 |
Description  |
|
|
BACKGROUND OF THE INVENTION
The present invention relates to an image forming device and, more
particularly, to a digital full-color image forming device which is
capable of reading and storing original image data, making color
corrections on the stored image and producing a color reproduction from
the stored image.
The conventional digital full-color copying machine includes the
combination of a scanner and a printer, may which be substituted by a
printing machine and a monitor. The original image is scanned and a color
correction is performed whereby the image is processed with a matrix
having the color characteristic of --H reverse to the color characteristic
H of the scanner and the printer. The processed colored image is printed
out by the printer. The matrix of --H is selected in such a way that the
color data O (r, g, b) to be printed out may be the same as those of the
original.
The color correction most frequently utilizes a primary matrix and a
secondary matrix which are expressed as follows:
##EQU1##
The expression (1) is a primary color correcting matrix, where r, g and b
are color data to be corrected, K.sub.01, K.sub.02, K.sub.03, K.sub.11,
K.sub.12, K.sub.13, K.sub.21, K.sub.22, K.sub.23 --factors, Cr, Cg,
Cb--constants, r, g, b--color data after the color correction.
##EQU2##
The expression (2) is a primary color correcting matrix, where r, g and b
are color data to be corrected, K.sub.01, K.sub.02, K.sub.03, K.sub.04,
K.sub.05, K.sub.06, K.sub.07, K.sub.08, K.sub.09, K.sub.11, K.sub.12,
K.sub.13 . . . K.sub.29 --factors, Cr, Cg, Cb--constants, r, g, b--color
data after the color correction.
In the final adjustment of the full-color digital copying machines to be
shipped from the manufacturer, the color correction has been made by
adjusting the above-mentioned factors and constants to the values at which
a minimal color difference exists between an original and a copy. Such
values have usually been determined by multiple-regression analysis or a
like method.
Multiple-regression analysis is one of the methods which derive a
regression equation explaining dependent variables through the means of a
plurality of independent variables using the method of least squares. It
can be performed as follows:
To decrease the factors and constants of a secondary color correction for
r=K.sub.01 r+K.sub.02 g+K.sub.03 b+K.sub.04 r.sup.2 +K.sub.05 g.sup.2
+K.sub.06 b.sup.2 +K.sub.07 rg+K.sub.08 gb+K.sub.09 br+Cr,
let us minimize the following equation as to K.sub.01, K.sub.02, K.sub.03,
K.sub.04, K.sub.5, K.sub.06, K.sub.07, K.sub.08, K.sub.09 and Cr:
##EQU3##
Where Y(j) represents a variable in the order of J.
The minimization of the above-mentioned equation (3) means to minimize a
total of the differences between the right side and the left side of the
expression (2) for R determined as to every sample color Y(j), i.e., to
minimize an error.
The determination of the factors and the constants by the above-mentioned
multiple-regression analysis can be made in practice as follows:
A standard pattern including a variety of colors is first generated and
then printed out. A printed standard pattern is input, by the use of a
scanner, into the color correction unit which calculates the factors and
the constants in the equation (1) or (2) by a multiple-regression
analysis. The equation (2) or a higher degree equation may be considered
for application in principle, but the equation (1) is mostly used to
obtain printing speed and a circuit scale of the device. The determination
of the factors and constants may be done, not in real time, but the
practice requires real-time processing with the equation (1) or (2).
The original image is input by the scanner into the color correction unit
wherein the input image is processed with a matrix --H inverse to the
color characteristic H of the scanner and the printer. Thus, the processed
image is printed out. The color difference between the original and its
copy can be minimized by multiple-regression analysis and the equation (1)
or (2) to find the inverse matrix --H so that the color O (r, g, b), to be
printed out, may coincide with the original color.
Among a variety of colors, flesh-color, sward's-green and sky-blue are very
familiar to us in our daily life. Therefore, in color printing, it is very
important to reproduce finely these specific colors.
The following is described in a paper "Development of a color reproduction
theory in hard copying", Y. MIYAKE, Journal of Electrophotography, Vol.
29, No. 3, 1990. P284-292.
In color correcting, the flesh-color area of an original copy is first
extracted and the area of a face is then extracted therefrom through a
process such as removing separate points, expanding, contracting, labeling
and recognizing the shape. The extracted face area is corrected with a
color correcting factor which assures a fine reproduction of the flesh
color and the rest is corrected by a normal color correcting factor. A
correcting matrix for flesh-color includes a variety of flesh-color
patterns processed by multiple regression analysis in order to minimize
the difference in the color between an original and a print's sample.
