JP2017017424A - Image processing apparatus, image reading apparatus, and image forming apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing apparatus, an image reading apparatus, and an image forming apparatus which detect an image formed by an image forming mode in which a developer is reduced without regard to the type and color of an image.SOLUTION: This invention comprises: an image processing apparatus, an image reading apparatus comprising the image processing apparatus, and an image processing apparatus comprising the image reading apparatus. The image processing apparatus comprises: detection means which detects whether a document image is subjected to tonner saving; and correction means which performs correction processing on an image where toner saving is detected by the detection means.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an image processing apparatus, an image reading apparatus, and an image forming apparatus. For example, the present invention is applied to an apparatus that reads and corrects a document printed in a toner save mode (printed in a print mode with a toner amount reduced than usual). Can do.

  An image processing device that supports the conventional image reading function (scanner function) detects a black area printed in the toner save mode in which the amount of toner (the amount of developer) is reduced than usual when reading a document. Thus, there is a device that corrects the black density of the area (correction that increases the black amount).

  For example, in the apparatus disclosed in Patent Document 1, a black edge of a character is detected in a read document, a halftone dot area is detected around the black edge, and the image density is less than a threshold ( With respect to (region), it is determined that the image is saved with toner, and correction processing for increasing the density (black amount) is performed.

Special Review 2014-87009

  However, in an apparatus corresponding to an image reading function (scanner function), toner save determination is limited to black characters printed using only black among toner colors corresponding to a color printer. For example, in a conventional image processing apparatus, cyan (hereinafter also simply referred to as “C”), magenta (hereinafter also simply referred to as “M”), yellow (hereinafter also simply referred to as “Y”), black (black) ( In the following, when reading a document printed using toner of four colors (simply referred to as “K”), the target for toner save determination is limited to black characters printed using only K. Therefore, in the conventional image processing apparatus, toner saving is performed for black characters printed using black (so-called “composite black”) expressed by multiplying the colors of CMYK, color characters (for example, red characters), and the like. Judgment could not be made. Further, in the conventional image processing apparatus, toner save determination cannot be performed for elements other than black characters (for example, photographs, figures (graphics, graphs, tables, etc.)).

  Therefore, an image processing apparatus, an image reading apparatus, and an image forming apparatus that can detect an image formed in an image forming mode with a reduced developer regardless of the type and color of the image are desired.

  The image processing apparatus according to the first aspect of the present invention includes (1) detection means for detecting whether or not an image of a document has been saved with toner, and (2) correction processing for an image for which toner save has been detected by the detection means. And correction means for performing the above.

  According to a second aspect of the present invention, there is provided an image reading apparatus including a document reading unit that reads a document and generates an image of the document, and an image processing unit that processes an image generated by the document reading unit. An image reading apparatus to which the image processing apparatus of the first invention is applied.

  A third aspect of the present invention includes a document reading unit that reads a document and generates an image of the document, an image processing unit that processes an image generated by the document reading unit, and an image forming unit that forms an image on a medium. In the image forming apparatus, the image processing apparatus according to the first aspect of the present invention is applied as the image processing unit.

  According to the present invention, it is possible to provide an image processing apparatus, an image reading apparatus, and an image forming apparatus that detect an image formed in an image forming mode in which a developer is reduced regardless of the type and color of the image.

