 |
Description  |
|
|
BACKGROUND OF THE INVENTION
The present invention relates to an apparatus for erasing and extracting
image data from a particular region of an original document, in which the
particular region is specified by an exclusive marking sheet, by reading
the sheet prior to the document, storing data associated with the sheet in
a memory, and processing an image carried on the document based on the
particular region which is stored in the memory. More particularly, the
present invention is concerned with such an apparatus which is applicable
to a digital copier, a digital printer, a printer using a mimeograph and
the like.
An apparatus of the kind described is disclosed in, for example, Japanese
Laid-Open Patent Publication (Kokai) No. 61-13867/1986. The apparatus
there disclosed reads an original document and an exclusive marking sheet
at the same density and stores the resulting data in a memory. For
example, assuming that the density at which a document is read is sixteen
dots per millimeter, data associated with the marking sheet of format A4
amounts to 15.9 megabits, i.e., 16.times.16.times.210.times.297=15.9
megabits, resulting in the need for an extremely large memory capacity. To
reduce the memory capacity required, this prior art apparatus compresses
data before storing the data memory and decodes the data after outputting
the data from the memory. Such a scheme, however, has various drawbacks in
that compressing and decoding data cannot be implemented without resorting
to complicated and expensive circuits. Further, since the data are
compressed, considerable difficulty is experienced in identifying a
particular region specified by a frame in distinction from the rest of the
data identifying region located outside of the particular region. Further,
the processing applied to data which have been decoded is time-consuming
because the amount of data is necessarily increased.
On the other hand, the content of processing may be varied from one kind of
document to another, as has been practiced with some prior art digital
copiers of the type using such a marking sheet. For example, a different
kind of image processing may be applied to each of a text document and a
photograph or like graphic document, which includes halftone, for the
purpose of enhancing the reproducibility of an output image. Hereinafter,
the processing applied to a text document will be referred to as a text
mode, and the processing applied to a graphic document as a graphic mode.
While in the text mode, data having undergone MTF (Modulation Transfer
Function) correction are transformed into two-level data, i.e.,
black-and-white data by discriminating such data with respect to a
predetermined threshold value, whereas in the graphic mode data are
transformed by reproducing halftone based on a dither method or the like.
However, should the processing associated with the text mode be directly
applied to the graphic mode or vice versa, the reproducibility would be
lowered to a critical degree. For example, in the case that a solid black
image or characters are reproduced by processing them by the dither method
which is associated with the graphic mode, the resulting image becomes
mere fragments which are discontinuous due to local omission of data. A
person, therefore, has to select either one of the text and graphic modes
depending upon the kind of a document to be used.
Since an exclusive marking sheet previously described is similar in
condition to a text document, it naturally has to be processed in the text
mode. However, there is a fear that the sheet is inadvertently processed
in the graphic mode, bringing about the above-described occurrence which
would lead to erroneous recognition of a marked region of a document.
SUMMARY OF THE INVENTION
It is, therefore, an object of the present invention to provide an
apparatus for erasing and extracting a particular region of a document
which needs a minimum of memory capacity.
It is another object of the present invention to provide an apparatus for
erasing and extracting a particular region of a document which is capable
of rapidly identifying the particular region in distinction from the other
region.
It is another object of the present invention to provide an apparatus for
erasing and extracting a particular region of a document which
automatically applies image processing in the text mode to an exclusive
marking sheet.
It is another object of the present invention to provide an apparatus for
erasing and extracting a particular region of a document which is
applicable to a digital copier, a digital printer, a printer using a
mimeograph and the like and selectively sets up the text and graphic modes
in matching relation to the kind of a document.
It is another object of the present invention to provide a generally
improved apparatus for erasing and extracting a particular region of a
document.
