Presentation 2015-03-03
Derivation of important basis using ICA for improvement of coding performance in JPEG
Kazuya KAWAMURA, Masashi KAMEDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Discrete cosine transform (DCT) in JPEG is the one of effective image coding methods based on the orthogonal transform with the energy compaction. However, visual quality of the encoded image is deteriorated in the edge region when the data compression ratio is higher. On the other hand, Independent component analysis (ICA) can obtain a set of basis which corresponding to the structural feature in the a given input image, and the image coding using ICA has been developed by focusing on the sparseness of the ICA coefficients. In our study, hybrid transform image coding method using both ICA and DCT has been proposed. However, our previous method has been needed more additional information to preserve all ICA basis to send the decoder side. In order to solve such a problem, Important Basis Score (IBS) which is an evaluation measure for selection of the ICA basis in the view point of contribution to the coding performance improvement has been proposed.IBS However,IBS has a problem that can not determine the number of important ICA basis which is dependent with improvement of coding performance. In this paper, we propose Redundancy Basis Score (RBS) which is an evaluation measure for selection of the ICA basis in the view point of redundancy of ICA basis. In our experimental results, the number of important ICA basis which improve coding performance is determined, and coding performance of proposed method is slightly better than the method using only DCT.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Still image coding / Discrete cosine transform / Independent component analysis / Image segmentation
Paper # IMQ2014-42,IE2014-103,MVE2014-90
Date of Issue

Conference Information
Committee MVE
Conference Date 2015/2/24(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 Media Experience and Virtual Environment (MVE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Derivation of important basis using ICA for improvement of coding performance in JPEG
Sub Title (in English)
Keyword(1) Still image coding
Keyword(2) Discrete cosine transform
Keyword(3) Independent component analysis
Keyword(4) Image segmentation
1st Author's Name Kazuya KAWAMURA
1st Author's Affiliation Graduate School of Software and Information Science, Iwate Prefectural University()
2nd Author's Name Masashi KAMEDA
2nd Author's Affiliation Graduate School of Software and Information Science, Iwate Prefectural University
Date 2015-03-03
Paper # IMQ2014-42,IE2014-103,MVE2014-90
Volume (vol) vol.114
Number (no) 487
Page pp.pp.-
#Pages 6
Date of Issue