 |
Description  |
|
|
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a digital copying apparatus. The present
invention more specifically relates to a digital copying apparatus capable
of specifying a binding position as a relative position relating to an
original image to prepare binding at said same relative position relating
to a copy image.
2. Description of the Related Art
When binding of copy sheets is desired, a binding position (direction) is
specified on an operation panel, and thereafter an original document is
placed at a predetermined position on a document platen whereupon a copy
is made. In such cases, the binding position (direction) is the relative
position (direction) relating to an original document sheet placed at a
predetermined position on the document platen, and is not a relative
position (direction) relating to the original document image.
Bound copy images are formed on a copy sheet in the following manner.
When specifying the binding position (direction) at the leading edge or
trailing edge of the copy sheet in the copy sheet feeding direction, a
non-image region for binding is formed at said leading edge portion or
trailing edge portion by adjusting the original document read timing and
sheet feed timing of the copy sheet. When specifying the binding position
(direction) at the top edge or bottom edge of the copy sheet in a
direction perpendicular to the copy sheet feeding direction, the
electrostatic latent image formed on said top edge portion or bottom edge
portion is exposed to light so as to erase said electrostatic latent image
and form a non-image region.
Conventionally, the specification of a binding position (direction) is
accomplished relative to an original document sheet placed at a
predetermined position on a document platen, and is not accomplished
relative to the original document image. Therefore, the placement
direction and placement position, or original document image direction
(top and bottom) may be mis-positioned whenever an original document is
placed on the document platen, such that binding of the copy sheet at a
desired position is not possible. In order to place the binding at a
desired position, it is necessary to exercise sufficient care in the
placement of the original document.
SUMMARY OF THE INVENTION
A main object of the present invention is to provide a copying apparatus
capable of forming a binding reliably at a desired position relative to
the copy image.
A further object of the present invention is to provide a copying apparatus
capable of forming a binding at a desired position without regard to the
original document placement position, or the top and bottom orientation of
said original document or placement position.
These and other objects of the present invention are accomplished by
providing a digital copying apparatus which reads an original image of an
original document placed on a document platen, generates digital image
data in accordance with said read image, and forms a copy image on a copy
sheet based on said image data, said digital copying apparatus comprising
a binding input device for inputting a binding position on a copy sheet as
a relative position in relation to said original image, a first detector
for detecting an orientation of the original document and an orientation
of the original image based on said image data, a second detector for
detecting an orientation of a copy sheet and a controller for controlling
formation of the copy image based on the orientations of the original
document and original document image detected by said first detector and
the orientation of the copy sheet detected by said second detector so as
to provide a binding at said relative position in relation to the copy
image.
These and other objects, advantages and features of the present invention
will become apparent from the following description thereof taken in
conjunction with the accompanying drawings which illustrate specific
embodiments of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a plan view showing the construction of the copying apparatus of
an embodiment of the present invention;
FIG. 2 is an illustration showing the operation panel in the application
mode;
FIG. 3 is an illustration showing the condition when the manual mode is
selected in the special purpose mode;
FIG. 4 is an illustration showing the condition when the auto mode is
selected in the special purpose mode;
FIG. 5 is a block diagram showing the construction of the control circuit
of the copying apparatus of FIG. 1;
FIG. 6 is a block diagram showing the image processing signal section of
FIG. 5;
FIG. 7 is a block diagram showing functions of the image processing section
of FIG. 6;
FIG. 8 is a block diagram showing the main scan movement of FIG. 7;
FIGS. 9a and 9b are illustrations of the main scan movement;
FIG. 10 is a flow chart showing the main routine of CPU 2;
FIG. 11 is a flow chart showing step S33 of FIG. 10;
