Hierarchical image vector quantizer using gradient-Laplacian subspace distortion

Lai-man PO; Chok-ki CHAN

doi:10.1109/TENCON.1990.152656

Hierarchical image vector quantizer using gradient-Laplacian subspace distortion

Lai-man PO
, Chok-ki CHAN

Department of Electrical Engineering

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

Abstract

A new hierarchical mean-residual image vector quantization (HMRIVQ) technique using a general four-dimensional gradient-Laplacian subspace distortion (GLSD) is proposed. The GLSD is used for design and implementation of different block sizes of the HMRIVQ. Therefore, the computational and storage costs of the encoder are significantly reduced. The HMRIVQ can also achieve a very high compression rate for low detailed images. A performance comparison of the HMRIVQ using Euclidean distortion and the general four-dimensional GLSD is also given. © 1990 IEEE

Original language	English
Title of host publication	IEEE TENCON'90: 1990 IEEE Region 10 Conference on Computer and Communication Systems
Subtitle of host publication	Conference proceedings
Publisher	IEEE
Pages	481-485
ISBN (Print)	0879425563
DOIs	https://doi.org/10.1109/TENCON.1990.152656
Publication status	Published - Sept 1990
Event	1990 IEEE Region 10 Conference on Computer and Communication Systems - IEEE TENCON '90 - Hong Kong Duration: 24 Sept 1990 → 27 Sept 1990

Conference

Conference	1990 IEEE Region 10 Conference on Computer and Communication Systems - IEEE TENCON '90
City	Hong Kong
Period	24/09/90 → 27/09/90

Access Output

10.1109/TENCON.1990.152656

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{9c7ebcbe2fd941438f18e2a3b9c46fc8,

title = "Hierarchical image vector quantizer using gradient-Laplacian subspace distortion",

abstract = "A new hierarchical mean-residual image vector quantization (HMRIVQ) technique using a general four-dimensional gradient-Laplacian subspace distortion (GLSD) is proposed. The GLSD is used for design and implementation of different block sizes of the HMRIVQ. Therefore, the computational and storage costs of the encoder are significantly reduced. The HMRIVQ can also achieve a very high compression rate for low detailed images. A performance comparison of the HMRIVQ using Euclidean distortion and the general four-dimensional GLSD is also given. {\textcopyright} 1990 IEEE",

author = "Lai-man PO and Chok-ki CHAN",

year = "1990",

month = sep,

doi = "10.1109/TENCON.1990.152656",

language = "English",

isbn = "0879425563",

pages = "481--485",

booktitle = "IEEE TENCON'90: 1990 IEEE Region 10 Conference on Computer and Communication Systems",

publisher = "IEEE",

address = "United States",

note = "1990 IEEE Region 10 Conference on Computer and Communication Systems - IEEE TENCON '90 ; Conference date: 24-09-1990 Through 27-09-1990",

}

PO, L & CHAN, C 1990, Hierarchical image vector quantizer using gradient-Laplacian subspace distortion. in IEEE TENCON'90: 1990 IEEE Region 10 Conference on Computer and Communication Systems: Conference proceedings. IEEE, pp. 481-485, 1990 IEEE Region 10 Conference on Computer and Communication Systems - IEEE TENCON '90, Hong Kong, 24/09/90. https://doi.org/10.1109/TENCON.1990.152656

Hierarchical image vector quantizer using gradient-Laplacian subspace distortion. / PO, Lai-man; CHAN, Chok-ki.
IEEE TENCON'90: 1990 IEEE Region 10 Conference on Computer and Communication Systems: Conference proceedings. IEEE, 1990. p. 481-485.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Hierarchical image vector quantizer using gradient-Laplacian subspace distortion

AU - PO, Lai-man

AU - CHAN, Chok-ki

PY - 1990/9

Y1 - 1990/9

N2 - A new hierarchical mean-residual image vector quantization (HMRIVQ) technique using a general four-dimensional gradient-Laplacian subspace distortion (GLSD) is proposed. The GLSD is used for design and implementation of different block sizes of the HMRIVQ. Therefore, the computational and storage costs of the encoder are significantly reduced. The HMRIVQ can also achieve a very high compression rate for low detailed images. A performance comparison of the HMRIVQ using Euclidean distortion and the general four-dimensional GLSD is also given. © 1990 IEEE

AB - A new hierarchical mean-residual image vector quantization (HMRIVQ) technique using a general four-dimensional gradient-Laplacian subspace distortion (GLSD) is proposed. The GLSD is used for design and implementation of different block sizes of the HMRIVQ. Therefore, the computational and storage costs of the encoder are significantly reduced. The HMRIVQ can also achieve a very high compression rate for low detailed images. A performance comparison of the HMRIVQ using Euclidean distortion and the general four-dimensional GLSD is also given. © 1990 IEEE

UR - https://www.scopus.com/pages/publications/0025533042

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-0025533042&origin=recordpage

U2 - 10.1109/TENCON.1990.152656

DO - 10.1109/TENCON.1990.152656

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 0879425563

SP - 481

EP - 485

BT - IEEE TENCON'90: 1990 IEEE Region 10 Conference on Computer and Communication Systems

PB - IEEE

T2 - 1990 IEEE Region 10 Conference on Computer and Communication Systems - IEEE TENCON '90

Y2 - 24 September 1990 through 27 September 1990

ER -

Hierarchical image vector quantizer using gradient-Laplacian subspace distortion

Abstract

Conference

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this