Therefore, it can assure a fine correction of the flesh-color. However,
this matrix cannot be applicable for correcting any color other than
flesh-color. Accordingly, color correction is effectively done by applying
the specially prepared matrix to the flesh-color area only and by applying
a standard matrix to the other normal colored areas.
Original image data is input by a scanner into a color correction unit
which extracts each color area of the original image from the data input
therein and separately selects a suitable color correction matrix for each
extracted area, e.g., a flesh-color correcting matrix for a flesh-color
area and a sky-blue correcting matrix for a sky-blue area. This makes it
possible to finely correct specific colors requiring precise reproduction
for the human eye by selectively using corresponding matrices especially
prepared for them and to limit the correction effect to only an extracted
area with no effect on the rest of the areas.
Three color correcting blocks A, B and C perform color correction,
respectively, with a flesh-color correcting matrix, a sky-blue color
correcting matrix and a normal color correcting matrix made from a wide
range of color samples. The flesh colored area and the sky blue area are
separately extracted and corrected by using the color correcting matrices
A (for the flesh-color area) and B (for the sky-blue color area)
respectively. In the case of an example of a color image original, image
data of the upper sky-blue area is corrected by the use of the color
correcting matrix B, image data of the flesh-color areas of a face and
hands is corrected by the use of the color correcting matrix A and image
data of the remaining normal colored areas is corrected by the use of the
color correcting matrix C.
A conventional type digital, full-color copying machine comprises an
original photographic image, an image sensor comprising a charge coupled
device (CCD), an A-D converter, a shading correction, a color converter, a
portion for UCR (Under Color Removal) and BP (Black Paint) memory and
laser unit.
The CCD image sensor reads by scanning the original image and transfers the
image data "r", "g" and "b" (sampling) to the A-D converter which in turn
converts the analog image data into digital signals R, G, B (quantized).
These digital signals, which include an error induced by variations of the
CCD elements and the uneven luminosity of a lamp, are corrected in the
shading correction portion and then transferred to the color converter
which in turn converts the digital image signals into those having a gray
level suitable for the visual properties of the human eye by logarithmic
conversion and converting three primary colors (R, G, B) of light into
three primary colors (Y, M, C) of the toner. The UCR and BP portions
perform under color removal from the converted data and by black painting
thereon and then enter the processed data into the memory. When copying is
required, the digital image data is subsequently read out from the memory
and transferred to the laser unit which outputs a full-color image.
The Japanese publications of unexamined applications JP,A, 60-91770 and
JP,A, 4-323957 describe, respectively, a color image processing device and
a color image processing method, which are the prior art devices and
methods which the present invention is concerned with. These prior art
references involve extracting an area of a specific color from an original
specimen to include even a small amount of a specific color, e.g., flesh
color and conducting special processing such as fuzzy optimal processing.
This may prolong the process and/or increase its performance cost in view
of the fact that most of the originals have a small area of the specific
color and may not be effectively processed for an increased time of the
special processing. On the contrary, the present invention proposes to
determine an area of a specific color of an original image and to conduct
a masking process on the original color data only when the determined area
of the specific color is large or to conduct normal masking of the
original image if the determined area is small, thereby making it possible
to quickly correct most of the normal color originals and to highly
correct any original image that is rich with the specific color to get an
image of high quality. This may improve the performance cost of the color
image processing device.
FIG. 1 shows the principle scheme of a conventional digital full-color
copying machine. The shown case includes a combination of a scanner and a
printer, which, however, may be substituted by a combination of a printing
machine and a monitor. An original image is scanned by the scanner to put
in a color correction (step 1) wherein the image is processed with a
matrix having a color characteristic of --H reversed to a color
characteristic H of the scanner and the printer (step 2). The processed
color image is printed out by the printer (step 3). The matrix of --H is
selected in such a way that the color data O (r, g, b) to be printed out
may be the same as those of the original.
As shown in FIG. 1, an original image is input by the scanner into the
color correction unit wherein the input image is processed with a matrix
--H, inverse to the color characteristic H, of the scanner and the
printer. Thus the processed image is printed out. A color difference
between the original and its copy can be minimized by multiple-regression
analysis and the equation (1) or (2) to find the inverse matrix --H for
making that the color O (r, g, b) to be printed out to coincide with the
original color.
FIG. 2 is a view showing the system of a conventional digital, full-color
copying machine.
