition and transform coders independently and consequently underestimate the potential advantage of using adaptive transform coding methods. We propose a new algorithm that concurrently optimizes the signal space partition and local transform coders. This algorithm is simply a constrained version of the LBG algorithm for vector quantizer design. Image compression experiments show that adaptive transform coders designed with our integrated algorithm compress an image with less distortion than previous related methods. We saw improvements in compressed image signal-to-noise ratio of 0.5 to 2.0 dB compared to other tested adaptive methods and 2.5 to 3.0 dB compared to global PCA transform coding.

Original languageEnglish (US)
Title of host publicationNeural Networks for Signal Processing - Proceedings of the IEEE Workshop
PublisherIEEE
Pages308-317
Number of pages10
Volume1
StatePublished - 2000
Event10th IEEE Workshop on Neural Netwoks for Signal Processing (NNSP2000) - Sydney, Australia
Duration: Dec 11 2000Dec 13 2000

Other

Other10th IEEE Workshop on Neural Netwoks for Signal Processing (NNSP2000)
CitySydney, Australia
Period12/11/0012/13/00

Fingerprint

Vector quantization
Principal component analysis
Image compression
Signal to noise ratio
Experiments

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Archer, C., & Leen, T. K. (2000). Adaptive transform coding as constrained vector quantization. In Neural Networks for Signal Processing - Proceedings of the IEEE Workshop (Vol. 1, pp. 308-317). IEEE.

Adaptive transform coding as constrained vector quantization. / Archer, Cynthia; Leen, Todd K.

Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. Vol. 1 IEEE, 2000. p. 308-317.

Research output: Chapter in Book/Report/Conference proceedingChapter

Archer, C & Leen, TK 2000, Adaptive transform coding as constrained vector quantization. in Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. vol. 1, IEEE, pp. 308-317, 10th IEEE Workshop on Neural Netwoks for Signal Processing (NNSP2000), Sydney, Australia, 12/11/00.
Archer C, Leen TK. Adaptive transform coding as constrained vector quantization. In Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. Vol. 1. IEEE. 2000. p. 308-317
Archer, Cynthia ; Leen, Todd K. / Adaptive transform coding as constrained vector quantization. Neural Networks for Signal Processing - Proceedings of the IEEE Workshop. Vol. 1 IEEE, 2000. pp. 308-317
@inbook{7b1ba976b25d450381f1caa0b67b04d4,
title = "Adaptive transform coding as constrained vector quantization",
abstract = "We investigate the application of local Principal Component Analysis (PCA) to transform coding for fixed-rate image compression. Local PCA transform coding adapts to differences in correlations between signal components by partitioning the signal space into regions and compressing signal vectors in each region with a separate local transform coder. Previous researchers optimize the signal space partition and transform coders independently and consequently underestimate the potential advantage of using adaptive transform coding methods. We propose a new algorithm that concurrently optimizes the signal space partition and local transform coders. This algorithm is simply a constrained version of the LBG algorithm for vector quantizer design. Image compression experiments show that adaptive transform coders designed with our integrated algorithm compress an image with less distortion than previous related methods. We saw improvements in compressed image signal-to-noise ratio of 0.5 to 2.0 dB compared to other tested adaptive methods and 2.5 to 3.0 dB compared to global PCA transform coding.",
author = "Cynthia Archer and Leen, {Todd K.}",
year = "2000",
language = "English (US)",
volume = "1",
pages = "308--317",
booktitle = "Neural Networks for Signal Processing - Proceedings of the IEEE Workshop",
publisher = "IEEE",

}

TY - CHAP

T1 - Adaptive transform coding as constrained vector quantization

AU - Archer, Cynthia

AU - Leen, Todd K.

PY - 2000

Y1 - 2000

N2 - We investigate the application of local Principal Component Analysis (PCA) to transform coding for fixed-rate image compression. Local PCA transform coding adapts to differences in correlations between signal components by partitioning the signal space into regions and compressing signal vectors in each region with a separate local transform coder. Previous researchers optimize the signal space partition and transform coders independently and consequently underestimate the potential advantage of using adaptive transform coding methods. We propose a new algorithm that concurrently optimizes the signal space partition and local transform coders. This algorithm is simply a constrained version of the LBG algorithm for vector quantizer design. Image compression experiments show that adaptive transform coders designed with our integrated algorithm compress an image with less distortion than previous related methods. We saw improvements in compressed image signal-to-noise ratio of 0.5 to 2.0 dB compared to other tested adaptive methods and 2.5 to 3.0 dB compared to global PCA transform coding.

AB - We investigate the application of local Principal Component Analysis (PCA) to transform coding for fixed-rate image compression. Local PCA transform coding adapts to differences in correlations between signal components by partitioning the signal space into regions and compressing signal vectors in each region with a separate local transform coder. Previous researchers optimize the signal space partition and transform coders independently and consequently underestimate the potential advantage of using adaptive transform coding methods. We propose a new algorithm that concurrently optimizes the signal space partition and local transform coders. This algorithm is simply a constrained version of the LBG algorithm for vector quantizer design. Image compression experiments show that adaptive transform coders designed with our integrated algorithm compress an image with less distortion than previous related methods. We saw improvements in compressed image signal-to-noise ratio of 0.5 to 2.0 dB compared to other tested adaptive methods and 2.5 to 3.0 dB compared to global PCA transform coding.

UR - http://www.scopus.com/inward/record.url?scp=0034497276&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034497276&partnerID=8YFLogxK

M3 - Chapter

AN - SCOPUS:0034497276

VL - 1

SP - 308

EP - 317

BT - Neural Networks for Signal Processing - Proceedings of the IEEE Workshop

PB - IEEE

ER -