It is the block diagram shown about the functional structure of the composite apparatus which concerns on embodiment. It is a schematic sectional drawing of the composite apparatus which concerns on embodiment. It is explanatory drawing shown about schematic structure of the image sensor unit which concerns on embodiment. FIG. 10 is an explanatory diagram illustrating a data flow when the multifunction apparatus according to the embodiment operates with toner save detection off. FIG. 6 is an explanatory diagram illustrating a data flow when the multifunction apparatus according to the embodiment operates with toner save detection on. It is the block diagram shown about the internal structure of the lightness value calculation part which concerns on embodiment It is the block diagram shown about the internal structure of the area | region cutting part which concerns on embodiment. It is the block diagram shown about the internal structure of the area | region matching part which concerns on embodiment. It is explanatory drawing shown about the determination conditions and the example of a determination result of the area | region determination part which concern on embodiment. It is the block diagram shown about the internal structure of the identification part which concerns on embodiment. It is explanatory drawing shown about the example of the determination result according to the difference number and the same number in the determination part which concerns on embodiment. It is the block diagram shown about the internal structure of the edge detection part which concerns on embodiment. It is the block diagram shown about the internal structure of the character edge detection part which concerns on embodiment. It is explanatory drawing which shows the matching range of the character edge by the character matching part which concerns on embodiment. It is explanatory drawing shown about the example of the matching pattern of the character edge which the character matching part which concerns on embodiment performs. It is explanatory drawing which shows the example of the process which detects the white-out (temporary) on the character edge which the character edge determination part which concerns on embodiment performs. It is explanatory drawing which shows the example of the image of the character cut out by the character edge detection part which concerns on embodiment. It is the block diagram shown about the internal structure of the chart edge detection part which concerns on embodiment. It is explanatory drawing shown about the example of the matching pattern of the character edge which the chart matching part which concerns on embodiment performs. It is explanatory drawing shown about the process which detects the white-out (temporary) on the chart edge which the chart edge determination part which concerns on embodiment performs. It is the block diagram shown about the internal structure of the photograph edge detection part which concerns on embodiment. It is explanatory drawing shown about the example of the matching pattern of the photograph edge used by the photograph edge determination part which concerns on embodiment. 3 is a block diagram illustrating an internal configuration of a toner save detection unit according to the embodiment. FIG. It is the block diagram shown about the internal structure of the character toner save detection part which concerns on embodiment. It is the block diagram shown about the internal structure of the character histogram part which concerns on embodiment. It is explanatory drawing shown about the example of the histogram used by the character toner save detection part which concerns on embodiment, and a detection condition. It is the block diagram shown about the internal structure of the character toner save rate calculation part which concerns on embodiment. 3 is a block diagram illustrating an internal configuration of a chart toner save detection unit according to the embodiment. FIG. FIG. 3 is a block diagram illustrating an internal configuration of a photographic toner save detection unit according to the embodiment. It is explanatory drawing shown about the example of the photograph histogram which concerns on embodiment. It is a block diagram which shows the internal structure of the photographic toner save detection part which concerns on embodiment. It is explanatory drawing shown about the determination conditions and determination result of the determination which the determination part which concerns on embodiment performs. It is explanatory drawing for verifying the determination process which the determination part which concerns on embodiment performs. It is the block diagram shown about the internal structure of the correction | amendment part which concerns on embodiment. It is the block diagram shown about the internal structure of the character correction part which concerns on embodiment. It is explanatory drawing shown about the example of the correction process performed in the character correction part (R color character correction part / G color character correction part / B color character correction part) which concerns on embodiment. It is the block diagram shown about the internal structure of the chart correction | amendment part which concerns on embodiment. It is explanatory drawing shown about the example of the correction process performed in the chart correction part (R color chart correction part / G color chart correction part / B color chart correction part) which concerns on embodiment. It is the block diagram shown about the internal structure of the photograph correction | amendment part which concerns on embodiment. It is explanatory drawing shown about the example of the tone curve for correction | amendment used in the photographic correction part (R color photographic correction part / G color photographic correction part / B color photographic correction part) concerning an embodiment. 5 is a flowchart illustrating an example of an operation of the multifunction peripheral according to the embodiment (an example of an operation when performing color copying with toner save detection ON).

(A) Main Embodiment Hereinafter, an embodiment of an image reading apparatus and an image forming apparatus according to the present invention will be described in detail with reference to the drawings. Hereinafter, an example in which the image processing apparatus, the image reading apparatus, and the image forming apparatus of the present invention are applied to a composite apparatus will be described.

(A-1) Configuration of Embodiment FIG. 2 is a schematic cross-sectional view of the composite apparatus 1 of this embodiment.

  The multi-function apparatus 1 is an apparatus (so-called MFP (Multi Function Peripheral)) that supports a plurality of functions such as a printer, FAX, and copy.

  As shown in FIG. 2, the composite apparatus 1 includes an image reading apparatus 2 and an image forming apparatus 3. On the upper part of the image forming apparatus 3, the image reading apparatus 2 is arranged.

  The image reading apparatus 2 has a flatbed portion (Flatbed, hereinafter referred to as “FB”) 4. The FB 4 includes a top cover 83, a white reference plate 85, a document table 84 on which the document P is placed, a contact glass 86 that forms the document table 84, a pulley 89, a conveyor belt 90, a stepping motor 91, An image sensor unit 87 fixed to the conveyance belt 90 and a home sensor 88 for detecting the position of the image sensor unit 87 are provided.