These and other objects are achieved in accordance with the present
invention by providing a novel apparatus for reading a document image
printed on an original document and an image of a marking sheet which
marks a desired region of the document image together with a mark provided
on the marking sheet, storing data associated with the marking sheet in a
memory, and processing the document image, which is typically read after
the marking sheet, in a predetermined editing mode, including a binarizer
for reading the marking sheet and document image to convert the sheet and
document image into two-level pixel data, a density conversion circuit for
converting density of the two-level image data associated with an image of
the marking sheet and controlling the memory such that the pixel data are
written in the memory, an inside/outside decision circuit for
discriminating the pixel data stored in the memory with respect to whether
or not the pixel data relate to locations on the document image inside or
outside of the marked region and applying processing to the pixel data
inside and outside of the marked region based on the predetermined editing
mode, and an image processor for reading the pixel data associated with
the marking sheet and stored in the memory and those associated with the
document and outputted by the binarizer while synchronizing the pixel data
to each other, and applying image processing to the document in the
predetermined editing mode.
Further, in accordance with the present invention, there is provided a
novel apparatus for reading a document image printed on a document and
which is read by a scanner and an image of a marking sheet which marks a
desired region of the document image, storing data associated with the
marked region in a memory, and applying image processing to the document
image, which is typically read after the marking sheet, in a predetermined
editing mode and a predetermined image mode, including an AD
(analog-to-digital) converter for reading the image of the marking sheet
and the document image and converting resulting image data into digital
image data, an image signal processor for processing the digital image
data in the predetermined image mode to transform the digital image data
into two-level pixel data and writing the pixel data in the memory, a
sequence controller for determining if it is the document image or the
image of the marking sheet that has been read by the scanner and
commanding the image signal processor a selected image mode, and an
editing circuit supplied with the pixel data associated with the image of
the marking sheet and read out of the memory and the pixel data associated
with the document image and outputted by the image signal processor for
applying image processing to the document image in the predetermined
editing mode.
BRIEF DESCRIPTION OF THE DRAWINGS
A more complete appreciation of the invention and many of the attendant
advantages thereof will be readily obtained as the same become better
understood by reference to the following detailed description when
considered in connection with the accompanying drawings, wherein:
FIG. 1 is a schematic illustration showing processing for erasing and
extracting a particular region of a document as marked by an marking
sheet;
FIG. 2 is a fragmentary view of an operation board of an apparatus which is
representative of a first embodiment of the apparatus in accordance with
the present invention;
FIG. 3 is a schematic block diagram showing the overall construction of the
apparatus of FIG. 2;
FIG. 4 is a schematic illustration showing two-level pixel data which are
stored in a memory;
FIGS. 5 and 6 are schematic illustrations representative of decision
processing applied to pixel data which are stored in the memory;
FIG. 7 is a schematic block diagram showing a specific construction of a
density conversion timing circuit and that of a memory in accordance with
the first embodiment;
FIG. 8 is a timing chart showing signals which appear in various portions
of the circuitry shown in FIG. 7;
FIG. 9 is a schematic block diagram showing a specific construction of an
inside/outside decision circuit;
FIGS. 10A to 10C are flowcharts each demonstrating a particular timing of
the inside/outside decision circuit during execution of its processing;
FIG. 11 is a schematic block diagram showing another specific construction
for the decision processing;
FIG. 12 is a schematic diagram showing a specific construction of an image
processing circuit;
FIG. 13 is a schematic block diagram showing the overall construction of a
second embodiment of the present invention;
FIG. 14 is a fragmentary view of an operating section which is included in
the embodiment of FIG. 13;
FIG. 15 is a flowchart demonstrating the operation of the second
embodiment;
FIG. 16 is a schematic block diagram showing a specific construction of an
image signal processing circuit of the second embodiment;
FIG. 17 is a schematic block diagram showing a specific construction of a
sequence control circuit;
FIG. 18 is a schematic block diagram showing the combined construction of a
memory control circuit and memory;
FIG. 19 is a timing chart showing various signals which appear in a memory
control circuit of FIG. 18;
FIG. 20 is a schematic block diagram showing an editing circuit;
FIG. 21 is a perspective view of a printer using a mimeograph to which the
present invention pertains; and
FIG. 22 is a vertical section showing a detailed construction of the
printer shown in FIG. 21.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
Referring now to the drawings, wherein like reference numerals designate
identical or corresponding parts throughout the several views, two
different embodiments of the apparatus in accordance with the present
invention will next be described in detail.
