U.S. patent application number 09/136929 was filed with the patent office on 2001-08-23 for image processing apparatus for discriminating image field of original document plural times and method therefor.
Invention is credited to KAWAKAMI, HARUKO, RAO, GURURAJ, YAMAMOTO, NAOFUMI.
Application Number | 20010016065 09/136929 |
Document ID | / |
Family ID | 16801861 |
Filed Date | 2001-08-23 |
United States Patent
Application |
20010016065 |
Kind Code |
A1 |
YAMAMOTO, NAOFUMI ; et
al. |
August 23, 2001 |
IMAGE PROCESSING APPARATUS FOR DISCRIMINATING IMAGE FIELD OF
ORIGINAL DOCUMENT PLURAL TIMES AND METHOD THEREFOR
Abstract
An image processing apparatus incorporating a field separating
section for separating an original image into plural types of
fields in response to a first image signal obtained at a rough
density of the supplied original image, a characteristic value
calculating section for calculating a characteristic value of the
original image in response to a second image signal of the original
image obtained at a density which is higher than the rough density,
and a discrimination section for discriminating an image field of
the original image in accordance with the characteristic value to
correspond to the type of the field.
Inventors: |
YAMAMOTO, NAOFUMI; (TOKYO,
JP) ; KAWAKAMI, HARUKO; (YOKOHAMA-SHI, JP) ;
RAO, GURURAJ; (YOKOHAMA-SHI, JP) |
Correspondence
Address: |
FOLEY & LARDNER
3000 K STREET NW
SUITE 500
WASHINGTON
DC
200075109
|
Family ID: |
16801861 |
Appl. No.: |
09/136929 |
Filed: |
August 20, 1998 |
Current U.S.
Class: |
382/173 |
Current CPC
Class: |
G06V 10/28 20220101;
G06T 7/11 20170101; G06V 30/10 20220101; G06V 30/414 20220101; G06V
30/162 20220101; G06V 30/18105 20220101; G06T 2207/10008 20130101;
G06V 10/56 20220101; G06T 2207/30176 20130101 |
Class at
Publication: |
382/173 |
International
Class: |
G06K 009/34 |
Foreign Application Data
Date |
Code |
Application Number |
Aug 20, 1997 |
JP |
9-223673 |
Claims
1. An image processing apparatus comprising: field separating means
for separating an original image into plural types of fields in
response to a first image signal obtained at a first density of the
original image; characteristic value calculating means for
calculating a characteristic value of the original image in
response to a second image signal of the original image obtained at
a second density which is higher than the first density;
discrimination means for discriminating an image field of the
original image in accordance with the characteristic value
calculated by the characteristic value calculating means to
correspond to the type of the field by the field separated by the
separating means; and an image processing means for processing a
predetermined image processes corresponding to a result of
discrimination of the image field performed by the discrimination
means, on the second image signal.
2. An image processing apparatus according to claim 1, wherein each
of the field separation means and the characteristic value
calculating means further includes scanning means for scanning the
original image at the first density and then scanning the same at
the second density.
3. An image processing apparatus according to claim 1, wherein the
field separating means includes means for separating the original
image into at least a usual character field, characters on a
background, a continuous gradation field and a dot gradation
field.
4. An image processing apparatus according to claim 1, wherein
means is provided which deletes a gradation image existing in a
field which has not been discriminated by the field separating
means as a gradation field and a gradation image existing in a
field discriminated as a continuous gradation field.
5. An image processing apparatus according to claim 1, wherein the
characteristic value calculating means includes characteristic
value calculating means for calculating change in the density, an
average density and chroma which are characteristic values of the
original image in response to the second image signal of the
original image obtained at the second density which is higher than
the first density.
6. An image processing apparatus according to claim 1, wherein the
characteristic value calculating means includes characteristic
value calculating means for calculating a characteristic value of
the original image in each of unit regions having different lengths
in a main scanning direction and a sub-scanning direction in
response to the second image signal of the original image obtained
at the second density which is higher than the first density.
7. An image processing apparatus according to claim 1, wherein the
discrimination means includes second discrimination means for
discriminating the image field of the original image by means of
comparing the characteristic value calculated by the characteristic
value calculating means with a threshold value corresponding to the
type of the field separated by the separating means.
8. An image processing apparatus according to claim 1, wherein the
characteristic value calculating means includes a second
characteristic value calculating means for calculating a segment
component of the original image in a main scanning direction and a
sub-scanning direction as the characteristic value in response to
the second image signal of the original image obtained at the
second density which is higher than the first density.
9. An image processing method comprising the steps of: a field
separating step for separating an original image into plural types
of fields in response to a first image signal obtained at a first
density of the supplied original image; a characteristic value
calculating step for calculating a characteristic value of the
original image in response to a second image signal of the original
image obtained at a second density which is higher than the first
density; a discrimination step for discriminating an image field of
the original image in accordance with the characteristic value
calculated in the characteristic value calculating step to
correspond to the type of the field separated in the field
separating step; and an image processing step for processing a
predetermined image processes corresponding to a result of
discrimination of the image field performed by the discrimination
means, on the second image signal.
10. An image processing method according to claim 9, wherein each
of the field separation step and the characteristic value
calculating step further includes a scanning step for scanning the
original image at the first density and then scanning the same at
the second density.
11. An image processing method according to claim 9, wherein the
field separating step includes a step for separating the original
image into at least a usual character field, characters on a
background, a continuous gradation field and a dot gradation
field.
12. An image processing method according to claim 9, wherein a step
is provided which deletes a gradation image existing in a field
which has not been discriminated in the field separating step as a
gradation field and a gradation image existing in a field
discriminated as a continuous gradation field.
13. An image processing method according to claim 9, wherein the
characteristic value calculating steps includes a characteristic
value calculating step for calculating change in the density, an
average density and chroma which are characteristic values of the
original image in response to the second image signal of the
original image obtained at the second density which is higher than
the first density.
14. An image processing method according to claim 9, wherein the
characteristic value calculating step includes a characteristic
value calculating step for calculating a characteristic value of
the original image in each of unit regions having different lengths
in a main scanning direction and a sub-scanning direction in
response to the second image signal of the original image obtained
at the second density which is higher than the first density.
15. An image processing method according to claim 9, wherein the
discrimination step includes a second discrimination step for
discriminating the image field of the original image by means of
comparing the characteristic value calculated by the characteristic
value calculating step with a threshold value corresponding to the
type of the field separated by the separating step.
16. An image processing method according to claim 9, wherein the
characteristic value calculating step includes a second
characteristic value calculating step for calculating a segment
component of the original image in a main scanning direction and a
sub-scanning direction as the characteristic value in response to
the second image signal of the original image obtained at the
second density which is higher than the first density.
Description
BACKGROUND OF THE INVENTION
[0001] The present invention relates to a method of discriminating
the attribute of an image, an image processing apparatus adopting
the discrimination method and an image forming apparatus.
[0002] Recently, when an image incorporating characters and
gradation images mixed therein is treated, recording for obtaining
a hard copy is an issue that needs to be resolved. An
electrophotographic method is in common use as a method of
recording a digital image. The foregoing method has unsatisfactory
performance of expressing a density of two to several levels per
picture point to be recorded. Therefore, when a gradation image is
expressed, a pulse width modulation method or the like must
employed. The pulse width modulation method is broadly divided into
a one-pixel modulation method and a two-pixel modulation method in
terms of the period of pulses in the pulse width modulation.
Although the former method is able to clearly record characters
because of excellent resolution, the foregoing method is of
inferior gradation expressing characteristic. On the other hand,
the latter method excels in gradation expressing characteristic
which enables a photograph or the like to smoothly be reproduced.