FIG. 12 is a flow chart showing step S34 of FIG. 10;
FIG. 13 is a flow chart showing the main routine of CPU 5;
FIG. 14 is an illustration showing the original document orientation (size)
detection method;
FIG. 15 is an illustration showing the data used for original document
orientation (size) detection;
FIG. 16 is an original document management table created by CPU 1;
FIG. 17 is a block diagram of the memory unit section;
FIG. 18 is a memory map of the code memory of the memory unit section;
FIG. 19 is a management table for the code memory created by CPU 6;
FIGS. 20a, 20b, 20c, and 20d are illustrations of the conditions of
calculating the number of black image elements in the main scan direction
and sub-scan direction to determine the top and bottom of the image and
portrait or landscape orientation of the image via the process of CPU 6;
FIGS. 21a, 2lb, 21c, and 21d are illustrations of the conditions of
calculating the number of black image elements in a direction
perpendicular to the first line to extract each character contained in the
first line for use in a pattern check via the process of CPU 6;
FIG. 22 is a flow chart showing the image orientation detection process
performed by CPU 6;
FIG. 23 is an original document management table created by CPU 6;
FIG. 24 is an illustration showing the relationship between the rotations
of the original document image and the copy image;
FIG. 25 is a flow chart showing the main routine of CPU 1;
FIG. 26 is a flow chart showing step S16 of FIG. 25;
FIG. 27 is an illustration showing the image orientation and original
document orientation in manual mode and auto mode;
FIG. 28 is a flow chart showing step S115 of FIG. 26;
FIG. 29 is a flow chart showing step S117 of FIG. 26;
FIG. 30 is a flow chart showing step S121 of FIG. 26.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
The preferred embodiments of the present invention are described
hereinafter.
(1) Binding specification and preparation method summary
In the copying apparatus of the present embodiment, binding position
specification input is accomplished as shown in FIGS. 3 and 4.
FIG. 3 shows the manual mode, wherein the binding can be specified at a
position at the top or bottom and left or right sides of an original
document image in portrait or landscape orientation. In this case, when
the original document is disposed in portrait orientation the short edge
is aligned along the sheet feeding direction of the copy sheet, whereas
when the original document is disposed in landscape orientation the long
edge is disposed along the sheet feeding direction of the copy sheet. In
FIG. 3, a 10 mm binding is prepared at the left edge of the original
document image in portrait orientation, such that touch switch 2 is
specified.
FIG. 4 shows the auto mode, wherein the binding can be specified at a
position at the top or bottom and left or right sides of an original
document image regardless of the document orientation. In FIG. 4, a 15 mm
binding is prepared at the left edge of the original document image
regardless of document orientation, such that touch switch 2 is specified.
Binding specified for the left edge or right edge of an original image is
formed at the leading edge or trailing edge of the copy sheet in the copy
sheet transport direction by regulating the timing by which the original
document is read and the feed timing of the copy sheet via the timing
roller 82 (refer to FIG. 1).
When the binding is formed at the leading edge of the copy sheet, the copy
sheet is fed from the timing roller 82 by a length corresponding to the
binding width only whenever a copy sheet is on standby at the timing
roller 82 position. The timing by which the copy sheet is fed via the
timing roller 82 and the timing for reading the image data via the image
reader IR (refer to FIG. 1) are controlled in the same manner when binding
is not normally specified. Thus, a white strip for binding is formed at
the leading edge portion of the copy sheet in the copy sheet feed
direction.
When a binding is formed at the trailing edge of the copy sheet, the
standby position of the copy sheet at the position of the timing roller 82
is controlled in the same manner as normally when no binding is specified.
In relation to the timing for reading image data by the image reader IR
(refer to FIG. 1), the timing by which the copy sheet is fed by the timing
roller 82 is delayed only for a period corresponding to the binding width.
After completion of image data transmission, the white data corresponding
to the binding width is transmitted. Thus, image data can be transmitted
relatively quickly, to form the white area for binding at the trailing
edge of the copy sheet in the sheet transport direction.
Binding specified for the top edge or bottom edge of an original image is
formed at the top edge or bottom edge of the copy image by moving the
image data in the main scan direction.
