Presentation 2007-10-18
Image compression using nonlinear prediction by RBF networks and DPCM
Keisuke NARISAWA, Takashi HOSHINO, Tohru IKEGUCHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Differential pulse code modulation (DPCM) is one of the most famous methods to realize lossless image compression. Due to strong autocorrelation structures in image data, the DPCM reduces redundancy of information in the image data and exhibits good compression performance. However, the prediction accuracy of the DPCM becomes worse around edges of the image because the DPCM uses a linear function to predict pixel values. To avoid such worse situation, we propose a new method which combines a radial basis function (RBF) network and the DPCM. The RBF network is a nonlinear prediction method. As a result, the proposed method shows higher prediction accuracy and compression rate than the original DPCM.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) DPCM / RBF / image compression / nonlinear prediction
Paper # CAS2007-39,NLP2007-67
Date of Issue

Conference Information
Committee NLP
Conference Date 2007/10/11(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 Nonlinear Problems (NLP)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Image compression using nonlinear prediction by RBF networks and DPCM
Sub Title (in English)
Keyword(1) DPCM
Keyword(2) RBF
Keyword(3) image compression
Keyword(4) nonlinear prediction
1st Author's Name Keisuke NARISAWA
1st Author's Affiliation Graduate School of Science and Engineering, Saitama University()
2nd Author's Name Takashi HOSHINO
2nd Author's Affiliation Graduate School of Science and Engineering, Saitama University
3rd Author's Name Tohru IKEGUCHI
3rd Author's Affiliation Graduate School of Science and Engineering, Saitama University
Date 2007-10-18
Paper # CAS2007-39,NLP2007-67
Volume (vol) vol.107
Number (no) 266
Page pp.pp.-
#Pages 6
Date of Issue