FIRST EMBODIMENT
The first embodiment which will be explained promotes rapid processing for
the discrimination between the inside and the outside of a particular
region marked and reduces memory capacity required.
FIG. 1 schematically shows how an image which lies in a certain region of a
document marked may be entirely erased or omitted. First, a person lays on
a document 12 a transparent or semitransparent marking sheet 10 which is
adapted to specify a desired region of the document 12 to be erased. Then,
the operator writes with a felt pen or like implement 16 a solid line 14
which marks the particular region in distinction from the rest, i.e.,
character "E". Subsequently, the operator operates an erase key 20 of an
operation board 18, FIG. 2, for setting up an erase mode which erases the
whole image lying inside of the line 14. A light emitting diode (LED) 22
is associated with the erase key 20 and turns on when the latter is
operated. When the operator desires to extract only the image which lies
in the region marked by the sheet 10, the operator may depress an extract
key 24 of the operation board 18 for selecting an extract mode. An LED 26
is associated with the extract key 24. Also provided on the operation
board 18 is a start key 28. In the above condition, an image scanner as
will described later is activated to read the sheet 10 and the document 12
in this order, and the resulting data are processed independently of each
other. Finally, an image printer is driven to produce an image 12a in
which the marked region, i.e., character "E" is omitted.
Referring to FIG. 3, an apparatus in accordance with this particular
embodiment is shown in a block diagram and generally designated by the
reference numeral 30. The marking sheet 10 and the document 12 are read by
an image scanner 32 in this order. The image scanner 32 is made up of a
light source 34, a mirror 36 and a CCD (charge coupled device) 40. The
marking sheet 10 is illuminated by light which issues from the light
source 34 and, therefore, read together with the line 14 by the CCD 40 by
way of the mirror 36 and lens 38. The output of the CCD 40 is fed to a
binarizer 42 to be thereby converted into two-level pixel data having
density of 16 dots per millimeter in both the main and subscanning
directions, i.e. pixel data of sixteen dots per millimeter. The sixteen
dots per millimeter, two-level pixel data are applied to a memory 44 and
an image processing circuit, or image processor, 46. A controller 48 is
adapted to control the entire apparatus 30 and is constituted by a CPU
(central processing unit), a ROM (read only memory), a RAM (random access
memory), an I/O (input/output) interface and others. A density conversion
timing circuit 50 delivers read/write commands and address signals to the
memory 44 in response to instructions which are generated by the
controller 48. The two-level data sequentially fed from the binarizer 42
to the memory 44 are thinned at a rate of seven out of eight, thereby
transforming the pixel data to a density which is as low as two dots per
millimeter.
In FIG. 4, there is shown how the two-level image data are written in the
memory 44. An area A indicated by hatching in the figure is representative
of a size in which the line 14 which would be written if the sixteen dots
per millimeter, two-level pixel data from the binarizer 42 were not
thinned. On the other hand, an area B is representative a size in which
the line 14 is produced when the sixteen dots per millimeter pixel data
are thinned at the rate of seven out of eight into two dots permillimeter
pixel data, as previously stated. Specifically, the density conversion
timing circuit 50 extracts only one pixel out of each eight pixels of the
hatched area A in both the column and row address directions. The pixel
data extracted so, i.e., two dots per millimeter pixel data are
represented by solid dots b in FIG. 4. These pixel data b only are written
in the memory 44 in a small configuration, as indicated by the area B.
This reduces the memory capacity required to 1/64, compared to the sixteen
dots per millimeter pixel data without thinning. This is because the
thinning processing is effected in both the main and subscanning
directions, i.e. 1/8.times.1/8=1/64. Since the pixel data undergone
thinning are not coded at all, the shape of data written in the memory 44
as represented by B appears analogous to that of original sixteen dots per
millimeter pixel data as represented by A, meaning a reduction of the
image to 1/8.