  Next, each part of FB4 will be described in detail.

  The conveyor belt 90 is stretched by a stepping motor 91 and a pulley 89, and the image sensor unit 87 is moved along the contact glass 86 by the rotation of the stepping motor 91. The home sensor 88 detects that the image sensor unit 87 has moved to a predetermined position.

  The image sensor unit 87 is moved along the document table 84 by the conveyance belt 90 that operates in accordance with the rotation of the stepping motor 91, and scans the document P placed on the document table 84 to obtain the analog data of the document. obtain. The configuration of the image sensor unit 87 will be described in detail with reference to FIG.

  The white reference plate 85 is read when acquiring data of the white reference plate 85 of the reference image sensor at the time of shading correction for smoothing variations in the voltage output by the image sensor in the image sensor unit 87. It is. The white reference plate 85 is disposed outside the reading range of the document table 84 on the contact glass 86.

  The image forming apparatus 3 is an apparatus (for example, a printer) that can print an image based on input image data on a sheet 93 as a recording medium.

  The image forming apparatus 3 includes a sheet storage cassette 92 that is detachably attached to a lower portion thereof, and a hopping roller 94 disposed on the upper portion of the sheet storage cassette 92. Further, the image forming apparatus 3 includes registration rollers 97, 101, 104 and a conveyance belt 113 starting from a hopping roller 94. Further, the image forming apparatus 3 includes a sheet conveyance path 98 formed in a substantially S shape with the discharge roller 120 as an end point, and a pinch roller 96 disposed so as to be in pressure contact with the registration roller 97. doing. Furthermore, the image forming apparatus 3 includes a pinch roller 100 disposed so as to be in pressure contact with the registration roller 101, and a sensor 99 disposed in front of the sheet conveyance path 98 of the registration roller 101. Yes. In addition, the image forming apparatus 3 includes a pinch roller 103 disposed so as to be in pressure contact with the registration roller 104 and sensors 102 and 106 disposed around the sheet conveyance path 98 of the registration roller 104. ing. Further, the image forming apparatus 3 is detachably mounted on the conveyance belt 113 from the downstream side to the upstream side of the paper conveyance path 98, the driving roller 112 disposed at the position where the conveyance belt 113 is stretched, the driven roller 111. Four image forming units 200 (200K, 200Y, 200M, and 200C). The image forming units 200K, 200Y, 200M, and 200C are electrophotographic LED prints for recording images corresponding to black (K), yellow (Y), magenta (M), and cyan (C) toner colors, respectively. Mechanism.

  The image forming apparatus 3 includes LED (Light Emitting Diode) heads 131 and 132 that irradiate light based on an image signal onto the surface of an image carrier (photosensitive drum 109 described later) of each of the image forming units 200K, 200Y, 200M, and 200C. 133, 134. Further, the image forming apparatus 3 performs a fixing process for fixing the toner image on the transfer rollers 110, 122, 123, and 124, and the paper sheet 93 arranged so as to be in pressure contact with the respective photosensitive drums 109 via the conveyance belt 113. A fixing unit 115 is provided.

  In addition, the image forming apparatus 3 includes a sensor 118 disposed at a subsequent stage of the heat roller 117 (a subsequent stage of the sheet conveyance path 98), a pinch roller 119 disposed so as to be in pressure contact with the discharge roller 120, And a stacker 121 formed using the outer surface of the casing of the image forming apparatus 3.

  Further, the image forming apparatus 3 is disposed at a position where the cleaning toner 126 disposed so as to sandwich the driven roller 111 between the lower surface portion of the conveying belt 113 and the waste toner scraped off by the cleaning blade 126 fall. A waste toner tank 127 and a density sensor 128 disposed in a non-contact manner at a position facing the lower surface of the transport belt 113 are provided.

  The fixing unit 115 includes a heat roller 117 heated by a heat source, a pressure roller 116 disposed so as to be in pressure contact with the heat roller 117, and a thermistor disposed in a non-contact manner in the vicinity of the surface of the heat roller 117. 125.

  Next, each part of the image forming apparatus 3 will be described in detail.

  The sheet storage cassette 92 is provided with a separating means (not shown) at the top thereof, and separates the sheets 93 stored in the sheet storage cassette 92 one by one from the top.

