Method and device of skin tone optimization in a color gamut mapping system
In a method of skin tone optimization in a color gamut mapping system, an image signal is first transformed from a predetermined color domain to an HSV color domain for generating an HSV image signal. Next, a skin tone optimization is performed on the HSV image signal for generating an adjusted saturation gain. Then, a color enhancement is performed on the HSV image signal according to the adjusted saturation gain and a color shift signal so as to generate a color enhancement signal. Finally, the color enhancement signal is transformed from the HSV color domain to the predetermined color domain.
1. A method of skin tone optimization in a color gamut mapping system, which is implemented in an image processor for performing the steps of:
(A) transforming an image signal from a predetermined color domain to an HSV (Hue, Saturation, Value) color domain for generating an HSV image signal;
(B) performing a skin tone optimization on the HSV image signal for generating an adjusted saturation gain, wherein step (B) comprises the steps of:
(B1) receiving the HSV image signal, color boundaries, saturation boundaries, and brightness boundaries to generate a hue smooth factor, a saturation smooth factor, and a brightness smooth factor;
(B2) generating a gain adjusting factor by performing a mean operation according to the hue smooth factor, the saturation smooth factor, and the brightness smooth factor; and
(B3) performing a weight adjustment on a saturation gain and a skin tone gain according to the gain adjusting factor;
(C) performing a color enhancement on the HSV image signal according to the adjusted saturation gain and a color shift signal so as to generate a color enhancement signal; and
(D) transforming the color enhancement signal from the HSV color domain to the predetermined color domain.
2. The method as claimed in claim 1 , wherein the hue smooth factor is expressed as:
factor_H
=
1
-
H
-
1
/
2
(
Hue_L
+
Hue_H
)
1
/
2
(
Hue_L
+
Hue_H
)
,
the saturation smooth factor is expressed as:
factor_S
=
1
-
S
-
1
/
2
(
S_L
+
S_H
)
1
/
2
(
S_L
+
S_H
)
,
and the brightness smooth factor is expressed as:
factor_V
=
1
-
V
-
1
/
2
(
V_L
+
V_H
)
1
/
2
(
V_L
+
V_H
)
,
where factor_H indicates the hue smooth factor; factor_S indicates the saturation smooth factor; factor_V indicates the brightness smooth factor; H, S, V indicate hue, saturation, brightness components of the HSV image signal; Hue_L, Hue_H indicate the color boundaries; S_L, S_H indicate the saturation boundaries; V_L, V_H indicate the brightness boundaries.
3. The method as claimed in claim 2 , wherein the gain adjusting factor is expressed as:
alpha=⅓(factor_ H +factor_ S +factor_ V ),
where alpha indicates the gain adjusting factor.
4. The method as claimed in claim 3 , wherein the adjusted saturation gain is expressed as:
S _gain′( H )=(1−alpha)× S _gain+alpha×Skin_gain,
where S_gain′(H) indicates the adjusted saturation gain, S_gain indicates the saturation gain, and skin_gain indicates the skin tone gain.
5. The method as claimed in claim 4 , wherein step (C) comprises the steps of:
(C1) receiving the HSV image signal, the adjusted saturation gain, and the color shift signal to perform an enhancement operation on a saturation component of the HSV image signal, so as to generate a saturation value of the color enhancement signal; and
(C2) performing a shift operation on a hue component of the HSV image signal for generating a hue value of the color enhancement signal.
6. The method as claimed in claim 5 , wherein the saturation value of the color enhancement signal is expressed as:
S′=S×S _gain′( H ),
where S′ indicates the saturation value, S_gain′(H) indicates the adjusted saturation gain, and S indicates the saturation component of the HSV image signal, and the hue value of the color enhancement signal is expressed as:
H′=H+H _shift( H ),
where H′ indicates the hue value, H_shift(H) indicates the color shift signal, and H indicates the hue component of the HSV image signal.
7. The method as claimed in claim 1 , wherein the predetermined color domain is an RGB (Red, Green, Blue) color domain.
8. The method as claimed in claim 7 , wherein transforming an image signal from a predetermined color domain to an HSV color domain in step (A) is expressed as:
H
1
=
cos
-
1
{
0.5
[
(
R
-
G
)
+
(
R
-
B
)
]
(
R
-
G
)
2
+
(
R
-
B
)
(
G
-
B
)
}
,
H
=
{
H
1
,
B
≤
G
360
°
-
H
1
,
B
>
G
,
S
=
Max
(
R
,
G
,
B
)
-
Min
(
R
,
G
,
B
)
Max
(
R
,
G
,
B
)
,
V
=
Max
(
R
,
G
,
B
)
255
,
where H1 indicates a temporary hue value, R indicates a red component of the image signal, G indicates a green component of the image signal, and B indicates a blue component of the image signal.