When forming a binding at the top edge of a copy sheet, data specifying a
movement amount equivalent to the binding width and movement toward the
front side are transmitted from the CPU 2 of FIG. 6 relative to the main
scan movement block (refer to FIG. 7) of the image process section 211
(refer to FIG. 12).
The timing for reading the image data by the image reader IR (refer to FIG.
1) and the timing by which the copy sheet is transported by the timing
roller 82 are controlled in the same manner as normally when no binding is
specified.
The combinations of one-sided originals/duplex originals and one-sided
copies/duplex copies are described hereinafter in relation to binding.
In forming a binding at the left edge of a copy image in the case of
combined single-sided/duplex copies, said binding is formed on the
trailing edge side of the copy sheet relative to odd number pages of the
original, and on the leading edge side of the copy sheet relative to even
number pages of the original. Similarly, when forming a binding on the
left edge of a copy image in the case of duplex originals/duplex copies,
said binding is formed on the trailing edge side of the copy sheet
relative to a first surface of said original, and on the leading edge side
of the copy sheet relative to a second surface of said original.
When forming a binding on the top edge of a copy image in the case of one
sided/duplex copies, a binding is provided on the top edge side of the
copy image relative to odd number pages of the original, and on the bottom
edge side of the copy image relative to the even number pages of the
original. Similarly, when forming a binding on the top edge of a copy
image in the case of duplex originals/duplex copies, said binding is
formed on the top edge side of the copy sheet relative to a first surface
of said original, and on the bottom edge side of the copy sheet relative
to a second surface of said original.
(2) Outline of copy apparatus construction
FIG. 1 is a plan view in section showing the construction of an embodiment
of the copying apparatus of the present invention.
The copying apparatus in the drawing is provided with an image reader
section IR and a printer section PRT.
The image reader section IR is provided with a scanning unit 10 for
exposing via a lamp 11 an original document disposed face downward on a
document platen 18 and scanning said document in a sub-scan direction
during said exposure, Primary image sensor (CCD) 16 for photoelectrically
converting the light reflected by the original document and directed
thereto by the scanning unit 10 and creating electrical signals via said
photoelectric conversion, image signal processing section 20 for
processing the electrical signals output from said CCD 16 and creating
image data D2 based on said signals, and memory unit section 30 for
accommodating said image data D2 transmitted from the image signal
processor 20 and for rotation processing and the like.
The printer section PRT is provided with a print process section 40 which
modulates the output of the laser diode 62 in accordance with the image
data D3 transmitted from the memory unit section, laser optical unit 60
for scanning in an axial direction (main scan direction) the surface of a
photosensitive drum 71 via laser light emitted from said laser diode 62,
and an image forming section 70. The image forming section 70 is provided
with a member for developing via toner an electrostatic latent image
formed on the surface of the photosensitive drum 71 via laser light, a
member for transporting a copy sheet to the photosensitive drum 71 and
onto which is transferred the developed toner image,and a member for
fixing said toner image transferred onto the surface of said transfer
sheet.
The copy sheet size and directional orientation (i.e., portrait or
landscape directions) is detected by means of a sensor 81a and sensor 81b.
Although, in FIG. 1, the placement of an original document on the document
platen 18 is accomplished via a manual operation by an operator, it may
similarly be achieved by providing an automatic document feeder (ADF) or
the like.
Furthermore, although a device is shown for forming an image on first side
of a copy sheet only, a copy sheet refeeding unit (a unit used to recycle
a copy sheet which has had a copy image fixed on a first side thereof) may
be provided to allow formation of copy images on both sides of a copy
sheet.
(3) Control circuit outline
FIG. 5 shows the general construction of the control circuit of the present
apparatus.
The control circuit of the present apparatus provides six central
processing units (CPU 1-6)which are mutually connected for executing
processes. Each CPU has connected thereto a read only memory (ROM) for
storing various control programs, and random access memory (RAM) as a
working area. The control circuit of the present apparatus is further
provided with an image signal processor 20 for processing image signals
output from the CCD 16 and generating image data D2, and print process
section 40 for controlling the laser diode 61b in accordance with the
image data D3 read from the memory unit section.