Referring to FIG. 3 again, an inside/outside decision circuit 54 applies
decision processing to the pixel data which are written in the memory 44
as stated above. By decision processing is meant filling the entire region
which is marked by the line 14 with the same data as that of the line 14,
i.e. logical ONE. More specifically, as shown in FIG. 5, in the raster
data constituted by the pixel data of line 14 as represented by ONEs and
those of the region inside of the line 14 as represented by ZEROs, ZEROs
are changed to ONEs by the decision processing. For such decision
processing, among those pixels of a surrounding pixel matrix shown in FIG.
6, pixels A, B, C, D, F, G, H and I which surround a particular pixel E to
be processed are used. For example, the particular pixel represented by E
is discriminated by using an equation E=E+(C+F).times.(D+G), in which the
symbols "+" and ".times." are representative of ORing and ANDing,
respectively. Details of the decision processing will be described in
detail later.
After the marking sheet 10, the document 12 is read by the image scanner
32. The output of the image scanner 32, i.e., that of CCD 40 is fed to the
binarizer 42 to be thereby converted into sixteen dots per millimeter
pixel data. Simultaneously, in response to a command from the density
conversion timing signal 50, the two dots per millimeter data previously
written in the memory 44 and representative of the line 14 and the region
inside of the line 14 are read out timed such that they coincide with the
sixteen dots per millimeter pixel data which are representative of the
document 12. Specifically, the same data are repeatedly read out eight
consecutive times in the main scanning direction and, likewise, the same
data are read eight consecutive times in the subscanning direction. The
data associated with the sheet 10 as read out of the memory 44 and the
data associated with the document 12 as fed from the binarizer 42 are
routed to the image processor 46. The image processor 46 erases that
region of the document 12 which is defined by the line 14 that is written
on the sheet 10. The resulting image data are delivered to a laser printer
or like image printer 52 to be printed out thereby. The resulting document
image 12a, therefore, is void of the image region which is marked by the
line 14, i.e. character E.
Hereinafter will be described specific constructions of the memory 44,
density conversion timing circuit 50, inside/outside decision circuit 54
and image processor 46. It is to be noted that the other structural
elements can be implemented with those which per se are known in the art
and, therefore, will not be described for the sake of simplicity.
FIG. 7 shows a combined specific construction of the density conversion
timing circuit 50 and memory 44 while FIG. 8 shows various signals in a
timing chart. As shown, the memory 44 is implemented with a so-called bit
map memory in which pixels are associated with addresses in one-to-one
correspondence. To write data associated with the marking sheet 10 of at
least format B4 at the density of two dots per millimeter, the memory 44
is sized 256 (millimeters).times.2 (dots per millimeter)=512 (dots) in the
column direction and 350.times.2=700 (1024 (dots) in the row direction,
i.e. 512 (columns).times.1024 (rows). The density conversion timing
circuit 50 is made up of four D flip-flops 502, 504, 506 and 508 which in
combination constitute a first 1/8 frequency division network, four D
flip-flops 510, 512, 514 and 516 constituting a second 1/8 frequency
division network, a 9-bit column address counter 518, a 10-bit row address
counter 520, and gates 522, 524, 526 and 528. Two-level data DATA from the
binarizer 42 are fed to the memory 44 clocked by synchronizing signals CK
and LCK which are generated by the CCD 40. While the synchronizing signal
CK is outputted by the CCD 40 for each pixel in the main scanning
direction, the synchronizing signal LCK is outputted by the same on a
line-by-line basis with respect to the subscanning direction. A clear
signal CL and a write signal WR are fed from the controller 48.
First, the operation for writing pixel data in the memory 44 will be
described. The synchronizing signal CK is applied to the four D flip-flops
502, 504, 506 and 508 so as to cause the flip-flops 502 and 504 to produce
signals Q.sub.1 and Q.sub.2, respectively. These signals Q.sub.1 and
Q.sub.2 are fed to the gate 522 which then ANDs these signals, i.e.