Original image data is input by a scanner into a color correcting matrix
(step 1) which extracts every color area of the original image from the
data input therein (step 2) and separately selects a suitable color
correcting matrix for each extracted area (step 3), and the color image is
printed out by the printer (step 4), e.g., a flesh-color correcting matrix
for a flesh-color area and a sky-blue correcting matrix for a sky-blue
area. This makes it possible to finely correct specific colors requiring
the precise reproduction for the human eye by selectively using
corresponding matrices especially prepared for them and to limit the
correction effect to only an extracted area with no effect on the rest of
the areas.
In FIG. 2, there are shown three color correcting blocks, A, B and C which
perform color corrections, respectively, with a flesh-color correcting
matrix, a sky blue-color correcting matrix and a normal color correcting
matrix made from a wide range of color samples.
The flesh-color and sky-blue color areas are separately extracted and
corrected by using the color correcting matrices A (for the flesh-color
area) and B (for the sky-blue color area) respectively. In the case of an
example of a color image original shown in FIG. 3, image data of the upper
sky-blue area is corrected by the use of the color correcting matrix B,
image data of the flesh-color areas of a face and hands are corrected by
the use of the color correcting matrix A and image data of the rest of the
normal colored areas are corrected by the use of the color correcting
matrix c (see FIG. 2).
FIG. 4 is a construction block-diagram of a conventional type digital
full-color copying machine, wherein numeral 1 designates a photographic
(image) original, and which comprises an image sensor 2 comprising a
charge coupled device (CCD), an A-D converter 3, a shading correction 4, a
color converter 5, a portion for UCR (Under Color Removal) and BP (Black
Paint) 6, a memory 7 and a laser unit 8.
The CCD image sensor 2 reads by scanning the original image 1 (step a) and
transfers the image data "r", "g" and "b" (sampling) to the A-D converter
3 (step b) which in turn converts the analog image data into digital
signals R, G, B (quantized). These digital signals, which include an error
induced by variations of CCD elements and the uneven luminosity of a lamp,
are corrected in the shading correction unit 4 and then transferred to the
color converter 5 (step c) which in turn converts the digital image
signals into those having a gray level suitable for the visual property of
the human eye by use of logarithmic conversion and converts three primary
colors (R, G, B) of light into three primary colors (Y, M, C) of toner.
The UCR and BP portion 6 performs under-color removal from the converted
data and black painting thereon and then enters the processed data into
the memory 7. When copying is required, the digital image data is
subsequently read out from the memory 7 (step d) and transferred to the
laser unit 8 which outputs a full-color image.
As mentioned above, the conventional image forming device employs an
advanced system that extracts a flesh-color area and a sky-blue area and
separately corrects the extracted areas by using specially selected
correction values in a correction table in order to match them with the
visual properties of the human eye. This device, however, has a drawback
that, if the device is used for color correction in a full-color digital
copying system with a full-color digital scanner, it requires much time
for processing and reduces the speed of color correction. An attempt to
increase the processing speed of the device has led to its increased size
with an increase in cost to manufacture because the color area extraction
contains many operations such as removing separate points, expanding and
contracting, labeling and so on. The problem is that any area of specific
color, e.g., flesh-color must be extracted even if the original has only a
small area of said color. In other words, the device always performs the
extraction of the specified color areas on any kind of originals, e.g., an
original mostly containing characters and having no need of the
above-mentioned operations. In this case, the time is consumed with no
effect on printing quality. The separate correction of an extracted area
of any specified color, e.g., flesh-color, by using a factor selected for
masking the color is effective to improve the color reproduction's quality
but it makes the device expensive and time-consuming. This is the reason
the device has not heretofore been employed in practical applications.
In a digital full-color copying machine, a full-color image copy to be
output has several specified colors, e.g., human flesh-color and sky-blue,
which must be faithfully reproduced. The specific colors are familiar to
us in our daily life and, therefore, are easily recognized by the human
eye. However, the conventional digital full color copier performs
one-patterned color correction and offers poor reproduction of the
specific colors. It cannot provide a full-color image with finely
reproduced flesh-color and sky-blue color therein. This may totally effect
the quality of the copy products of a the digital copying machine.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a color correcting
device for use in a full-color copying system with color masking, which is
capable of determining areas of specific colors (e.g. flesh-color,
sky-blue) determined during the time of pre-scanning an original image;
selectively uses a plurality of correction matrices for correcting colors
(flesh-color, sky-blue and normal colors) at the time of color masking,
depending on whether the specific color area is larger than the specific
value or not; dynamically switches the color correcting matrices according
to pixels' colors; characterized by excellent color reproduction of the
original that can be obtained and capable of high speed printing of the
original having small amounts of the specific color.