However, the latter method is of inferior resolution. Although the
modulation methods are able to realize either of the resolution and
the gradation expressing characteristic, both of the foregoing
requirements cannot simultaneously be realized when a recording
operation is performed.
[0003] To record an image such that both of the resolution and the
gradation expressing characteristic are realized, an image field
discrimination process is performed. That is, an image which must
be recorded is discriminated into fields, for example, photographs,
in which the gradation expressing characteristic is of importance
and fields, for example, characters or line drawings, in which the
resolution is of importance. In accordance with a result of the
discrimination, the recording method is switched.
[0004] As an image field discrimination method, a method is known
in which the difference in change in local densities between
gradation fields and character fields or the difference in the
local patterns is used. As an example of the former method, a
method has been disclosed in Jpn. Pat. Appln. KOKAI Publication No.
58-3374. The method has steps of dividing an image into small
blocks, and calculating the difference between a highest density
and a lowest density in each block. Moreover, if the difference is
larger than a threshold value, a discrimination is made that the
subject block is a character image field. If the difference is
smaller than the threshold value, a discrimination is made that the
subject block is a gradation image field. The above-mentioned
method is able to perform accurate discrimination if the image is
composed of only continuous gradation images, such as photographs,
and characters. However, there arises a problem of unsatisfactory
discrimination accuracy in a field, for example, a dot image, in
which local density change frequently occurs. There arises another
problem in that a gradation image having a sharp edge thereon is
incorrectly discriminated as a character field. As an example of
the latter method, a method has been disclosed in Jpn. Pat. Appln.
KOKAI Publication No. 60-204177. The foregoing method has steps of
Laplacian-filtering an image; binarizing the image; and performing
discrimination in accordance with a shape of, for example, a
4.times.4 pattern. The above-mentioned method is able to
discriminate even a dot image. However, also the above-mentioned
method had a problem in that an edge portion on a gradation image
can frequently incorrectly be discriminated as a character
image.
[0005] If the foregoing methods are combined with each other or if
a method is furthermore employed in which correction is performed
in accordance with results of discrimination of surrounding pixels
by using a characteristic that an image field is constant over a
somewhat wide area, the discrimination accuracy can be improved.
However, the scale of the circuit results in limitation of
reference fields to several pixels. Thus, a satisfactory
discrimination accuracy cannot be realized.
[0006] As described above, the method using information of the
local density among the conventional image field discrimination
methods suffers from a problem in that discrimination accuracy of
an edge portion of a gradation image having a micro structure
similar to that of a character and a rough dot image portion
deteriorates. Another problem arises in that a character formed by
thick lines and a portion in lines cannot easily be discriminated
as a character.
[0007] The known method of macroscopically analyzing the structure
of a document image, which is able to accurately discriminate
existence of a character or a gradation image, suffers from
unsatisfactory position-resolving power. Thus, accurate
discrimination in pixel unit cannot easily be performed.
BRIEF SUMMARY OF THE INVENTION
[0008] An object of the present invention is to provide an image
field discrimination method which is capable of realizing both
excellent discrimination accuracy and excellent position resolving
power and with which a portion in a characteristic can correctly be
discriminated, an image processing method using the same and an
image forming apparatus.
[0009] According to one aspect of the present invention, there is
provided an image processing apparatus comprising: the field
separating means for separating an original image into plural types
of fields in response to a first image signal obtained at a first
density of the supplied original image; characteristic value
calculating means for calculating a characteristic value of the
original image in response to a second image signal of the original
image obtained at a second density which is higher than the first
density; discrimination means for discriminating an image field of
the original image in accordance with the characteristic value
calculated by the characteristic value calculating means to
correspond to the type of the field separated by the field
separating means; and an image processing means for processing a
predetermined image processes corresponding to a result of
discrimination of the image field performed by the discrimination
means, on the second image signal.
[0010] The present invention having the above-mentioned structure
is different from a conventional method in which discrimination of
an image field is performed by one time. The present invention has
the structure that discrimination of a field in response to a rough
signal of an original image is performed. Moreover, a
characteristic value indicated by a dense signal of the original
image is obtained. Then, final discrimination of the field is, in
each field indicated by the rough signal, performed in accordance
with the characteristic value. As a result, further accurate
discrimination of a field can be performed.
[0011] That is, the conventional method has the structure that
discrimination of a field is performed only one time in response to
a dense signal. Thus, incorrect discrimination of an edge in a
photograph field as a character field can easily be made. However,
the present invention arranged to first perform macro field
discrimination in accordance with the rough signal is able to
prevent incorrect discrimination that an edge in the foregoing
photograph field is a character. Since the second process is
performed such that the field is discriminated in accordance with
the characteristic value indicated by the dense signal, a micro
character field in a character field and a background field are
discriminated from each other. Thus, the character field and the
background field can individually be detected. Since high contrast
similar to that of a character field is not provided for the
background field, that is, since low contrast is provided for the
background field, generation of noise on the background can be
prevented.
[0012] Therefore, the field discrimination according to the present
invention and arranged to perform two steps is able to prevent
incorrect discrimination between a photograph field and a character
field, detect a background field in a character field and thus
realize a background field free from any noise.
[0013] Also a method according to the present invention enables
both of a character field and a gradation field to satisfactorily
accurately be discriminated from each other because of the
above-mentioned reason.
[0014] The discrimination of an image field according to the
present invention can favorably be employed in a digital color
image forming apparatus. In this case, accurate discrimination of
an image field can be performed and an image in a satisfactory
state can be formed.
[0015] Additional objects and advantages of the invention will be
set forth in the description which follows, and in part will be
obvious from the description, or may be learned by practice of the
invention. The objects and advantages of the invention may be
realized and obtained by means of the instrumentalities and
combinations particularly pointed out hereinafter.
BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING
[0016] The accompanying drawings, which are incorporated in and
constitute a part of the specification, illustrate presently
preferred embodiments of the invention, and together with the
general description given above and the detailed description of the
preferred embodiments given below, serve to explain the principles
of the invention.
[0017] FIG. 1 is a block diagram showing the structure of an
essential portion of a digital color copying machine according to a
first embodiment of the present invention;
[0018] FIG. 2 is a block diagram showing an example of the
structure of an image field discrimination section;
[0019] FIG. 3 is a graph showing characteristics of one-pixel
modulation process and a two-pixel modulation process (the
relationship between density signals and pulse widths);
[0020] FIG. 4 is a flow chart of an operation of the digital color
copying machine shown in FIG. 4;
[0021] FIG. 5 is a histogram about brightness values in a general
document;
[0022] FIG. 6 is a flow chart of an example of a procedure which is
performed by a macro discrimination section and in which the type
of a field of pixels in which connection fields of different types
overlap is discriminated;
[0023] FIG. 7 is a block diagram showing the structure of a micro
discrimination section;
[0024] FIG. 8 is diagram showing a method which is employed by the
micro discrimination section and in which an image field is
discriminated;
[0025] FIG. 9 is a diagram showing an example of an image of an
original;
[0026] FIG. 10 is a diagram showing an example of a result of field
separation of the image of the original document shown in FIGS. 12A
and 12B;
[0027] FIG. 11 is a graph showing an example of distribution of
characteristic values (density change values and average densities)
in an usual character field;
[0028] FIGS. 12A and 12B are graphs showing an example of
distribution of characteristic values (density change values,
average density values and chroma) in characters on a
background;
[0029] FIG. 13 is a graph showing an example of distribution of
characteristic values (density change values and average densities)
in a continuous gradation field;
[0030] FIG. 14 is a graph showing an example of distribution of
characteristic values (density change values and average densities)
in a dot image field;
[0031] FIGS. 15A, 15B, 15C and 15D are diagrams showing examples of
results of micro discrimination in each field of the image of the
original document shown in FIG. 9 and separated in the macro
discrimination section;
[0032] FIG. 16 is a diagram showing a final result of the
discrimination;
[0033] FIG. 17 is a diagram showing the structure of a micro
discrimination section of a digital color copying machine according
to a second embodiment of the present invention;
[0034] FIG. 18 is a diagram showing the structure of an essential
portion of a digital color copying machine according to a third
embodiment of the present invention;
[0035] FIG. 19 is a diagram showing the structure of a micro
discrimination section according to a fourth embodiment of the
present invention; and
[0036] FIG. 20 is a diagram showing an image field discrimination
method adapted to the micro discrimination section according to the
fourth embodiment.
