Compression of color facial images using feature correction two-stage vector quantization

Jincheng Huang; Yao Wang

doi:10.1109/83.736696

Compression of color facial images using feature correction two-stage vector quantization

Jincheng Huang, Yao Wang

Electrical and Computer Engineering

Research output: Contribution to journal › Article

Abstract

A feature correction two-stage vector quantization (FC2VQ) algorithm was previously developed to compress gray-scale photo identification (ID) pictures. This algorithm is extended to color images in this work. Three options are compared, which apply the FC2VQ algorithm in RGB, YCbCr, and Karhunen-Loeve transform (KLT) color spaces, respectively. The RGB-FC2VQ algorithm is found to yield better image quality than KLT-FC2VQ or YCbCr-FC2VQ at similar bit rates. With the RGB-FC2VQ algorithm, a 128 × 128 24-b color ID image (49 152 bytes) can be compressed down to about 500 bytes with satisfactory quality. When the codeword indices are further compressed losslessy using a first order Huffman coder, this size is further reduced to about 450 bytes.

Original language	English (US)
Pages (from-to)	102-109
Number of pages	8
Journal	IEEE Transactions on Image Processing
Volume	8
Issue number	1
DOIs	https://doi.org/10.1109/83.736696
State	Published - 1999

ASJC Scopus subject areas

Software
Computer Graphics and Computer-Aided Design

Access to Document

10.1109/83.736696

Fingerprint Dive into the research topics of 'Compression of color facial images using feature correction two-stage vector quantization'. Together they form a unique fingerprint.

Cite this

Huang, J., & Wang, Y. (1999). Compression of color facial images using feature correction two-stage vector quantization. IEEE Transactions on Image Processing, 8(1), 102-109. https://doi.org/10.1109/83.736696

@article{028884a47941466eb770b6d5086eb753,

title = "Compression of color facial images using feature correction two-stage vector quantization",

abstract = "A feature correction two-stage vector quantization (FC2VQ) algorithm was previously developed to compress gray-scale photo identification (ID) pictures. This algorithm is extended to color images in this work. Three options are compared, which apply the FC2VQ algorithm in RGB, YCbCr, and Karhunen-Loeve transform (KLT) color spaces, respectively. The RGB-FC2VQ algorithm is found to yield better image quality than KLT-FC2VQ or YCbCr-FC2VQ at similar bit rates. With the RGB-FC2VQ algorithm, a 128 × 128 24-b color ID image (49 152 bytes) can be compressed down to about 500 bytes with satisfactory quality. When the codeword indices are further compressed losslessy using a first order Huffman coder, this size is further reduced to about 450 bytes.",

author = "Jincheng Huang and Yao Wang",

year = "1999",

doi = "10.1109/83.736696",

language = "English (US)",

volume = "8",

pages = "102--109",

journal = "IEEE Transactions on Image Processing",

issn = "1057-7149",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "1",

}

TY - JOUR

T1 - Compression of color facial images using feature correction two-stage vector quantization

AU - Huang, Jincheng

AU - Wang, Yao

PY - 1999

Y1 - 1999

N2 - A feature correction two-stage vector quantization (FC2VQ) algorithm was previously developed to compress gray-scale photo identification (ID) pictures. This algorithm is extended to color images in this work. Three options are compared, which apply the FC2VQ algorithm in RGB, YCbCr, and Karhunen-Loeve transform (KLT) color spaces, respectively. The RGB-FC2VQ algorithm is found to yield better image quality than KLT-FC2VQ or YCbCr-FC2VQ at similar bit rates. With the RGB-FC2VQ algorithm, a 128 × 128 24-b color ID image (49 152 bytes) can be compressed down to about 500 bytes with satisfactory quality. When the codeword indices are further compressed losslessy using a first order Huffman coder, this size is further reduced to about 450 bytes.

AB - A feature correction two-stage vector quantization (FC2VQ) algorithm was previously developed to compress gray-scale photo identification (ID) pictures. This algorithm is extended to color images in this work. Three options are compared, which apply the FC2VQ algorithm in RGB, YCbCr, and Karhunen-Loeve transform (KLT) color spaces, respectively. The RGB-FC2VQ algorithm is found to yield better image quality than KLT-FC2VQ or YCbCr-FC2VQ at similar bit rates. With the RGB-FC2VQ algorithm, a 128 × 128 24-b color ID image (49 152 bytes) can be compressed down to about 500 bytes with satisfactory quality. When the codeword indices are further compressed losslessy using a first order Huffman coder, this size is further reduced to about 450 bytes.

UR - http://www.scopus.com/inward/record.url?scp=0032761668&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032761668&partnerID=8YFLogxK

U2 - 10.1109/83.736696

DO - 10.1109/83.736696

M3 - Article

C2 - 18262869

AN - SCOPUS:0032761668

VL - 8

SP - 102

EP - 109

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 1

ER -