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 journalReview articlepeer-review

117 Scopus citations


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
Issue number11
StatePublished - 1997


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

ASJC Scopus subject areas

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this