DETAILED DESCRIPTION OF THE INVENTION
[0037] Embodiments of the present invention will now be described
with reference to the drawings.
[0038] First Embodiment
[0039] FIG. 1 is a diagram showing an example of the structure of
an essential portion of an image forming apparatus (a digital color
copying machine which is hereinafter simply called an image copying
machine or a copying machine) having an image processing apparatus
to which an image field discrimination method according to the
present invention is applied.
[0040] The image copying machine incorporates an image input
section 1001, a color converting section 1002, an image field
discrimination section 1004, a filtering section 104, a signal
selector section 1005, an inking process section 1006, a gradation
process section 1007 and an image recording section 1008. The image
field discrimination method according to the present invention is
applied to the image field discrimination section 1004. Note that
editing processes including an expansion/reduction process and
trimming and masking processes not shown do not concern the present
invention. Therefore, sections for performing the above-mentioned
processes are disposed, for example, immediately posterior to the
image input section 1001.
[0041] The image input section 1001 reads an image of an original
document so as to produce an output of a color image signal 1101.
The color image signal 1101 indicates, for example, each
reflectance of R, G and B of each pixel of the original document,
the output of the color image signal 1101 being produced in the
form of three time-sequential signals obtained by two-dimensionally
scanning information of each pixel. At this time, the number of
read pixels per unit length is called a pixel density. The read
density in this embodiment is, for example, 600 dpi, that is, 600
pixels per 25.4 mm. Note that prescanning is performed at a low
density of, for example, 200 dpi in a vertical direction (in a
sub-scanning direction) as described later.
[0042] The color converting section 1002 converts the color image
signal 1101 indicating the reflectance of RGB into a color image
signal 1102 denoting the density of a coloring material (for
example, YMC) to be recorded. The reflectance of RGB and that of
YMC usually hold a very complicated non-linear relationship.
Therefore, a 3D table lookup method or a method in which a 1D table
lookup and a 3.times.3 matrix are combined with each other is
employed to perform the foregoing converting process. Specific
structures of the foregoing methods are disclosed in, for example,
Jpn. Pat. Appln. KOKAI Publication No. 1-055245 and Jpn. Pat.
Appln. KOKAI Publication No. 61-007774.
[0043] The image field discrimination section 1004 discriminates
the attribute of the pixel in the supplied (color) image signal
1102 to produce an output of an image field signal 1103. In this
embodiment, the attribute of a pixel includes three types which are
"character", "edge of gradation" and "smooth gradation". Therefore,
the image field signal 1103 is a signal having any one of values of
the three types.
[0044] The schematic structure of the image field discrimination
section 1004 will now be described. As shown in FIG. 5, the image
field discrimination section 1004 incorporates a macro
discrimination section 1201 and a micro discrimination section
1202. The macro discrimination section 1201 incorporates an image
separator section 1211, an image memory 1212, a CPU 1213, a program
memory 1214 and a field signal output section 1215. The micro
discrimination section 1202 incorporates a characteristic value
abstracting section 1311 for abstracting a plurality of (for
example, three) characteristic values, an image field
discrimination section 1312 for discriminating image fields of a
plurality of (for example, five) types and a discrimination signal
selector section 1313. The image field discrimination section 1004
is a section to which the image field discrimination method
according to the present invention is applied. Therefore, the
detailed structure and operation of the image field discrimination
section 1004 will be described later.
[0045] The filtering section 1003, subjects the YMC color image
signals 1102 to a plurality of filtering processes including a
sharpening process and a smoothing process in parallel. In this
embodiment, a three processes including a strong edge emphasizing
process, a weak edge emphasizing process and a smoothing filter
process are performed so as to produce results of the processes as
signals 1104, 1105 and 1106.
[0046] A copying machine usually treats a document image. An image
of the foregoing type contains character images and gradation
images mixed therein. The character image must sharply be
reproduced, while the gradation of the gradation image must
smoothly be reproduced. Since industrial printers and marketed
printers usually use dot images to express gradation, dot image
components must be removed. Therefore, the signal selector section
1005 responds to an image field signal 1103 transmitted from the
image field discrimination section 1004 to selectively switch
outputs of the various filtering processes from the filtering
section 1003. When the image field signal 1103 indicates a
character, the YMC color image signal 1102 is subjected to the
strong edge emphasizing filter so as to produce an output of a
result of the emphasizing process as the signal 1104 to a following
section. When the image field signal 1103 indicates an edge of
gradation, an output of a result obtained by subjecting the YMC
color image signal 1102 to the weak edge emphasizing filter is
produced as a signal 1105 to the following section. When the image
field signal 1103 indicates smooth gradation, the YMC color image
signal 1102 is subjected to the smoothing filter so as to produce
an output of a signal 1106 obtained by removing noise and dot image
components to the following section. As a result, the character
image can sharply be reproduced and the gradation image can
smoothly be reproduced.
[0047] Note that the filtering section 1003 may be arranged to
receive the image field signal 1103 so as to selectively switch the
filtering processes of the plural types in response to the image
field signal 1103. In the foregoing case, the signal selector
section 1005 may be omitted from the structure.
[0048] The inking process section 1006 converts the filtered YMC
color image signals into YMCK signals. Although black can be
expressed by superimposing coloring materials in YMC, a general
color recording process is performed by using YMCK coloring
materials including a black coloring material because the black
coloring material excels high density as compared with that
realized by stacking YMC coloring materials and the black coloring
material is a low cost material.
[0049] As a specific converting method, an UCR (Under Color
Reduction) method and a GCR method are known and actually used. A
calculating formula of the GCR method is expressed in the following
formula (1). In formula (1), CMY density signals are expressed as
CMY and CMYK density signals to be transmitted are expressed as C',
M', Y' and K'.
K'=k.multidot.min (C, M, Y)
C'=(C-K')/(1-K')
M'=(M-K')/(1-K')
Y'=(Y-K')/(1-K') (1)
[0050] The gradation process section 1007 will now be described.
When an electrophotograph image is recorded, time for which a laser
beam is turned on/off is modulated to express an intermediate
density. The gradation process section 1007 performs the modulation
process. Specifically, a pulse signal having a width in response to
the density signal is generated. In response to the pulse signal,
the laser beam is turned on/off. The structure is arranged such
that a method in which the pulse position is shifted forwards and a
method in which the pulse position is shifted rearwards can be
switched.
[0051] The modulation includes a two-pixel modulation method and a
one-pixel modulation method. The two-pixel modulation method is
performed such that the pulse positions for odd-numbered pixels are
shifted forwards and those for even-numbered pixels are shifted
rearwards. On the other hand, the one-pixel modulation method is
performed such that all of pixels are shifted forwards so as to be
recorded. Since the one-pixel modulation method has a structure
that pulses are turned on/off at cycles in one pixel units,
recording can be performed at a resolution in one pixel units. On
the other hand, the two-pixel modulation method arranged to have
cycles in two pixel units encounters deterioration in the
resolution as compared with the one-pixel modulation method.