It is another object of the present invention to provide the possibility of
correcting image data of specific colors (easily recognizable by human eye
but not easily reproducible by the full-color copying machine) by using
respective correction tables prepared on the basis of data obtained by
measuring a number of samples; to correct image data other than the
specified colors by using a normal correction table (to make the color be
similar to the original one); to extract only the specific color when the
user designates it (the conventional method employs the time-consuming
automatic extraction of all colors); to effectively perform the color
correction in a shorter time.
It is another object of the present invention to obtain an output image
that satisfies the user's requirements by providing the possibility of
freely selecting the necessity to correct the image's color in order to
remove the drawback of the conventional method that provides only
automated color correction, not allowing the user to select a desired
color (some users desire the correct reproduction of the original color,
as for example, in the case where an automatically corrected image of a
human figure represents somewhat colored skin while the original
represents white skin).
The present invention has as its primary object to provide a color image
forming device for use in a digital full-color copying system, which is
capable of more precisely correcting such colors, e.g., flesh-color,
garden-green, sky-blue, all of which are familiar to us in daily life and
therefore require high quality color reproduction using the actual colors
of the original, and which is also capable of high-speed color correction
for most of the usual originals, e.g., of character images, which do not
require high-quality color reproduction but only economical high speed
printing. According to the present invention, it is possible to get a
print sample of high quality color reproduction from an original image
being rich with specific colors, e.g. flesh-color, garden-green, sky-blue
and so on, by determining an area of each specific color at the time of
pre-scanning the original and by separately extracting and correcting each
specific colored area if the area is large, and it is also possible to
quickly get a printing sample from the original image including a small
amount of any specific color. The device according to the present
invention can automatically detect, at the time of pre-scanning, whether
an original image is requiring the separate correction of any specific
color or not. There is no need to worry about the decision for separate
color correction. Pre-scanning is conducted to detect the size, position
and monochrome or color state of an original without determining any
specific color area, i.e., with no loss of pre-scanning speed. It is also
possible to determine an area of specific color on an original at a
reduced speed (cost) by thinning the image data.
The present invention proposes to prepare a plurality of specific color
correcting tables (for flesh-color, sky-blue and so on), each containing
image data obtained from measurements of image samples of only a specific
color (e.g. flesh-color or sky-blue); selectively extracting any desired
area of any specific color on a full-color original image; and of
performing color correction with color masking of the extracted area
according to the designated color correction table and of not-extracted
areas according to a table for normal color correction. This feature is
effective to provide a full-color copy image of a desired color quality
and to improve the image quality of a digital full-color copying machine.
Looking at a monitor screen, one can carry out the above-mentioned color
correcting operations. This makes it possible to reduce the circuity of
the device in comparison with the circuity of a device which performs
automatic area separation. Accordingly, the device may be manufactured at
lower cost and may work at higher speed (without extracting a specific
color area, e.g., from an image of a person wearing a flesh-colored or
sky-blue suit which would be extracted when automatic area separation is
performed).
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 is a view showing a system of a conventional digital color copier.
FIG. 2 is a view showing another system of a conventional digital color
copier.
FIG. 3 is a view showing a conventional color correction.
FIG. 4 is a construction view of a conventional digital color copier.
FIG. 5 is a block diagram for explaining an example of a circuit for
extracting a specific color for use in an image forming device according
to the present invention.
FIG. 6 shows an example of a circuit of a comparator shown in FIG. 5.
FIG. 7 shows an example of a circuit for determining an area of a specific
color, according to the present invention.
FIG. 8 shows a color correcting circuit, according to the present
invention.
FIG. 9 shows another example of a color correcting circuit, according to
the present invention.
FIG. 10 is a flow chart for explaining the operations of an image forming
device, according to the present invention.
FIGS. 11 and 12 are a block diagram for explaining the structure of a
full-color digital copying machine utilizing a color correcting function,
according to the present invention.
FIG. 13 shows another embodiment of an image forming device according to
the present invention.
FIG. 14 is a construction view of a control portion for the color
correction of an image's data.
FIGS. 15A, 15B, 15C, 15D, 15E, 15F are a view for explaining how to
designate a specific color according to the present invention.