Q.sub.1 .times.Q.sub.2. As shown in FIG. 8, the signal Q.sub.1
.times.Q.sub.2 remains high level only for the duration of one pixel out
of eight consecutive pixels in the main scanning direction. At each
positive-going edge of the signal Q.sub.1 .times.Q.sub.2, the 9-bit column
address counter 518 which is generating a column address is incremented
(see wave-forms of signals A.sub.0 to A.sub.8 shown in FIG. 8). The column
address counter 518 and four D flip-flops 502, 504, 506 and 508 are
cleared by the synchronizing signal LCK so as to set up synchronization in
the main scanning direction, the column address in the main scanning
direction being incremented once for each eight pixels. The synchronizing
signal LCK, like the signal CK, is fed to the four D flip-flops 510, 512,
514 and 516 to cause the flip-flops 510 and 512 to produce signals Q.sub.5
and Q.sub.6, respectively. The signals Q.sub.5 and Q.sub.6 are ANDded by
the gate 524 to become a signal Q.sub.5 .times.Q.sub.6. This output
Q.sub.5 .times.Q.sub.6 of the gate 524 remains high level only for the
duration of one line out of eight consecutive lines in the subscanning
direction. At each positive-going edge of the signal Q.sub.5
.times.Q.sub.6, the 10-bit row address counter 520 which is generating a
row address is incremented. Cleared by the clear signal CL, the
row-address counter 520 and four D flip-flops 510, 512, 514 and 516 set up
synchronization in the subscanning direction, the row address being
incremented once for each eight lines. The output Q.sub.1 .times.Q.sub.2
of the gate 522 and that Q.sub.5 .times.Q.sub.6 of the gate 524 are ANDded
by the gate 526, and the resulting AND is fed to the gate 528. Also
applied to the gate 528 is the write signal WR which is outputted by the
controller 48. The output WR of the gate 528 is delivered to the memory
44. When the signal WR has a low level, the memory 44 stores two-level
pixel data DATA applied to its terminal DATA IN in a particular memory
cell which is being accessed. When the signal WR has a high level, the
memory 44 outputs data from a particular memory cell being accessed via
its terminal DATA OUT. In this manner, since the signal WR fed to the
memory 44 causes the sigal Q.sub.1 .times.Q.sub.2 which appears once for
each eight pixels and the signal Q.sub.5 .times.Q.sub.6 which appears once
for each eight lines to be ANDded, data are produced at a rate of one
pixel per sixty-four (=8.times.8) pixels and stored in the memory 44. The
other data are not stored in the memory 44. Stated another way, the other
data are thinned. Needless to mention, it is only when the controller 48
generates the write signal WR that data are written in the memory 44.
Image data stored in the memory 44 are read thereoutof in the same order as
the document 12 is read by the image scanner 32 and written in the memory
44. Specifically, since data can be always read so long as the output WR
of the gate 528 has a high level, all that is required is sequentially
incrementing the address. The column address counter 518 generating a
column address is not changed throughout eight pixels of image data which
are read out of the document 12, so that data are read out of the memory
44 with the address unchanged. This is true with the row address also,
i.e., the row address is unchanged throughout eight consecutive rows.
Referring to FIG. 9, a specific construction of the inside/outside decision
circuit 54 is shown. As shown, the decision circuit 54 is constituted by D
flip-flops 542, 544, 546, 548, 550, 552, 554, 556 and 558 which are
arranged in association with the surrounding pixel matrix of FIG. 6 in
order to latch the surrounding pixels, FIFO (first-in-first-out) memories
560 and 562 adapted to store data of the immediately preceding line and
those of the line occurred two lines before, and gates 564, 566, 568, and
570. The image data DATA are fed from the memory 44 to the flip-flop 558
and subjected to the processing which fills the line 14 and the whole
region defined by the line 14, FIG. 5, with ONEs, the resulting data being
delivered via the gate 570 to the memory 44. For details of the procedure
for replacing ZEROs with ONEs in a particular area of an image memory as
mentioned, a reference may be made to Japanese Laid-Open Patent
Publication (Kokai) No. 62-58508/1987.
The decision performed by the inside/outside decision circuit 54 as
discussed above may be programmed in any of three different flows which
are shown in FIGS. 10A to 10C. In FIG. 10A, the decision is effected at
the same time as the image data associated with the marking sheet 10 are
read out, i.e., as the two-level pixel data from the binarizer 42 are
thinned by the density conversion timing circuit 50 to the density of two
dots per millimeter, the result being written in the memory 44. In FIG.