However, the pulse width for expressing the same density can be
doubled, stability of the density can be improved. Thus, the
gradation expressing characteristic can be improved as compared
with the one-pixel modulation. An example of the relationship
between density signals and recordable gradations is shown in FIG.
6. Referring to FIG. 6, a curve 11 indicates the relationship
between density signals and pulse widths realized in a case of the
one-pixel modulation method. A curve 12 indicates the relationship
between density signals and pulse widths realized in a case of the
two-pixel modulation. The one-pixel modulation method is a method
suitable to record a character image, while the two-pixel
modulation method is a method suitable to record a gradation
image.
[0052] In this embodiment, selection of the two-pixel modulation
process or the one-pixel modulation process is performed in
response to the image field signal 1103. Specifically, when the
image field signal 1103 indicates a character, the one-pixel
modulation process is selected. When the image field signal 1103
indicates an edge of a gradation image or a smooth section, the
two-pixel modulation process is selected. As a result, an image of
a gradation field can be expressed with smooth gradation and a
multiplicity of gradation levels. A sharp image of a character
field can be recorded with a high resolution.
[0053] The image recording section 1008 will now be described. In
this embodiment, the image recording section 1008 is adapted to an
electrophotography method. The principle of the electrophotographic
method will be described briefly. Initially, a laser beam or the
like is modulated in response to an image density signal. Then, the
modulated laser beam is applied to a photosensitive drum. An
electric charge corresponding to a quantity of applied light is
generated on the photosensitive surface of the photosensitive drum.
Therefore, a laser beam is applied to scan the axial direction of
the photosensitive drum to correspond to the scanning position of
the image signal. Moreover, the photosensitive drum is rotated so
as to be scanned. Thus, a two-dimensional charge distribution
corresponding to the image signals is formed on the photosensitive
drum. Then, the toner electrically charged by a developing unit is
allowed to adhere to the surface of the photosensitive drum. At
this time, toner in a quantity corresponding to the potential is
allowed to adhere to the surface so that an image is formed. Then,
the toner on the photosensitive drum is transferred to the surface
of recording paper through a transfer belt. Finally, the toner is
melted by a fixing unit so as to be fixed on the recording paper.
The above-mentioned process is sequentially performed for each of
the YMCK toner so that a full color image is recorded on the
surface of the recording paper.
[0054] Operation of Image Forming Apparatus
[0055] The operation of the copying machine shown in FIG. 1 will
now be described with reference to a flow chart shown in FIG. 7.
The copying machine according to this embodiment is arranged to
perform an operation for copying an image of an original document
such that the image input section 1001 reads (scans) an image two
times.
[0056] When a first operation for scanning an image is performed,
the scanning operation is performed at a high velocity. Thus, the
image is read at a rough density in the sub-scanning direction. A
read image signal is color-converted by the color converting
section 1002, and then supplied to the macro discrimination section
1201 of the image field discrimination section 1004 (steps S1 and
S2). The image separator section 1211 in the macro discrimination
section 1201 converts the image signal into a plurality of
characteristic value signals (step S3). Thus, the contents of one
page of the original document are written on the image memory 1212
(step S4). The above-mentioned process is performed simultaneously
with the operation of the image input section 1001 for scanning the
original document. After the original document has been scanned and
image information has been recorded in the image memory 1212, the
CPU 1213 performs a field separation process (step S5). A result of
the field separation process is stored in the image memory 1212
(step S6).
[0057] After the field separation process has been completed by the
CPU 1213, the image input section 1001 starts second scanning of
the image (step S7). The second image scanning operation is
performed such that the image is read at a low velocity. The image
signal read by the image input section 1001 is subjected to a color
conversion process in the color converting section 1002 (step S8),
and then supplied to the image field discrimination section 1004
and the filtering section 1003. The image signal supplied to the
image field discrimination section 1004 is supplied to the micro
discrimination section 1202 so as to be supplied to the
discrimination process (step S9). In synchronization with the
second image scanning process, the field signal stored in the image
memory 1212 is supplied through the field signal output section
1215 of the macro discrimination section 1201. As a result, the
image field signal 1103 is transmitted from the discrimination
signal selector section 1313 (step S10).
[0058] On the other hand, the image signal transmitted from the
color converting section 1002 is allowed to pass through the
filtering section 1003, the signal selector section 1005, the
inking process section 1006 and the gradation process section 1007,
and then transmitted to the image recording section 1008. Each of
the signal selector section 1005, the inking process section 1006
and the gradation process section 1007 selects a signal and a
process in response to the image field signal 1103 supplied in
synchronization with execution of each process (steps S11 to
S14).
[0059] Description of Image Field Discrimination Section
[0060] The image field discrimination section 1004 shown in FIG. 2
will now be described.
[0061] Structure and Operation of Macro Discrimination Section
[0062] The macro discrimination section 1201 performs field
separation in accordance with the major structure of the image. In
this embodiment, an image of an original document is separated into
the following five types of fields.
[0063] 1. Usual Character Field
[0064] 2. Characters on Background
[0065] 3. Continuous Gradation Field
[0066] 4. Dot Gradation Field
[0067] 5. Other Field
[0068] The usual character field is a field in which a white
background and characters and graphics are written on the white
background. A major portion of usual documents are included in the
above-mentioned field. The dot gradation field on a background has
a gradation background which exists as the background of
characters. The dot gradation field on a background is a field in
which the background is divided in terms of colors to emphasize or
classify a character or characters are superimposed on a gradation
image to make a description. The former case is exemplified by a
catalog, while the latter case is exemplified by a map. The
continuous gradation field is a field having gradation, such as a
human figure or a background image, and realized by recording the
gradation image by a continuous gradation method, such as a silver
salt photograph method or a sublimation type transfer method. Even
if a dot image printing method is employed, an image field in which
the dot images have sufficiently high frequencies with which dot
image components are eliminated from the image signal is included
in the continuous gradation field. Similarly to the continuous
gradation field, the dot gradation field is a field of an image of
a human figure or a background image. The dot gradation field is a
field in which dot image printing is performed to express a
gradation image.
[0069] A large portion of images is basically classified into any
one of the four fields. Some images are not included in the
foregoing classification or are difficult to be classified. For
example, an image produced by computer graphics, such as a
character string having gradation given to the entire field
thereto, applies to the foregoing case. A field of the foregoing
type is classified as "the other field".
[0070] The structure and operation of the macro discrimination
section 1201 will now be described. The image separator section
1211 separates the color image signal 1102 transmitted from the
color converting section 1002 into image data in a plurality of
planes in accordance with the difference in the density of
peripheral pixels and a state of chroma. Separated image data is
sequentially stored in the image memory 1212. In this embodiment,
the image memory 1212 has a capacity corresponding to the number of
the planes of the images. Thus, the separated image signals for one
page are completely stored in the image memory 1212.
[0071] The image separator section 1211 calculates brightness I and
chroma S from the YMC color image signal 1102 in accordance with
the following formula (10):
I=(C+M+Y)/3
S=(C-M)2+(M-Y)2+(Y-C)2 (2)
[0072] The brightness value I is a quantity indicating the density
of an image. The brightness value I is "0" in a case of a white
image, while the same is "1" in a case of a black image. The chroma
S is "0" in a case of achromatic color, while the same is enlarged
in proportion to the degree of chromatic color. Then, change in the
brightness value I in the scanning direction is detected, and then
discriminates a field in which the brightness value I is changed as
a dot image. Then, histogram with respect to the brightness value I
in a certain field is produced. As shown in FIG. 8, the histogram
of a usual original document has three peaks, that is, a background
density, a gradation density and a black density. Lowest densities
among the foregoing peaks are made to be th1 and th2. Pixels other
than the dot images are discriminated such that pixels having the
brightness value I which is not larger than the threshold value th1
are discriminated as the background pixels, those having the
threshold value larger than the threshold value th1 and not larger
than the threshold value th2 are discriminated as the gradation
pixels and those having the threshold value which is larger than
the threshold value th2 are discriminated as the character pixels.