PREFERRED EMBODIMENT OF THE INVENTION
FIG. 1 shows the principle scheme of a conventional digital full-color
copying machine. The shown case includes a combination of a scanner and a
printer, which, however, may be substituted by a combination of a printing
machine and a monitor. An original image is scanned by the scanner to put
in a color correction (step 1) wherein the image is processed with a
matrix having a color characteristic of --H reversed to a color
characteristic H of the scanner and the printer (step 2). The processed
color image is printed out by the printer (step 3). The matrix of --H is
selected in such a way that the color data O (r, g, b) to be printed out
may be the same as those of the original.
As shown in FIG. 1, an original image is input by the scanner into the
color correction wherein the input image is processed with a matrix --H,
inverse to the color characteristic H, of the scanner and the printer.
Thus the processed image is printed out. A color difference between the
original and its copy can be minimized by multiple-regression analysis and
the equation (1) or (2) to find the inverse matrix --H for making that the
color O (r, g, b) to be printed out to coincide with the original color.
FIG. 2 is a view showing the system of a conventional digital, full-color
copying machine.
Original image data is input by a scanner into a color correcting matrix
(step 1) which extracts every color area of the original image from the
data putt therein (step 2) and separately selects a suitable color
correcting matrix for each extracted are (step 3), and the color image is
printed out by the printer (step 4), e.g., a flesh-color correcting matrix
for a flesh-color area and a sky-blue correcting matrix for a sky-blue
area. This makes it possible to finely correct specific colors requiring
the precise reproduction for the human eye by selectively using
corresponding matrices especially prepared for them and to limit the
correction effect to only an extracted area with no effect on the rest of
the areas.
In FIG. 2, there are shown three color correcting blocks, A, B and C which
perform color corrections, respectively, with a flesh-color correcting
matrix, a sky blue-color correcting matrix and a normal color correcting
matrix made from a wide range of color samples.
The flesh-color and sky-blue color areas are separately extracted and
corrected by using the color correcting matrices A (for the flesh-color
area) and B (for the sky-blue color area) respectively. In the case of an
example of a color image original shown in FIG. 3, image data of the upper
sky-blue area is corrected by the use of the color correcting matrix B,
image data of the flesh-color areas of a face and hands are corrected by
the use of the color correcting matrix A and image data of the rest of the
normal colored areas are corrected by the use of the color correcting
matrix C (see FIG. 2).
FIG. 4 is a construction block-diagram of a conventional type digital
full-color copying machine, wherein numeral 1 designates a photographic
(image) original, and which comprises an image sensor 2 comprising a
charge coupled device (CCD), an A-D converter 3, a shading correction 4, a
color converter 5, a portion for UCR (Under Color Removal) and BP (Black
Paint) 6, a memory 7 and a laser unit 8.
The CCD image sensor 2 reads by scanning the original image 1 (step a) and
transfers the image data "r", "g" and "b" (sampling) to the A-D converter
3 (step b) which in turn converts the analog image data into digital
signals R, G, B (quantized). These digital signals, which include an error
induced by variations of CCD elements and the uneven luminosity of a lamp,
are corrected in the shading correction 4 and then transferred to the
color converter 5 (step c) which in turn converts the digital image
signals into those having a gray level suitable for the visual property of
the human eye by use of logarithmic conversion and converts three primary
colors (R, G, B) of light into three primary colors (Y, M, C) of toner.
The UCR and BP portion 6 performs under-color removal from the converted
data and black painting thereon and then enters the process data into the
memory 7. When copying is required, the digital image data is subsequently
read out from the memory 7 (step d) and transferred to the laser unit 8
which outputs a full-color image.
Referring now to the accompanying drawings, preferred embodiments of the
present invention will be described in detail as follows:
FIG. 5 is a construction view for explaining an example of a circuit for
extracting a specific color area on an original image in an image forming
device embodied in the present invention. The circuit comprises a register
11a for extracting flesh-color (r), a register 11b for extracting
flesh-color (g), a register 11c for extracting flesh-color (b),
comparators 12a, 12b, 12c and an AND-circuit 13.
The circuit first performs pre-scanning of an original image to extract a
specific color, e.g., flesh-color therefrom. The specific color extraction
can be made by a simple circuit which, by way of example, comprises three
flesh-color extracting registers r, g and b which preset thereat
extractable primary color components, red, green and blue, and of
flesh-color respectively. The comparators 12a, 12b and 12c compare
components R, G and B of the image data, respectively, with the registers
r, g and b. They have an output "1" only if their inputs are r-C.sub.0
<R<r+C.sub.1, g-C.sub.2 <G<g+C.sub.3 and b-C.sub.4 <B<b+C.sub.5
respectively. The AND-circuit 13 with three inputs has an output "1" only
if its inputs are r-C.sub.0 <R<r+C.sub.1, g-C.sub.2 <G<g+C.sub.3 and
b-C.sub.4 <B<b+C.sub.5 simultaneously. C.sub.0 -C.sub.5 are constants
which are defined depending upon a color range of color data used for
determining matrix factors (coefficients) used for correcting the
flesh-color. This means that the image data with flesh-color extracted by
a scanner can be corrected with the use of matrix factors for flesh-color
correction of a finely reproduced color.