10B, the thinned two dots per millimeter pixel data associated with the
sheet 10 are directly written in the memory 44 and discriminated later
when they are read out of the memory 44. In FIG. 10C, the thinned two dots
per millimeter pixel data which are directly written in the memory are
read out by raster-scanning the memory 44 when the data associated with
the document 12 are read by the image scanner 32 and, then, discriminated,
the result being written in the memory 44 again.
Alternatively, as shown in FIG. 11, the decision processing stated above
may be executed by a CPU 56 which accesses the memory 44. In this case,
the CPU 56 will read a plurality of times those pixels which surround a
pixel to be discriminated in the memory 44, store data of the surrounding
pixels in a register which is built in the CPU 56, perform computation
with those data, and writes the result in a particular address of the
memory 44. This corresponds to the flow of FIG. 10C.
Referring to FIG. 12, a specific construction of the image processor 46 is
shown. As shown, the image processor 46 is comprised of two gates 462 and
464. The gate 462 is supplied with the data ONEs representative of the
line 14 and the inside of the line 14, as shown in FIG. 5, from the memory
44 and an image data extract/erase signal from the controller 48,
delivering an output thereof to the gate 464. The gate 464, on the other
hand, is supplied with the two-level pixel data from the binarizer 42
while feeding its output to the image printer 52. When the extract/erase
signal from the controller 48 has a low level, the output of the memory 44
is directly routed to the gate 464. Since the whole region delimited by
the line 14 is a ONE, only those image data in the region of the document
which is located outside of the region stored in the register with the
marked region region are effective. Consequently, the region other than
the marked region becomes all ZERO, i.e. blank. When the extract/erase
signal from the controller 48 is high level, the region outside the marked
region is all ONE so that only the image data outside of the marked region
are effective, the marked region being all ZERO and, therefore, blank.
The operation board 18 shown in FIG. 2 is constructed as follows. When the
start key 28 on the operation board 18 is depressed under a usual
condition in which neither the erase key 20 nor the extract key 24 is
depressed, the document 12 is read by the image scanner 32 and, after a
series of processing stated earlier, it is reproduced by the image printer
52. When either the erase key 20 or the extract key 24, especially the
latter as illustrated, is depressed and, then, the start key 28, data
associated with the marking sheet 10 are written in the memory 44. In such
a case, as the start key 28 is depressed again, the document 12 which
follows the sheet 10 is subjected to the previously stated image
processing resulting that the image 12a to be produced by the image
printer 52 is void of the image as surrounded by the line 14, i.e.
character E.
In the illustrative embodiment, extra processing may be added for
preventing data which discriminate the desired region and the rest from
each other from being omitted. For example, there may be added processing
which causes even a single pixel out of eight pixels to be written in the
memory 44 if it is a ONE. Further, the equation for determining the pixel
E as previously stated may be modified as
E=[E+(C+F).times.(D+G)].times.(A+B+C+D+F+G+H+I) to add anti-noise
processing. Specifically, by adding ".times.(A+B+C+D+F+G+H+I)" to the
previously stated equation, it is possible to make E a ZERO even if it is
a ONE, when all the surrounding pixels are a ZERO.
As described earlier, the two-level pixel data are reduced in density from
the sixteen dots per millimeter to two dots per millimeter before being
written in the memory 44. Such a decrease in the amount of data
contributes not only to a decrease in memory capacity required but also to
rapid processing of the inside/outside decision circuit 54. Reading the
document 12 at the density of sixteen dots per millimeter is satisfactory
with regard to the quality of image reproduction and becoming popular as a
standard. In this regard, even if the marking sheet 10 is read at the
density of two dots per millimeter, it is needless to replace it with the
usual density of sixteen dots per millimeter because the specified region
is merely used to erase or extract an image which lies in that region and
not reproduced as it is. Further, the density of two dots per millimeter
suffices in practice considering the fact that the line 14 on the sheet 10
is written by hand by using the felt pen 16 or the like. A good
experimental result was achieved by using a felt pen which was about 1.0
to 1.5 millimeters in line width.
| | |