The CPU 1 processes input signals from the various types of keys and touch
panel switches, and executes processes relating to the displays of the
operation panel shown in FIGS. 2-4. The CPU 1 further specifies the
binding direction. The processes executed by the CPU 1 are shown in the
flow charts of FIGS. 25, 26, and FIGS. 28-30 and are described later with
reference to FIG. 27.
The CPU 2 executes processes for setting parameters for the image signal
processor 20, specifications for the CPU 3 of the scanning section, and
detection of original document directional orientation and size. In the
image signal processor 20, various image processes are executed such as
movement and the like in the main scanning direction of the image. The
functions of the image signal processor 20 and processes of the CPU 2 are
described later with reference to FIGS. 6-12, and FIGS. 14 and 15.
The CPU 3 executes sub-scan drive control.
The CPU 4 executes sequence control of the print process section 40 and
image formation unit 70.
The CPU 5 executes processes for regulating the overall timing of the
control circuit of the present apparatus, and setting the operation modes.
Binding creation is executed relative to the leading edge or trailing edge
of the copy sheet as shown in FIG. 13. These processes are described in
detail later.
The CPU 6 controls the memory unit section. Detection of the image
directional orientation and rotation of the image are accomplished by the
memory unit section. Processes of the memory unit section and the CPU 6
are described fully later with reference to FIGS. 16-23.
(4) Construction and functions of the image signal processor 20
FIG. 6 shows the construction of the image signal processor 20. FIG. 6
shows the functions of the image process section 211. In the image signal
processor 20, each block of the process is executed in accordance with
image reading synchronization signals received from the timing control
section, and in accordance with the parameter settings set by the CPU 2.
Firstly, the CCD 16 reads the original document in single line units, and
generates original document read signals which are converted into digital
data by an analog-to-digital (A/D), which are thereafter transmitted to
the image process section 211.
In the image process section 211, the following processes are sequentially
executed: image quality corrections such as shading correction, MTF
correction, .gamma. correction, and electrical variation, main scan
movement, main scan reversal, density correction, filtering and the like.
After the aforesaid processing, the image data D2 are transmitted to the
memory unit section.
Shading correction corrects for non-uniform light quantity and
irregularities in reading element sensitivity in the main scan direction.
It is possible to switch between the mode for executing shading correction
and the mode for not executing shading correction by a shading ON/OFF
signal set via the CPU 2. During output regulation by the CCD 16, the
non-shading correction mode is set to allow throughput of the CCD 16
output. Reading data when a standard white pattern is read are written to
a shading RAM in accordance with shading timing signals.
The variable magnification (density conversion) of the image data in the
main scan direction is executed in the electric conversion block. This
magnification rate is set by the CPU 2.
The image data are shifted in the main scan direction in the main scan
movement block. That is, a shift corresponding to the binding width formed
on the top edge or bottom edge of the copy image, i.e. , the movement of
the movement mode, is executed. The amount and direction of the aforesaid
shift is set by the CPU 2. The process is described in detail later.
The image data are reversed in the main scan direction in the main scan
reversal block. When a mirror image of the original image read by a normal
scan is desired, or when a normal image or the original image read by a
reverse scan is desired, the reversal request signal is transmitted from
CPU 2 and reversal is accomplished in accordance therewith.
Background removal and density reproducibility correction are accomplished
in the density correction block. The amount of background removal and
amount of reproducibility correction (density gradient) are set by CPU 2.
The edge highlight process, smoothing process, and combinations thereof are
accomplished in the filtering block. Filter process selection and mix
ratio are set by CPU 2.
After the previously mentioned shading correction, data are transmitted to
the image monitor memory specified by CPU 2, and said image data are
stored in one-line segments. The directional orientation and size of the
original document are detected based on said one-line segment image data
stored in memory in a manner described later.