Moreover, threshold value th3 for the chroma S is discriminated.
Pixels having a chroma S which is smaller than the threshold value
th3 and a brightness value I which is larger than the threshold
value th2 are discriminated as black pixels. The other pixels are
discriminated as gray pixels. That is, in accordance with the rough
image signals of the original image, the original image is
classified into seven types, which are a character image (an image
containing dense pixels), a gradation image (an image containing
pixels having a low density similar to that of a photograph), a
background image (an image containing pixels having a very low
density similar to that of a background), a color image (an image
containing colored pixels), a gray image (an image containing gray
pixels), a black image (an image containing black pixels) and a dot
image (an image having the density which is frequently and greatly
changed similarly to a dot image). Classified image data is
temporarily stored in the image memory 1212 together with a result
of the classification (information about the field separation).
Note that each image is made to be a binary image whether or not
the image has the foregoing characteristic.
[0073] In accordance with a program code stored in the program
memory (for example, a ROM) 1214, the CPU 1213 performs a field
discrimination process while the CPU 1213 makes a reference to the
contents of separated image data stored in the image memory 1212 so
that information about field separation is modified. Then, modified
information about field separation for e.g. each pixel written on
the image memory 1212. That is, continuous pixels in the three
image field, that is, the character image, the gradation image and
the dot image, are unified so that connection fields in rectangular
units are produced. Then, the characteristic value of each
connection field is calculated so that the type of the field is
discriminated. The types of the fields include, for example, the
usual character field mainly containing characters and a photograph
field which is a gradation image. The position and size of the
connection field (information of a connection field) and the type
of the field are again stored in the image memory 1212. Then,
pixels having overlapping connection fields of different types are
subjected to, for example, a process as shown in FIG. 9. Thus, in
accordance with information of a connection field and the type of
the field, the type of the pixel is discriminated.
[0074] The procedure of the flow chart shown in FIG. 6 is performed
such that a field discriminated as a photograph field (step S401)
is discriminated as a dot gradation field if a dot image exists in
the foregoing field (steps S405 and S406). If a gradation image
exists, the field is discriminated as a continuous gradation field
(steps S407 and S408). If both of the dot image and the gradation
image do not exist, the field is discriminated as characters on a
background (steps S407 and S409). If the field is not the
photograph field, gradation image data is deleted from the field
containing a gradation image and the field discriminated in step
S408 as a continuous gradation field (steps S402 to S404). As a
result of the above-mentioned process, at least four types of
modified information of field discrimination are stored in the
image memory 1212.
[0075] Information about field separation stored in the image
memory 1212 is read by the field signal output section 1215 in
synchronization with a second reading signal from the image input
section 1001 so as to be transmitted as a field separation signal.
Since the density of pixels indicated with the information about
field separation in the image memory 1212 and the density of pixels
indicated with the image signal from the image input section 1001
are different from each other, the density denoted by the field
separation signal is converted so that the two pixel densities are
matched with each other before transmittance.
[0076] In this embodiment, the field separation signal is expressed
in a 3-bit signal form. The relationship between the values of the
3-bit signals and the fields are as follows:
1 Values Of Region Separation Signals Regions "0" Usual Character
Field "1" Characters on Background "2" Continuous Gradation Field
"3" Dot Gradation Field "4" Other Fields
[0077] Another arrangement may be employed in which the field
separation signal is formed into a 5-bit signal and the 5-bit
signals represent the five fields, respectively.
[0078] Structure and Operation of Micro Discrimination Section
[0079] The micro discrimination section 1202 discriminates the
field by paying attention to a micro difference in the image. The
detailed structure of the micro discrimination section 1202 is
shown in FIG. 10. The micro discrimination section 1202 comprises a
characteristic value abstracting section 1311 for abstracting three
characteristic values, an image field discrimination section 1312
for discriminating five types of image fields and a discrimination
signal selector section 1313.
[0080] The characteristic value abstracting section 1311
incorporates a density calculation section 1311d, a density change
value abstracting section 1311a for abstracting the three
characteristic value, an average density abstracting section 1311b
and a chroma abstracting section 1311c.
[0081] The density change value abstracting section 1311a abstracts
the degree of change in the density of a portion around a pixel of
interest. Initially, the density calculation section 1311d
calculates density signal D from the YMC color image signal 1102.
The density is calculated such that a weighted linear sum of the
YMC density signals is calculated as indicated with the following
formula (2):
D=Ky.multidot.Y+Km.multidot.M+Kc.multidot.C (3)
[0082] where Ky=0.25, Km=0.5 and Kc=0.25
[0083] Then, the density change value abstracting section 1311a
calculates change in the density in a 3 pixel.times. 3 pixel block
in the vicinity of the pixel of interest so as to transmit a
density change value signal DD. Assuming that densities of pixels
in the 3 pixel.times. 3 pixel block are D1, D2, D3, . . . , D9, the
density change value signal DD can be expressed by the following
formula (3), in which Max (A1, A2, . . . , An) indicates a maximum
value among A1, A2, . . . , An.
DD=Max (.vertline.D1-D9.vertline., .vertline.D2-D8.vertline.,
.vertline.D3-D7.vertline., .vertline.D4-D6.vertline.) (4)
[0084] Note that the density change may be abstracted by another
method, such as a BAT method disclosed in Jan. Pat. Appln. KOKOKU
Publication No. 04-05305. A formula for calculating the density
change value for use in the BAT method is as shown in the following
formula (4):
DD=Max (D1, D2, . . . , D9)-Min (D1, D2, . . . , D9) (5)
[0085] Although this embodiment has the structure that the
reference field is 3 pixel.times.3 pixel range, another field may
be employed, for example, a larger field, for example, a 4
pixel.times.4 pixel field or a 5 pixel.times.5 pixel field or a
non-square 3 pixel.times.5 pixel field. If the reference field is
enlarged, accuracy in abstracting the characteristic value is
usually improved. However, the size of the hardware is enlarged.
Therefore, an appropriate size adaptable to an object and required
performance must be employed.
[0086] The average density abstracting section 1311b abstracts the
density of the pixel of interest. That is, the average density
abstracting section 1311b receives the density signal D transmitted
from the density calculation section 1311d so as to calculate an
average value of the density signals of 3 pixel.times. 3 pixel
field in the vicinity of the pixel of interest. Then, an output of
a result of the calculation is produced as an average density
signal DA. The average density signal DA indicates the density of a
portion around the pixel of interest.
[0087] The chroma abstracting section 1311c abstracts the chroma of
the pixel of interest. In response to YMC signals of the pixel of
interest, a chroma signal DS expressed with the following formula
(5) is generated:
DS=(C-M)2+(M-Y)2+(Y-C)2 (6)
[0088] The chroma signal DS indicates the chroma of the pixel of
interest, that is, whether or not the pixel of interest is colored.
In a case of a achromic pixel, such as white, black or gray, DS is
substantially "0". In a case of red or blue, DS is substantially a
maximum value of "2".
