Image analysis – Color image processing
Reexamination Certificate
1999-01-29
2001-10-30
Johnson, Timothy M. (Department: 2623)
Image analysis
Color image processing
C345S604000
Reexamination Certificate
active
06310969
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a color space conversion method for an image process, and more particularly to a color conversion from the HSV color space to a new color space.
2. Discussion of Related Art
Currently, extensive research for searching an image based upon the contents is actively being conducted. In response, commercial image searching apparatus and/or applications program are being developed to meet the demand for an improved content based image searching.
The most important information for use in the content based image searching is the color information. Accordingly, an effective performance of an image searching apparatus or applications program depends greatly on an accurate method for extracting the color information.
Generally, the number of different colors which can theoretically be expressed by a computer has been steadily increasing. However, the number of colors which can be displayed by the computer is limited by the available number of quantized colors. A problem arises because the number of quantized colors is less than the number of the colors which can be expressed. Quantization is a procedure of color mapping from M number of colors to N number of colors when M>>N, where M and N are positive integral numbers.
In computers, a color is expressed utilizing the RGB color model based upon the three primary colors of red R, green G, and blue B. However, the RGB space is hardware oriented and a limitation exists in expressing the color changes such that the change can be sensed by the human eye. Thus, the RGB space is often converted into a user oriented HSV color model based upon a hue H, saturation S and value V.
FIGS. 1A and 1B
illustrate an existing method for converting the RGB color space into the HSV color space in the related art. The conversion of the RGB color space into the HSV color space is completed by the following; where the values r, g, b, v and s ranges from 0 and 1, and the value h ranges from 0 to 360. Also, if max=min, h is undefined for achromatic color.
max
=
MAXIMUM
{
r
,
g
,
b
}
min
=
MINIMUM
{
r
,
g
,
b
}
v
=
max
s
=
(
max
-
min
)
/
max
h
=
(
g
-
b
)
/
(
max
-
min
)
*
60
,
if
(
r
=
max
⋂
(
g
-
b
)
>
0
)
(
g
-
b
)
/
(
max
-
min
)
*
60
+
360
,
if
(
r
=
max
⋂
(
g
-
b
)
<
0
)
(
2.0
+
(
b
-
r
)
)
/
(
max
-
min
)
*
60
,
if
(
g
=
max
)
(
4.0
+
(
r
-
g
)
)
/
(
max
-
min
)
*
60
,
if
(
b
=
max
)
&AutoRightMatch;
Upon conversion from the RGB color space into the HSV color space, a cylindrical shape results as shown in FIG.
1
B. On the cylindrical shape of HSV color space, a gray color is displayed towards the center of circle around the V axis, a solid color is displayed towards the circumference of the circle around the V axis, a brighter (white) color is displayed towards the directions of the axes S and V(+), a darker (black) color in the direction of the axis V(−), and the hue (H) is defined as the angle around the V axis.
However, even within the same HSV space, a range of color variation which can be visually recognized are differently displayed. Also, a simple quantization of the colors in the HSV space does not result in a uniform expression of colors. On the other hand, to quantize the saturation vector value in consideration of the color change sensed by the human eye, many complicated calculations are required. Moreover, selecting a quantization model would be difficulty. Thus, the performance of the content based image searching system using the HSV space deteriorates.
For example, referring to
FIG. 1B
, the black color is displayed towards the direction of the bottom surface of the cylindrical shape. However, if a color is distributed in the direction of the S axis, it would be difficult to visually differentiate the black color from the distributed color. Thus, even if a color quantization is possible, the color differentiation at low values of V cannot made. Accordingly, a problem still remains.
Therefore, a new color space and a color quantizing method is required in which all of the colors can be distributed in an even manner and the variation of the colors can be uniformly and visually recognized by a user.
FIG. 2
is a schematic view illustrating an HSV color coordinate space structure in the prior art.
Referring to
FIG. 2
, on the cylindrical column of HSV color coordinate space, the value V is the center axis of the cylindrical column, the saturation S is a normal vector from the center axis of the column to concentric circle, and the hue H is the angle around V. To divide a color area and a gray area on the HSV color space, the gray area is designated according to the saturation S. Particularly, a gray value &ggr; is defined as a constant value &tgr;
gray
=&ggr;, such that the gray area if S<&tgr;
gray
and the color area if S>&tgr;
gray
are divided.
However, taking into consideration the user's visual and sensitive characteristics, the variation of the gray area may be generated by the gray value &ggr; depending also upon the hue H or the value V. Nevertheless, in dividing the color area and the gray area of the conventional HSV color space, the gray area is designated according with only the saturation S of the HSV or the HSI, irrespective of the hue H or the value V. As a result, the color area visually recognizable is really displayed as the gray area.
In such case, unnecessary color signal processing has been added, increasing the loads of the color signal processing. In addition, if the color information to which the color signal process is necessary is in the gray area, the performance of an image search system becomes degraded.
SUMMARY OF THE INVENTION
Accordingly, an object of the present invention is to solve at least the problems and disadvantages of the related art.
An object of the present invention is to provide a new color space upon conversion of RGB color coordinate space into HSV color coordinate space.
Another object of the present invention is provide a color space to evenly distribute all of the color areas while uniformly displaying the color variation in consideration of a user's visual characteristic.
A further object of the present invention is to provide a color quantizing method to improve a content based image search.
Still a further object of the present invention is to provide a variable gray area designating method in which a gray value is variable according to hue H and value V.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objects and advantages of the invention may be realized and attained as particularly pointed out in the appended claims.
To achieve the objects and in accordance with the purposes of the invention, as embodied and broadly described herein, a color space according to the present invention is characterized by a reverse conical shape in a three-dimensional space and has color space coordinates comprising the value V in the direction of a longitudinal axis passing the center of the reverse cone, the hue H as the angle 0°-360° around the longitudinal axis passing the center of the reverse cone, and a difference value S′ between a maximum value and a minimum value on the shortest linear component in the direction of maximum circumference from the center of the circle.
According to another aspect of the present invention, a color quantizing method includes the steps of converting input RGB color information into the coordinate values on a three-dimensional reverse conical shape of color coordinate space comprised of the coordinates of hue H, a difference value S′ and value V, based upon the bigness and/or smallness and th
Kim Hyeon Jun
Lee Jin Soo
Fleshner & Kim LLP
Johnson Timothy M.
LG Electronics Inc.
LandOfFree
Color coordinate space structure and color quantizing and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Color coordinate space structure and color quantizing and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Color coordinate space structure and color quantizing and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2551651