  The hopping roller 94 is connected to the hopping motor 95 and is rotated by the driving force supplied from the hopping motor 95 to take out the sheet 93 separated by a separating means (not shown) and convey it to the registration roller 97.

  The registration roller 97 is rotated by the driving force of the registration motor 105 and corrects the skew of the sheet 93 conveyed by the hopping roller 94 together with the pinch roller 96 disposed so as to be in pressure contact with the registration roller 97. . The registration roller 101 is also connected to the registration motor 105, is rotated by the driving force supplied from the registration motor 105, and is moved by the registration roller 97 together with the pinch roller 100 disposed so as to be pressed against the registration roller 101. The skew of the conveyed sheet 93 is corrected.

  The sensor 99 detects the position of the sheet 93 conveyed by the registration roller 97.

  The registration roller 104 is rotated by the driving force of the registration motor 105 and corrects the skew of the sheet 93 conveyed by the registration roller 101 together with the pinch roller 103 disposed so as to be pressed against the registration roller 104. .

  Sensors 102 and 106 detect the position of the sheet 93 conveyed by the registration roller 101.

  The transport belt 113 electrostatically attracts the charged paper sheet 93 and transports it along the paper transport path 98.

  The driving roller 112 is connected to the belt motor 130 and rotates by the driving force of the belt motor 130 to drive the conveying belt 113 in the direction of arrow e in FIG.

  The image forming units 200K, 200Y, 200M, and 200C all have the same configuration, and are different only in the color of the toner of the toner cartridge to be mounted. Therefore, hereinafter, the configuration of the image forming unit 200 will be described with the image forming unit 200K having black (K) toner as a representative.

  The image forming unit 200 </ b> K includes a toner cartridge 107 that stores black (K) toner and a drum cartridge 108.

  The toner cartridge 107 is detachably mounted on the drum cartridge 108.

  The drum cartridge 108 is provided with a photosensitive drum 109 as an image carrier that is rotatably supported with respect to the frame of the drum cartridge 108, and a predetermined pressure contact amount on the surface of the photosensitive drum 109. A charging roller 135 for uniformly and uniformly charging the surface of the drum 109 and a surface of the photosensitive drum 109 having a predetermined pressure contact amount are disposed on the electrostatic latent image formed on the surface of the photosensitive drum 109 with toner. A developing roller 136 that supplies toner, a toner supply roller 137 that supplies toner to the developing roller 136, and a surface near the surface of the photosensitive drum 109. A neutralizing unit (not shown) that neutralizes the surface of the photosensitive drum 109 is included.

  The image forming units 200K, 200Y, 200M, and 200C are connected to a drum lift-up mechanism (not shown). The drum lift-up mechanism (not shown) is connected to a lift-up motor (not shown). During color printing, the lift-up motor (not shown) drives the drum lift-up mechanism (not shown) to form the image forming units 200K and 200Y. , 200M and 200C are lowered toward the conveyor belt 113, and color printing is possible. During monochrome printing, a lift-up motor (not shown) drives a drum lift-up mechanism (not shown) to lower only the image forming unit 200K toward the transport belt 113, and monochrome printing is possible.

  The LED heads 131, 132, 133, and 134 are each a board (not shown) on which an LED array (not shown), a drive IC (Integrated Circuit) (not shown) for driving the array, and a register group that holds image data are mounted. And a lens array (not shown) that collects light from the LED array.

  A black (K) image signal in the color image signal or the monochrome image signal is input to the LED head 131, and yellow (Y) and magenta (in the color image signal are input to the LED heads 132, 133, and 134. M) and cyan (C) image signals are supplied. The LED heads 131, 132, 133, 134 irradiate the surfaces of the respective photosensitive drums with light based on the input image signals to form electrostatic latent images.

  A bias voltage is applied to the transfer rollers 110, 122, 123, and 124 from the high-voltage power supply 142, and the toner images formed on the surfaces of the photosensitive drums are respectively transferred to the sheet 93 at a predetermined timing.

  The fixing unit 115 includes a heat roller 117, a pressure roller 116, and a thermistor 125.

  A heater 143 is disposed in the heat roller 117.

  The heat roller 117 is connected to the fixing motor 114 and is rotated by the driving force supplied from the fixing motor 114, and the pressure roller 116 is rotated along with the rotation of the heat roller 117.

