Flesh Tone Balance Algorithm for AWB of Facial Pictures

Tae-Wuk BAE; Sung-Hak LEE; Jung-Wook LEE; Kyu-Ik SOHNG

doi:10.1587/transele.E93.C.1616

Flesh Tone Balance Algorithm for AWB of Facial Pictures

Tae-Wuk BAE, Sung-Hak LEE, Jung-Wook LEE, Kyu-Ik SOHNG

Full Text Views

0

Cite this

Summary :

This paper proposes a new auto flesh tone balance algorithm for the picture that is taken for people. In this paper, the basis of auto white balance (AWB) is human face in photo. For experiment, the transfer characteristic of camera image sensor is analyzed and the camera output RGB is calculated by measuring the average face chromaticity under standard illuminant. For the face region taken under unknown illuminant, the proposed algorithm makes RGB output rate of face region become its rate of standard face color. For this, it adjusts the R and B channel and performs the chromaticity correction. Algorithm is applied to the light skin color (average face color) in Macbeth color chart and average color of various face colors that are actually measured.

Publication: IEICE TRANSACTIONS on Electronics Vol.E93-C No.11 pp.1616-1620

Publication Date: 2010/11/01

Publicized

Online ISSN: 1745-1353

DOI: 10.1587/transele.E93.C.1616

Type of Manuscript: BRIEF PAPER

Category

Cite this

Copy

Tae-Wuk BAE, Sung-Hak LEE, Jung-Wook LEE, Kyu-Ik SOHNG, "Flesh Tone Balance Algorithm for AWB of Facial Pictures" in IEICE TRANSACTIONS on Electronics, vol. E93-C, no. 11, pp. 1616-1620, November 2010, doi: 10.1587/transele.E93.C.1616.
Abstract: This paper proposes a new auto flesh tone balance algorithm for the picture that is taken for people. In this paper, the basis of auto white balance (AWB) is human face in photo. For experiment, the transfer characteristic of camera image sensor is analyzed and the camera output RGB is calculated by measuring the average face chromaticity under standard illuminant. For the face region taken under unknown illuminant, the proposed algorithm makes RGB output rate of face region become its rate of standard face color. For this, it adjusts the R and B channel and performs the chromaticity correction. Algorithm is applied to the light skin color (average face color) in Macbeth color chart and average color of various face colors that are actually measured.
URL: https://global.ieice.org/en_transactions/electronics/10.1587/transele.E93.C.1616/_p

Copy

@ARTICLE{e93-c_11_1616,
author={Tae-Wuk BAE, Sung-Hak LEE, Jung-Wook LEE, Kyu-Ik SOHNG, },
journal={IEICE TRANSACTIONS on Electronics},
title={Flesh Tone Balance Algorithm for AWB of Facial Pictures},
year={2010},
volume={E93-C},
number={11},
pages={1616-1620},
abstract={This paper proposes a new auto flesh tone balance algorithm for the picture that is taken for people. In this paper, the basis of auto white balance (AWB) is human face in photo. For experiment, the transfer characteristic of camera image sensor is analyzed and the camera output RGB is calculated by measuring the average face chromaticity under standard illuminant. For the face region taken under unknown illuminant, the proposed algorithm makes RGB output rate of face region become its rate of standard face color. For this, it adjusts the R and B channel and performs the chromaticity correction. Algorithm is applied to the light skin color (average face color) in Macbeth color chart and average color of various face colors that are actually measured.},
keywords={},
doi={10.1587/transele.E93.C.1616},
ISSN={1745-1353},
month={November},}

Copy

TY - JOUR
TI - Flesh Tone Balance Algorithm for AWB of Facial Pictures
T2 - IEICE TRANSACTIONS on Electronics
SP - 1616
EP - 1620
AU - Tae-Wuk BAE
AU - Sung-Hak LEE
AU - Jung-Wook LEE
AU - Kyu-Ik SOHNG
PY - 2010
DO - 10.1587/transele.E93.C.1616
JO - IEICE TRANSACTIONS on Electronics
SN - 1745-1353
VL - E93-C
IS - 11
JA - IEICE TRANSACTIONS on Electronics
Y1 - November 2010
AB - This paper proposes a new auto flesh tone balance algorithm for the picture that is taken for people. In this paper, the basis of auto white balance (AWB) is human face in photo. For experiment, the transfer characteristic of camera image sensor is analyzed and the camera output RGB is calculated by measuring the average face chromaticity under standard illuminant. For the face region taken under unknown illuminant, the proposed algorithm makes RGB output rate of face region become its rate of standard face color. For this, it adjusts the R and B channel and performs the chromaticity correction. Algorithm is applied to the light skin color (average face color) in Macbeth color chart and average color of various face colors that are actually measured.
ER -