(5) Image data shift (movement in main scan direction)
Image data are shifted as follows in the main scan direction in the main
scan movement block of the previously mentioned image process section 211.
This shift of image data occurs when, for example, a binding is formed on
the top edge or bottom edge of a copy image.
Firstly, the image data from the electric variation block are stored in the
line memory 301a, as shown in FIG. 8. Address control at this time is
accomplished by the write address generation section. The write address
generation section generates address data by counting clock signals. Clock
signals are the transfer clock SNYNCK of input image signals.
Each time the horizontal synchronization signal HSYNC is input, the
relationship is switched between the line memory 301a and line memory
301b, and the write address generation section and read address generation
section. That is, when the image data of a certain line have been stored
in the line memory and thereafter the next data of the next line are
input, the line memory in which is stored the image data of said certain
line are read out by the specification of the read address generation
section. This read address generation section also generates address data
by means of counting the transfer clock SYNCK signal.
In relation to the read address generation section, a read start position
signal FST.sub.-- POS signal is transmitted from the CPU 2 during the
binding shift (image shift). That is, when a binding position is specified
at the top edge section of a copy image (step S601: YES), the "foreground"
predetermined position is set as FST.sub.-- POS (step S603), as shown in
FIG. 19. When the bottom edge is specified (step S611: YES), the
"interior" predetermined position is set (step S613). The binding position
is set in accordance with the input from the operation panel; this is
described in more detail later with reference to FIG. 30.
The read start position signal FST.sub.-- POS is transmitted from the CPU 2
to the read address generation section to accomplish the image data shift.
For example, the read start position from the line memory 301a (30lb) is
normally address [0] of the line memory. When the aforesaid read start
position signal is transmitted, the readout starts from the set position
As (.noteq.0; refer to FIG. 9a), thereby shifting the image "A" (refer to
FIG. 9b). Although FIGS. 9a and 9b show the image shift leftward (As>0), a
rightward image shift can be realized when the setting is As<0.
(6) Original document directional orientation and size detection
The image signal process section 20 is controlled by the previously
mentioned CPU 2. Original document directional orientation and size
detection is accomplished by the CPU 2 in the manner described below.
Firstly, when a command for detect original document directional
orientation is issued from CPU 1 and said detection is executed by CPU 2
(Step S501: YES, FIG. 11), a command to execute a pre-scan is transmitted
from CPU 2 to CPU 3 (Step S503), and the directional orientation of the
original document is detected based on the signals derived from said
pre-scan (step S505).
The image data read by the aforesaid scan and corrected for shading were
periodically stored in line units in the memory used for image monitoring,
and thereafter said image data are read out, and the read image data are
scanned from the standard position side in the main scan direction. That
is, a scan is executed in the direction X shown in FIG. 14. FIG. 14 shows
the principle for detecting original document directional orientation and
size. The document platen standard position 0 is stored beforehand by the
CPU 2.
In the scan in the X direction, the address (FST.sub.-- WHT) of the first
"white" level detected in the X direction and the address (LST.sub.-- WHT)
of the last "white" level detected in the X direction are stored in
buffers corresponding to the sub-scan position (=line position, i.e., the
position in the Y direction from the document platen standard position 0).
A value [0] is stored at the lines where a white level is not detected
(refer to FIG. 15).
When the buffer data are stored, the size and position of the original
document are calculated based on said buffer data.
That is, the data stored in the buffer are sequentially read, and the
sub-scan position (address in the Y direction) k at which the first data
other than [0] is stored is determined to be the leading edge position of
the original document in the sub-scan direction (Y direction). The
sub-scan position (address in the Y direction) n at which the last data
other than [0] is stored is determined to be the trailing edge position of
the original document in the sub-scan direction. Thus, the dimensions of
the original document in the sub-scan direction are determined as [n-k].