  The thermistor 125 (not shown) is a means for detecting the surface temperature of the heat roller 117 and is disposed in a non-contact manner in the vicinity of the surface of the heat roller 117. The surface temperature signal of the heat roller 117 detected by the thermistor 125 is supplied to an engine control unit (not shown). The engine control unit described above controls on / off of the heater 143 based on the temperature signal detected by the thermistor 125, and maintains the surface temperature of the heat roller 117 at a predetermined temperature.

  The sensor 118 is a discharge sensor that monitors jamming of the paper 93 in the fixing unit 115 and winding of the paper 93 around the heat roller 117.

  The discharge roller 120 is connected to the fixing motor 114, is rotated by the driving force of the fixing motor 114, and discharges the sheet 93 conveyed from the fixing unit 115 to the stacker 121 along the conveyance path 98.

  FIG. 3 is an explanatory diagram (perspective view) showing a schematic configuration of the image sensor unit 87 in FIG.

  In the example of this embodiment, the image sensor unit 87 is a contact image sensor (hereinafter referred to as “CIS”) unit. The image sensor unit 87 includes an LED light source 150, a light guide plate 149 constituting an optical system, a rod lens 148, and a CIS 147.

  The longitudinal direction of the image sensor unit 87 is the main scanning direction X, and the short direction is the sub-scanning direction Y ′.

  The CIS 147 is composed of a plurality of image sensors, arranged in the main scanning direction X, which is the longitudinal direction of the image sensor unit 87, and configured to be able to read one line of the white reference plate 85 / original P in the main scanning direction X. It shall be. It is assumed that each of the plurality of imaging elements corresponds to each pixel px in the main scanning direction of the read white reference plate 85 / document P data.

  The LED light source 150 includes light emitting diodes (hereinafter referred to as “LEDs”) that are a plurality of light emitting elements that emit different colored light L.

  In the example of this embodiment, the LED light source 150 includes three (three colors) LEDs of red (hereinafter referred to as “R”), green (hereinafter referred to as “G”), and blue (hereinafter referred to as “B”). It shall have. That is, the LED light source 150 of this embodiment has RLED1 which emits R color, GLED1 which emits G color, and BLED1 which emits B color.

  Hereinafter, the color light L emitted from the RLED 1 is referred to as “Lr”. Hereinafter, the color light L emitted from the GLED 1 is referred to as “Lg”. Further, hereinafter, the emission color light L emitted from the BLED 1 is referred to as “Lb”.

  When the image reading apparatus 2 reads the document P, the emitted color lights Lr, Lg, and Lb respectively emitted by the three (three colors) LEDs reach the white reference plate 85 / document P via the light guide plate 149. Then, the reflected lights of Lr, Lg, and Lb from the white reference plate 85 / original P are guided to the CIS 147 through the rod lens 148 and reach the CIS 147. And CIS147 is comprised so that the optical signal of the reflected light may be received and converted into each electric analog signal (analog data).

  Hereinafter, the analog data acquired by the RLED 1 is referred to as “analog data R”. Hereinafter, the analog data acquired by the GLED 1 is referred to as “analog data G”. Further, hereinafter, the analog data acquired by the BLED 1 is referred to as “analog data B”. That is, the analog data acquired by one image sensor (pixel px) has three types of analog data R, G, and B.

  FIG. 1 is a block diagram showing the configuration of the control system of the composite apparatus 1.

  The composite apparatus 1 includes an image reading apparatus 2 and an image forming apparatus 3.

  The image reading device 2 includes a CPU (central processing unit) 5 that controls the entire image reading device 2, a ROM (non-volatile read-only memory) 15, an operation panel 16, a bus 17, a reading unit 6, RAM (volatile memory) 7, shading unit 157, lightness value calculation unit 8, threshold calculation unit 9, region cutout unit 10, edge detection unit 11, toner save detection unit 12, correction Section 13, image processing section 14, and I / F section (interface section) 146. In addition, the image forming apparatus 3 includes a printer unit 18.

  Next, each element in the control system of the composite apparatus 1 will be described.

  The bus 17 is a communication path for transmitting / receiving commands, control signals, and data output from the CPU 5. The CPU 5 connects (accesses) and controls each component via the bus 17.

  In the ROM 15, a control program for controlling the image reading device 2, an algorithm for image processing, data such as a threshold value (for example, firmware) are stored in advance. The CPU 5 controls the entire image reading apparatus 2 based on various control programs stored in the ROM 15.