FIG. 6 shows an example of a circuit of the comparator shown in FIG. 5. The
comparator includes EXNOR-circuits (A,B,C,D) 14a,14b,14c,14d and
AND-circuit 15.
The comparator can also be formed of simple circuity. For example, if
r-C.sub.0 <R<r+C.sub.1 is 38H with C0=9H and C1=8H, there will be
2FH<R<40H that means the necessity of extracting pixels having R-values
from 30H to 3FH can easily be achieved by using 4 EXNOR 14a,14b,14c,14d
and the AND-circuit 15 with four inputs. When an image data has 8 bits,
the most significant bit R.sub.7 of its components R of the image data and
the most significant bit r.sub.7 of the flesh-color extracting register r
are input into the EXNOR-circuit (D) 14d whose output is "1" only if
R.sub.7 =r.sub.7. The 6th bit, 5th bit and 4th bit of the image data and
those of the registers are put into the respective EXNOR-circuits
(C),(B),(A) which output is "1" only if the components R of the image data
are equal to the respective values preset at the registers r.
Outputs of the EXNOR-circuits 14a,14b,14c,14d are input into the
AND-circuit (E) 15 whose output is "1" only if all inputs are "1", i.e., 4
sets of input bits are equal to each other. In the shown case, if r.sub.4
=r.sub.5 =1 and r.sub.6 =r.sub.7 =0 are preset and 4 left significant bits
is 3H, the AND-circuit (E) 15 with four inputs has an output "1". The
embodiment is realized since the output of the comparator is "1" only in
the case of 2FH<R<40H. Accordingly, when three comparators for components
R, G, B prepared for fresh-color are connected at their outputs to an
AND-circuit with three inputs as shown in FIG. 5, the AND-circuit has an
output "1" (color extraction signal) only in the case of the image data
relating to "flesh-color".
FIG. 7 shows an example of a circuit for determining an area of a specific
color according to the present invention, which includes AND-circuits
16a,16b,16c, EXOR-circuits 17a,17b,17c,17d and flip-flops 18a,18b,18c,18d.
An area of specified color on an original image, which is extracted by the
specified color extracting circuit shown in FIG. 5, can be determined by
counting pixels (pictorial elements) included therein. This circuit is
simple and easily formed.
For example, a 4 bit counter is composed of 4 flip-flops 18a,18b,18c,18d, 4
EXOR-circuits 17a,17b,17c,17d, 2 AND-circuits 16a, 16b with two inputs and
1 AND-circuit 16c with three inputs. The above-mentioned color-extraction
signal is input into the EXOR-circuit (EXOR.sub.0) 17a which at its other
input receives an output signal q.sub.0 from the flip-flop F/F0. If one
input is "1", the EXOR-circuit inverts the level of its output for the
level of the other input. This output is applied to a data input of the
Flip-flop F/F0 which, therefore, inverts the level of its output in
synchronism with a data transfer clock only when the color extraction
signal is "1". As a usual binary counter, output signals q.sub.0 and
q.sub.1 of the flip-flops F/F0 and F/F1 are applied to the inputs of the
EXOR-circuit (EXOR.sub.1) 17b, but, unlike the usual binary counter, the
output of the AND-circuit (AND.sub.0) 16a is "1" only if the color
extraction signal and the signal q.sub.0 are all "1". In this case, the
output of the flip-flop F/F1 is inverted in synchronism with the data
transfer clock. The flip-flops F/F2 and F/F3 convert their data-outputs
when a lower-order bit is "1" and the color extraction signal is "1".
Accordingly, counting occurs only when the color extraction signal is "1".
It is also possible to reduce the number of bits for the counter and to
give a margin of counting speed by thinning pulses of data transferring
clock relative to the practical image data. The area of any specific color
can be easily counted by the above-described circuit.
FIGS. 8 and 9 show examples of color correcting circuits according to the
present invention. There are shown look-up tables (LUT) 19a,19b,19c, an
adder 20, registers of color correction factor | | |