[0089] The image field discrimination section 1312 will now be
described. The first to fifth image field discrimination sections
1312a to 1312e perform image field discrimination processes
suitable to the five fields separated by the image separator
section 1211 of the macro discrimination section 1201. The image
field discrimination sections 1312a to 1312e receive characteristic
value signals DD, DA and DS transmitted from the characteristic
value abstracting sections 1311a to 1311c, discriminate an image
field in response to the received signals and generate an image
field signal DT. The generated image field signal is a 2-bit signal
having values "0", "1" and "2". As for the relationship between the
values and the image fields, a smooth gradation field is expressed
when DT=0. An edge of a gradation image field is expressed when DT=
1. An inside portion of a character and an edge are expressed when
DT=2. The first to fifth image field discrimination sections 1312a
to 1312e have different discrimination methods as will now be
described with reference to FIG. 11. The first image field
discrimination section 1312a discriminates a point having the
density change value DD which is larger than a threshold value T1
and a point having an average density which is larger than a
threshold value T2 as a character field (DT=2). The second image
field discrimination section 1312b discriminates a point having the
density change value DD which is larger than a threshold value T3
and a point having an average density which is higher than a
threshold value T4 and chroma DS which is lower than a threshold
value T5 as a character field (DT=2). The threshold value T3 is
made to be a value larger than the threshold value T1. The third
image field discrimination section 1312c discriminates a point
having a density change value DD which is larger than a threshold
value T6 as an edge (DT=1) of a gradation image field. The third
image field discrimination section 1312c discriminates the point as
a smooth gradation image field (DT=0) in the other cases. The
fourth image field discrimination section 1312d is arranged to make
the image field signal DT to always be a value "0". That is, the
fourth image field discrimination section 1312d discriminates that
all of fields are smooth gradation fields. The fifth image field
discrimination section 1312e discriminates the point as an edge
field (DT=1) if the density change value DD is larger than a
threshold value T7 and discriminates the same as a gradation field
(DT=0) if the density change value DD is smaller than the threshold
value T7. Note that the threshold values T1 to T7 are predetermined
threshold value for performing discrimination which must be
discriminated appropriately adaptable to the resolution
characteristic of an input system and that of the color conversion
process section. Appropriate threshold values will be described
later.
[0090] The discrimination signal selector section 1313 selects the
five types of image field signals DT transmitted from the image
field discrimination sections 1312a to 1312e in response to the
field separation signal transmitted from the field signal output
section 1215 of the macro discrimination section 1201. That is,
when the fields separation signal is "0", an output signal from the
first image field discrimination section 1312a is selected. When
the field separation signal is "1", "2", "3" or "4", an output
signal from the second, third, fourth or fifth image field
discrimination sections is selected so as to be transmitted as the
image field signal 1103.
[0091] As a result, the two steps of the field discrimination
according to the present invention are performed. Thus, incorrect
discrimination between a photograph field and a character field can
be prevented. Moreover, a background field in the character field
can be detected so that a background field free from any noise is
realized.
[0092] Specific Operation
[0093] The specific operation will now be described such that the
original document shown in FIG. 9 is taken as an example.
[0094] The original document shown in FIG. 9 has a unnatural
structure as compared with a usual document image to easily
describe the structure. A field 1601 is a field in which a sentence
is written, the field 1601 being composed of black and red
characters. A field 1602 is a table field in which black characters
and ruled lines are drawn on a light color background which is
sectioned in terms of color. A field 1603 is a field to which a
gradation image realized by a silver salt photograph is pasted. A
field 1604 is a field in which a gradation image is recorded by a
dither method, that is, a dot image modulation method.
[0095] An operation will now be described with reference to FIG. 4,
the operation being performed when the image of the original
document is read by a scanner section not shown so that a copied
image of the original document is transmitted from the printer
section 2. As described above, the image of the original document
shown in FIG. 9 is read by the image input section 1001 shown in
FIG. 4, the image being read as an electric signal. Then, the
electric signal is converted into the color image signal 1102
indicating the quantity of YMC toner in color converting section
1002.
[0096] In response to the YMC color image signal 1102, the image
field discrimination section 1004 performs the image field
discrimination process. The macro discrimination section 1201
performs the field separation in accordance with the discriminated
characteristic of the structure. Since a reference to a wide field
is made when the discrimination is performed, the field separation
into the above-mentioned classifications can significantly
accurately be performed. An example of a field separation signal
denoting a result of the process performed by the macro
discrimination section 1201 is shown in FIG. 10. In FIG. 10, fields
having respective field separation signal values of "0", "1", "2"
and "3" are expressed with white (a field 1601), diagonal lines
facing lower left positions (a field 1602), diagonal lines facing
lower right positions (a field 1603) and a cross diagonal lines (a
field 1604). Note the illustrated example of the original document
is free from "the other field".
[0097] The micro discrimination section 1202 discriminates the
usual character field into a character field and the other field.
Moreover, the micro discrimination section 1202 discriminates the
characters on a background into a character field and the other
field (a background field). The discrimination is performed in
pixel units. The discrimination method employed by the micro
discrimination section 1202 will now be described such that a
characteristic value distribution in each field is described.
[0098] In the usual character field, characters are generally
recorded on a white background. Characters in yellow or blue having
very low densities and very fine characters having sizes of 5
points or smaller are sometimes recorded. An example of
two-dimensional distribution between the density change value
signals DD in the usual character fields and average density
signals DA is shown in FIG. 14. Characters are distributed in a
field 1801 shown in FIG. 14, while portions other than the
characters are distributed about a field 1802 which is the center.
Therefore, a boundary line 1803 causes the character field and the
field other than the characters to be discriminated. The position
of the boundary line corresponds to the threshold values T1 and T2
for performing discrimination.
[0099] In the characters on a background, the characters are placed
on a thin background. Although the background is sometimes formed
with thin ink, the background is usually formed by dot image
recording in the form of a usual printed matter. Since the
visibility of the characters on the background excessively
deteriorates, thin color and small characters are not usually
employed in the character portion. In a major portion of the
character portions, deep colors, such as black or red and a thick
characters having a size of 7 points or larger are used. An example
of two-dimensional distribution in the foregoing region between the
density change value signals DD and average density signals DA is
shown in FIG. 15A, and an example of two-dimensional distribution
between average density signals DA and Chroma signals DS in FIG.
15B. In FIG. 15B, black characters are distributed in a field 1901,
while color characters are distributed in a field 1902. In FIG.
15A, a distribution field of pixels in the background portion is
distributed in a field 1903. At this time, the character field and
the field other than the character field can be discriminated from
each other by dint of the threshold values T4 and T5 indicated by a
boundary line 1904, as shown in FIG. 15B. If the average density DA
is high despite the density change value being "0" as shown in FIG.
15A, the threshold value T3 indicated with a boundary line 1901
enables the discrimination from characters to be performed.
Therefore, the inside portion in a black character which cannot
easily be discriminated by the conventional method can correctly be
discriminated.
[0100] Examples of two-dimensional distribution between the density
change value signals DD and average density signals DA in the
continuous gradation field and the dot gradation field are shown in
FIGS. 16 and 17, respectively. As shown in FIG. 16, the overall
density change value DD is small in the continuous gradation field,
while the same is somewhat large in edge portions. In this
embodiment, the threshold value T6 for performing discrimination
indicated with a boundary line 1103 is used to discriminate a
gradation image field and edges of the gradation image field. Since
gradation is expressed with dot images in the dot gradation field
as shown in FIG. 17, the density change value DD is enlarged. Since
removal of the dot image component from the dot gradation field
causes the quality of the image to be improved, the image field
signal DT is made to be "0" regardless of the characteristic value
signals DD, DA and DS.
