Presentation 1997/12/19
2-Dimensional Optimum Frequency Band Partition With Human Visual Sensitivity for Subband Image Coding
Masashi KAMEDA, Kohhei OHTAKE, Makoto MIYAHARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents a 2-dimensional optimum frequency band partition method for subband image coding considering perceptual properties. Based on a set of the frequency pieces called the band blocks in 2-dimensional frequency domain, we propose the optimum band partition method which improves the quantization noise power weighted by the human visual sensitivity based on the band blocks. And, we show a filter bank configuration realizing it. Secondly, we derive the optimum band partition pattern in 2-dimensional frequency domain inherent to an input natural image. Finally, it is shown that the optimum band partition considering the human visual sensitivity is better than DCT which is used in JPEG coding system at the higher entropy rate using the PQS (Picture Quality Scale) as an evaluation measure of the picture quality considering perceptual properties. It is clarified that the proposed band partition method is effecitive.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) subband coding / quantization noise power / human visual sensitivity / PQS (Picture Quality Scale)
Paper # CQ97-65
Date of Issue

Conference Information
Committee CQ
Conference Date 1997/12/19(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Communication Quality (CQ)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) 2-Dimensional Optimum Frequency Band Partition With Human Visual Sensitivity for Subband Image Coding
Sub Title (in English)
Keyword(1) subband coding
Keyword(2) quantization noise power
Keyword(3) human visual sensitivity
Keyword(4) PQS (Picture Quality Scale)
1st Author's Name Masashi KAMEDA
1st Author's Affiliation Graduate School of Information Science, Japan Advanced Institute of Science and Technology (JAIST)()
2nd Author's Name Kohhei OHTAKE
2nd Author's Affiliation College of Engineering, Hosei University
3rd Author's Name Makoto MIYAHARA
3rd Author's Affiliation Graduate School of Information Science, Japan Advanced Institute of Science and Technology (JAIST)
Date 1997/12/19
Paper # CQ97-65
Volume (vol) vol.97
Number (no) 457
Page pp.pp.-
#Pages 8
Date of Issue