The minimum value of FST.sub.-- WHT, i.e. , the address of the first
"white" level detected in the X direction for each line, is calculated,
and said minimum value FST.sub.-- WHT1 is determined to be the leading
edge position of the original document in the main scan direction (X
direction). The maximum value of LST.sub.-- WHT, i.e., the address of the
last "white" level detected in the X direction for each line, is
calculated, and said maximum value LST.sub.-- WHTm is determined to be the
trailing edge position of the original document in the main scan direction
(X direction). Thus, the dimensions of the original document in the main
scan direction are determined as [LST.sub.-- WHTm.sub.-- FST.sub.-- WHT1].
When the leading edge position/trailing edge position/dimensions of the
original document in the sub-scan direction and the leading edge
position/trailing edge position/dimensions of the original document in the
main scan direction are determined, the portrait or landscape orientation
of the original document is identifiable based on said determinations.
Furthermore, a standard size close to the aforesaid dimensions is
searched, and said standard paper size is transmitted to CPU 1 as the
original document size.
In CPU 1, a document management table shown in FIG. 16 is generated based
on the aforesaid derived document size data. That is, for each original
document a table is generated which shows the size and portrait/landscape
orientation distinction of the original document.
In the previously described process, the method for identifying the
document region and non-document region from the read image may be (a) a
method wherein the document cover is provided with a mirror surface, and
the original document area is determined to be within a range of "white"
detected by the image scan, or (b) a method wherein a scan is performed
with the document cover open, and the original document area is similarly
determined to be within a range of "white" detected by the image scan.
The CPU 2 also executes other processes in addition to those described
above, such as a process for specifying a scan direction for a reverse
read out (step S35), and a process for specifying scanning via a main scan
reversal (step S36), as shown in FIG. 10.
(7) Memory unit section
As shown in FIG. 17, the memory unit section comprises a switch section,
binarizer, image memory, compressor, code memory, expander, rotation
section, multi-level converter, and variable magnification section. These
sections are controlled parameters set by the CPU 6.
The binarizer converts the image data D2 transmitted from the image signal
processor 20 into binary data. The image memory is provided with multiple
ports and has an A4-size two-page capacity at 400 dots per inch (dpi). The
compressor and expander of the encode section are capable of mutually
independent parallel operation. The code memory is provided with multiple
ports. The image rotation process is executed in the rotation section.
Binary data are converted to multi-level data in the multi-level
converter. The image is subjected to electrical variable magnification in
the variable magnification section.
When the binarizer writes the image data to the image memory, the
compressor reads and compresses said image data to generate encoded data
which are written to the code memory. The expander reads the data written
in the code memory in accordance with instructions from CPU 6, and expands
said data as image data which are written to the image memory. The data
are transmitted via direct memory access (DMA).
When the expander writes one page of image data to the image memory, CPU 6
reads the image data from the image memory, and discriminates the top and
bottom of the image as well as its landscape or portrait orientation.
Landscape/portrait orientation is determined based on the distribution of
black image elements in the main scan direction and sub-scan direction.
The detection of image directional orientation is described later.
Thereafter, the image data are supplied to the rotation section.
The aforesaid image data are rotated by the rotation section as necessary,
and converted to multi-level data by the multi-level converter,
magnification is varied as necessary by the variable magnification
section, and said image data D3 are transmitted to the print process
section 40 (refer to FIG. 5).
FIG. 18 shows the code memory, and FIG. 19 shows the management table for
managing said code memory. The code memory is divided into memory areas in
units of 32 kilobytes (kB). Code data of the same page contents are stored
in the aforesaid memory areas to allow simultaneous control during writing
time (image reading time) and reading time (image printing time). Stored
in the aforesaid management table are number expressing the memory area of
the code memory, page numbers, number of linked memory areas, and various
additional information required for the compression method, and the data
compression and expansion processes. The code memory is dynamically
managed based on the aforesaid information.
In the management table of FIG. 19, a preconcatenation is linked in the
forward direction to the memory area to each 32 kB within the same page.
That is, if the value is [00], it expresses the first memory area of the
page, whereas any other value expresses a number of the memory area linked
in a forward direction. A post-concatenation is the same. If the value is
[FF], it expresses the last memory area of the page, whereas any other
value expresses the number of the memory area linked in a backward
direction.