[0101] Since the conventional discrimination process using only the
micro characteristics has not employed the field separation process
according to this embodiment, the threshold values for the
discrimination cannot be switched to be adaptable to the five types
of the regions. Therefore, the conventional method has used
discrimination threshold values obtainable from the same
discrimination boundary as indicated by boundary lines 1810, 1910,
11010 and 11110 shown in FIGS. 14 to 15D. It leads to a fact that
fine characters which are usual characters and an edge field in a
gradation field do not satisfy an expected result of
discrimination. Therefore, incorrect discrimination takes place and
thus a satisfactory discrimination accuracy cannot be obtained.
[0102] This embodiment having the structure that the field
separation is performed by the macro discrimination process
followed by selecting a micro discrimination boundary suitable to
each region enables accurate discrimination to be performed with
high resolving power. That is, a result of discrimination (see FIG.
14) obtained from the first image field discrimination section
1312a of the micro discrimination section 1202 is selected for a
region discriminated by the macro discrimination section 1201 as
the usual character field (having a field separation signal "0"). A
result of discrimination (see FIG. 15A to 15D) obtained from the
second image field discrimination section 1312b of the micro
discrimination section 1202 is selected for a region discriminated
by the macro discrimination section 1201 as the characters on a
background (having a field separation signal "1"). A result of
discrimination (see FIG. 16) obtained from the third image field
discrimination section 1312c of the micro discrimination section
1202 is selected for a region discriminated by the macro
discrimination section 1201 as the continuous gradation field
(having a field separation signal "2"). A result of discrimination
(see FIG. 17) obtained from the fourth image field discrimination
section 1312d of the micro discrimination section 1202 is selected
for a region discriminated by the macro discrimination section 1201
as the dot gradation field (having a field separation signal "3").
A result of discrimination obtained from the fifth image field
discrimination section 1312e of the micro discrimination section
1202 is selected for a region discriminated by the macro
discrimination section 1201 as the other fields (having a field
separation signal "4").
[0103] Examples of discrimination of an image of an original
document as shown in FIG. 9 performed by the image field
discrimination sections 1312a to 1312e of the micro discrimination
section 1202 are schematically shown in FIGS. 15A, 15B, 15C and
15D. Examples of selection of the image field signal DT in the
discrimination signal selector section 1313 in response to the
field separation signal is shown in FIG. 19. In FIGS. 15A, 15B,
15C, 15D and 16, a region having the image field signal "2", that
is, a field discriminated as the character field is expressed in
black. The other fields are expressed in white. FIG. 15A shows a
result of discrimination performed by the first image field
discrimination section 1312a, FIG. 15B shows a result of
discrimination performed by the second image field discrimination
section 1312b and FIGS. 15C and 15D show result of discrimination
performed by the third and fourth image field discrimination
sections 1312c and 1312d. As can be understood from results of
comparison among FIGS. 15A, 15B, 15C, 15D and 16, the image field
signal DT does not realize an accurate discrimination in the fields
other than the adapted field. However, when the discrimination
signal selector section 1313 selects only image field signals
adaptable to the field separation signals which are results of
discrimination performed by the macro discrimination section 1201,
an accurate result of discrimination can be obtained with the final
image field signal.
[0104] Second Embodiment
[0105] A modification of the micro discrimination section according
to the first embodiment will now be described. Another example of
the structure of the micro discrimination section 1202 is shown in
FIG. 20. Note that the same elements as those shown in FIG. 7 are
given the same reference numerals and only different elements will
now be described.
[0106] As shown in FIG. 20, the micro discrimination section 1202
incorporates a density calculation section 1311d, three
characteristic value abstracting sections, that is, the density
change value abstracting section 1311a, average density abstracting
section 1311b and the chroma abstracting section 1311c, three
threshold value registers 1401a to 1401c, comparators 1402a to
1402c and a total discrimination section 1403. In each of the
threshold value registers 1401a to 1401c, five discrimination
threshold values corresponding to five types of fields are stored.
Any one of the threshold values is selected in response to a field
separation signal supplied from the macro discrimination section
1201 to the threshold value registers 1401a to 1401c. A threshold
value signal transmitted from the selected threshold value register
is subjected to a comparison with characteristic value signals DD,
DA and DS in the comparators 1402a to 1402c. Results of the
comparison are transmitted as binary comparison signals.
[0107] The comparison signals corresponding to the characteristic
values are subjected to predetermined logical calculations in the
total discrimination section 1403 so that a final image field
signal 1404 is transmitted. The total discrimination section 1403
makes a reference to both of the supplied comparison signal and the
field separation signal so as to transmit an image field
signal.
[0108] As described above, a structure in which the micro
discrimination section 1202 is composed of a pair of a total
discrimination section and threshold value registers enables a
discrimination process similar to that according to the first
embodiment to be realized. Since this modification has the
structure that the discrimination processes are commonly performed,
the flexibility is lowered. However, the size of the circuit can be
reduced as compared with the first embodiment.
[0109] Third Embodiment
[0110] Another example of a color copying machine incorporating an
image processing apparatus employing the image field discrimination
method according to the present invention will now be described.
FIG. 18 shows an example of the structure of an essential portion
of the color copying machine according to a third embodiment. The
same elements as those shown in FIG. 1 are given the same reference
numerals and only different elements will now be described. In the
structure shown in FIG. 21, a color image signal of an image of an
original document read by the image input section 1001 is allowed
to pass through the color converting section 1002 so as to be
stored in a page memory 1411. The foregoing structure is a
remarkable difference from the first embodiment. Then, the
structure and operation of this embodiment will be described
briefly.
[0111] Initially, an image of an original document is read by the
image input section 1001 as RGB image signals. Then, the color
converting section 1002 converts the RGB image signals into color
image signals indicating densities in YMC. The converted YMC color
image signals 1102 for one page are stored in the page memory 1411.
On the other hand, the YMC color image signal 1102 is also supplied
to a macro discrimination section 1412. The macro discrimination
section 1412 has a structure similar to that of the macro
discrimination section 1201 of the image field discrimination
section according to the first embodiment so as to perform a
similar operation. That is, the YMC color image signal 1102
supplied to the macro discrimination section 1412 is separated into
image data in a plurality of planes by the image separator section
1211 shown in FIG. 5. Separated image data is sequentially stored
in the image memory 1212. In accordance with the program code
stored in the program memory 1214, the CPU 1213 separates the
fields while the CPU 1213 makes a reference to the contents of
separated image data stored in the image memory 1212 so as to write
a result of separation on the image memory 1212.
[0112] After the discrimination process has been completed and thus
all of results of the field separation have been written on the
image memory 1212, the image signals stored in the page memory 1411
are sequentially read from the same. In synchronization with the
reading operation, information of field separation stored in the
image memory 1212 is read through the field signal output section
1215. Since the pixel density indicated by the field separation
information in the image memory 1212 and pixel density indicated by
the image signal in the page memory (1411) are different from each
other, the density indicated by the field separation signal is
converted so that the densities of the both signals are matched
with each other.
[0113] The YMC color image signal transmitted from the page memory
1411 and the field separation signal transmitted from the macro
discrimination section 1412 are supplied to a micro discrimination
section 1413. The structure and operation of the micro
discrimination section 1413 are similar to those of the micro
discrimination section 1202 according to the first embodiment. That
is, characteristic value signals DD, DA and DS are, by the three
characteristic value abstracting sections, generated from the
supplied YMC color image signals. Then, the five image field
discrimination sections 1312 generate respective image field
signals from the characteristic value signals. Finally, the
discrimination signal selector section 1313 selects results of
discrimination (image field signals) performed by the five image
field discrimination sections 1312 in response to the field
separation signal transmitted from the macro discrimination section
1412 so as to transmit a final image field signal 1103.
[0114] The YMC color image signal transmitted from the page memory
1411 is allowed to pass through the filtering section 1003, the
signal selector section 1005, the inking process section 1006 and
the gradation process section 1007 so as to be recorded by the
image recording section 1008. The signal selector section 1005 and
the gradation process section 1007 switch the process thereof in
response to the image field signal 1103 transmitted from the micro
discrimination section 1413. Since the foregoing switching
operation is similar to that according to the first embodiment, the
switching operation is omitted from description.
[0115] Also the third embodiment is able to perform an image
process similar to the image process which can be performed by the
first embodiment. Therefore, the image field signal 1103 which is
an accurate signal similar to that obtainable from the first
embodiment can be obtained. When a signal process suitable to the
type of the image is selected, a character field can be reproduced
with a high resolution. Moreover, a gradation image field can
smoothly be reproduced.
[0116] Since the third embodiment has the structure that the color
image signal is stored in the page memory 1411, the necessity of
reading and scanning an image of an original document two times as
is required for the first embodiment can be eliminated. Therefore,
the same signals can be employed to perform the macro
discrimination and to read the image of the original document.
Therefore, a necessity of considering an influence of deviation of
the reading position which is exerted at each scanning operation
can be eliminated. When the capacity of the page memory 1411 and so
forth is enlarged to correspond to a plurality of pages, another
original document can be read immediately after an original
document has been read. Therefore, in a case where an automatic
document feeder or the like is used to sequentially copy an
original document composed of a plurality of pages, the sequential
copying operation can be performed at a high velocity.
[0117] Although the foregoing embodiment uses change in the
density, the average density and chroma as the characteristic
values which are abstracted by the micro discrimination section,
the abstracted values are not limited to the foregoing factors. For
example, distribution of frequencies in a block or consistency with
a predetermined pattern may be employed as the characteristic
value.
[0118] As described above, the image copying machine according to
this embodiment has the structure that the macro discrimination
section 1201 uses a macro structural characteristic of an original
image to separate a character field and a gradation image field
from each other in response to a supplied rough image signal of an
original document. Then, a result of image field discrimination DT
of the original image performed by the micro discrimination section
1202 in response to a coarse image signal of the original image
adaptable to the result of the separation is selected. Then, an
image field signal denoting the result of the final image field
discrimination is transmitted. In response to the image field
signal, the processes which must be performed by the filtering
section 1003 and the gradation process section 1007 are selectively
switched. Thus, a character field is subjected to an edge
emphasizing process and a high-resolution recording process, while
a gradation image field is subjected to a multi-gradation
processes. As a result, an image in which both of a character field
and a gradation field in an image of the original document are
satisfactorily reproduced can be generated and/or recorded. That
is, the image field discrimination method and the image processing
apparatus according to this embodiment are able to perform
discrimination an edge of a gradation image and an edge of a
character from each other which has been difficult for the
conventional technique. Moreover, the method and apparatus
according to this embodiment are able to perform discrimination of
a character placed on a dot image and a field around the character.
Moreover, the foregoing discrimination can accurately be performed
with large resolving power.
[0119] When an image is encoded, switching of the encoding method
by using an image field signal denoting a result of final reference
numeral enables an image encoding method free from considerable
distortion occurring due to the encoding process and exhibiting a
high compression ratio to be realized.
[0120] Fourth Embodiment
[0121] A fourth embodiment will now be described which is a
modification of the first embodiment of the present invention. Also
this modification has a structure similar to that according to the
first embodiment. This embodiment is different from the first
embodiment in the structure of the micro discrimination
section.
[0122] The structure of the micro discrimination section according
to this modification is shown in FIG. 22. The micro discrimination
section incorporates a first segment characteristic abstracting
section 2201, a second segment characteristic abstracting section
2202, a density abstracting section 2203, a first image field
discrimination section 2204, a second image field discrimination
section 2205, a third image field discrimination section 2206, a
fourth image field discrimination section 2207, a fifth image field
discrimination section 2208 and a signal selector section 2209.
[0123] The first segment characteristic abstracting section 2201
makes a reference to an image signal in a rectangular field having
a size of 5 pixels.times.5 pixels in the vicinity of a pixel of
interest so as to detect a segment component in a vertical or
horizontal direction. Then, the first segment characteristic
abstracting section 2201 produces an output of a first segment
characteristic value signal (SX) 2251. The foregoing signal
indicates a degree of the segment structure in the rectangular
field. If a vertical or a horizontal segment structure exists, a
signal having a large value is transmitted. If no segment structure
exists or if the density is constant, a signal having a small value
is transmitted.
[0124] The second segment characteristic abstracting section 2202
detects a segment component in a diagonal direction existing in the
rectangular field in the vicinity of the pixel of interest. Then,
the second segment characteristic abstracting section 2202
transmits a second segment characteristic value signal (SB) 2252
indicating a degree of the diagonal segment structure in the
rectangular field.
[0125] The structure and discrimination principle of the first and
second segment characteristic abstracting sections are disclosed in
Jan. Pat. Appln. SHUTSUGAN Pubulication No. PH10-009480 (TITLE OF
THE INVENTION: IMAGE PROCESSING APPARATUS).
[0126] The density abstracting section 2204 abstracts a density
component (DD) 2254 in accordance with the following formula:
DD=Ky.multidot.Y+Km.multidot.M+Kc.multidot.C
[0127] where Ky=0.25, Km=0.5 and Kc=0.25
[0128] The image field discrimination sections will now be
described. The first to fifth image field discrimination sections
perform image field discrimination processes adaptable to five
types of fields separated by the field separator section, similarly
to those of the first embodiment. Each of the image field
discrimination sections receives the characteristic value signal
transmitted from the characteristic value abstracting section so as
to discriminate the image field signal. A discrimination process
which is performed in each image field discrimination section is
shown in FIG. 23.
[0129] This modification has a structure that the micro
discrimination section employs a discrimination method on the basis
of the linearity of an image element. Therefore, accuracy to
discriminate a character and a dot image from each other can be
improved. Therefore, this modification is suitable when
reproducibility of fine characters is required.
[0130] That is, the fourth embodiment has the structure that the
segment components in the main scanning direction and the
sub-scanning direction. Therefore, the background field can
furthermore accurately be detected. That is, if discrimination is
performed simply in accordance with change in the density or the
like, there is apprehension that incorrect discrimination is
performed such that a background in a constant color as a character
field. Since distribution of the segments (edges) in the main
scanning direction and the sub-scanning direction is detected, a
correct character field can be detected because a character image
contains a multiplicity of segments (edge components).
[0131] As described above, the present invention is able to
accurate image field discrimination of a supplied image with large
resolving power. Moreover, a macro structure characteristic of an
image is used to separate a character field and a gradation image
field from each other. Then, a result of an image field
discrimination suitable for each of the separated fields is
selected so as to obtain a result of final image field
discrimination. Thus, discrimination between an edge of a gradation
image and an edge of a character and that between a character
placed on a dot image and a field around the character, which have
been difficult for the conventional technique, can accurately be
performed with large resolving power.
[0132] In accordance with the result of the image field
discrimination, a character field is subjected to the edge
emphasizing process and a high resolution recording process. On the
other hand, the gradation field is subjected to the multi-gradation
process. Thus, the processes are selectively performed. Thus,
images of both of the character field and the gradation field can
satisfactorily be recorded.
[0133] When an image is encoded, the encoding method is switched in
accordance with the result of the image field discrimination. Thus,
image encoding can be performed without considerable distortion
occurring because of the encoding process such that a high
compression ratio is realized.
[0134] Additional advantages and modifications will readily occur
to those skilled in the art. Therefore, the invention in its
broader aspects is not limited to the specific details and
representative embodiments shown and described herein. Accordingly,
various modifications may be made without departing from the spirit
or scope of the general inventive concept as defined by the
appended claims and their equivalents.
* * * * *