The management table of FIG. 19 is created by the CPU 6 whenever image data
are read from the image memory, compressed by the compressor, and stored
in the code memory. The information of the management table is erased when
the number of the necessary sections are all normally discharged.
In the memory mode, the image data D2 are binarized and thereafter fetched
from the image memory and subjected to predetermined processing, and are
read from the image memory and output as image data D3 to the print
process section 40 as required. When reading image data from the image
memory, the rotation and variable magnification processes are executed in
accordance with instructions from the CPU 6 .
(8) Discrimination of image top and bottom and landscape/portrait
orientations
As previously described, when the expander writes one page of image data to
the image memory, the CPU 6 reads said image data and discriminates the
top and bottom as well as the landscape/portrait orientation of said
image, as shown in FIGS. 20-22.
As shown in FIG. 20a, the number of black image elements are counted for
each line in both the main scan and sub-scan directions (step S71). The
counting results in the main scan direction are shown for each image on
the left side, and counting results in the sub-scan direction are shown
for each image on the right side in FIGS. 20a-20d. Based on the
distribution of the counted black image elements, the landscape or
portrait orientation of the image is discriminated (step S72). That is,
when a band-like peak appears in the distribution of black image elements
in the main scan direction, a portrait image orientation is discriminated
(as in the cases of FIGS. 20a and 20b), whereas when a band-like peak
appears in the distribution of black image elements in the sub-scan
direction, a landscape image orientation is discriminated (as in the cases
of FIGS. 20c and 20d).
Then, the image data corresponding to the first line is extracted. The
first line is the region of the initial appearance of the band-like peak,
as indicated by the shaded area of oblique lines in FIGS. 21a-21d. After
the aforesaid extraction, the number of black image elements of each line
in a direction perpendicular to the aforesaid first line are counted
within the region of said first line. For example, with regard to the
band-like region which includes [ABCDEF], as shown in FIG. 21a, the number
of black image elements of each line is counted in the sub-scan direction,
i.e., a direction perpendicular to the band-like region. With regard to
the band-like region which includes [MNOPQR], as shown in FIG. 21d, the
number of black image elements of each line is counted in the main scan
direction, i.e., a direction perpendicular to the band-like region (step
S73). The counting results are shown on the bottom side of the image for
portrait orientations, and on the left side for landscape orientations.
The characters within the first line are extracted based on each peak (the
peaks corresponding to each character within the first line) appearing in
the counting results (step S74).
Next, the pattern of each extracted character is matched to standard
character patterns. The angle of the standard character patterns is
initially 0.degree., i.e., an normal upright state. After pattern
matching, the number of discriminatable characters (i.e., the number of
characters whose pattern matches a standard character pattern) is
determined, and stored in association with the standard character pattern
angle (initially 0.degree.) (step S75).
Subsequently, the angle of the standard character pattern is rotated
+90.degree.. Since the initial standard character angle was 0.degree., and
angle of 90.degree. is set. Thereafter, the process of step S75 is
executed, and similar pattern matching is executed using the standard
character pattern set at 90.degree.. After the aforesaid pattern matching,
the number of discriminatable characters is determined, and stored in
association with the standard character pattern angle (initially 90
.degree.) (step S75).
Thereafter, the standard character pattern angle is set at 180.degree. and
270.degree., and the same process is executed (steps S75, S76).
When the pattern matching is completed for each standard character pattern
at angles of 0.degree., 90.degree., 180.degree., and 270.degree. (step
S77: YES), the standard character pattern angle at which the maximum
number of discriminatable characters was found is determined, and that
angle is set as the image orientation direction (step S78).
Thus, the top and bottom as well as landscape and portrait orientations of
the image can be discriminated. The original document management table of
FIG. 23 is generated based on the aforesaid results.
(9) Readout from the image memory
When the CPU 5 determines | | |