  Note that the brightness value calculation unit 8, the threshold calculation unit 9, the region extraction unit 10, the edge detection unit 11, the toner save detection unit 12, the correction unit 13, the image processing unit 14, and the shading unit 157 are all hardware (dedicated. It may be configured by a chip, an electronic circuit, or the like, or part or all may be configured in software (for example, configured as a part of a program of firmware recorded on the ROM 15). In addition, the shading unit 157, the lightness value calculation unit 8, the threshold value calculation unit 9, the area extraction unit 10, the edge detection unit 11, the toner save detection unit 12, the correction unit 13, and the image processing unit 14 are provided with one dedicated ASIC. You may make it construct | assemble by (Application Specific Integrated Circuit).

  The image processing apparatus of the present invention can be realized by a component that processes the image data of the document P read by the image reading apparatus 2. The image processing apparatus of the present invention can be realized by components including, for example, a lightness value calculation unit 8, a threshold calculation unit 9, an area cutout unit 10, an edge detection unit 11, a toner save detection unit 12, and a correction unit 13. it can.

  Further, the image reading apparatus of the present invention can be realized by a component that reads a document P and processes the image. The image reading apparatus of the present invention includes, for example, the image reading apparatus 2, the image sensor unit 87, the lightness value calculation unit 8, the threshold value calculation unit 9, the area extraction unit 10, the edge detection unit 11, the toner save detection unit 12, and the correction. It can be realized by a component including the unit 13.

  Further, in this embodiment, whether or not toner is saved for each image of the document P by the lightness value calculation unit 8, the threshold calculation unit 9, the region cutout unit 10, the edge detection unit 11, and the toner save detection unit 12 is determined. Detection means for detecting is configured. In this embodiment, a correction unit that corrects an image saved with toner by the correction unit 13 is configured.

  The operation panel 16 displays an operation instruction from the user and the state of the composite apparatus 1. For example, the operation panel 16 can display an operation screen (menu) for setting ON / OFF of the toner save detection mode under the control of the CPU 5. The CPU 5 displays the above operation screen (toner save detection mode on / off setting menu) on the operation panel 16 and accepts the setting of whether the toner save detection mode is on or off. It may be. It is assumed that the CPU 5 stores and manages a flag F for managing the on / off state of the toner save detection mode in a predetermined storage unit. In FIG. 1, a flag F for managing the on / off state of the toner save detection mode is recorded on the RAM 7. For example, the CPU 5 may read the default value of the flag F from the ROM 15 and record the flag F (initial value) in the RAM 7 at startup. Then, the CPU 5 accepts the update of the content of the flag F in accordance with the operation of the operation panel 16 or the like after activation. The flag F may be stored in a nonvolatile memory (not shown). Further, the CPU 5 executes the toner save detection mode when the flag F is on, and does not execute the toner save detection mode when the flag F is off. Note that the default value of the flag F is not limited.

  The reading unit 6 includes a CIS 147 on the image sensor unit 87, a CIS control unit 151 for controlling the CIS 147, an LED light source 150, a light source control unit 152 for controlling the LED light source 150, and an analog / digital conversion circuit (Analog Front End). , Hereinafter referred to as “AFE”) 153.

  The AFE 153 converts the analog data of the white reference plate 85 / original P input from the CIS 147 into digital data.

  Hereinafter, digital data converted from the analog data R of the white reference plate 85 is also referred to as “white plate data R”. Hereinafter, digital data converted from the analog data G of the white reference plate 85 is also referred to as “white plate data G”. Further, hereinafter, digital data converted from the analog data B of the white reference plate 85 is also referred to as “white plate data B”. That is, one pixel px has three types of data, white plate data R, G, and B.

  Hereinafter, the digital data converted from the analog data R based on the original P is also referred to as “original data R”. Hereinafter, the digital data converted from the analog data G based on the original P is also referred to as “original data G”. Further, hereinafter, the digital data converted from the analog data B based on the original P is also referred to as “original data B”. That is, each pixel px has three types of document data R, G, and B.

  The above-mentioned white board data / original data R, G, B are respectively input to the RAM 7 and temporarily stored. White plate data / original data R, G, and B are supplied from the RAM 7 to the shading unit 157, respectively.

  The shading unit 157 performs shading correction for each reading line of the document P on the document data R, G, B based on the input white plate data R, G, B.

  Hereinafter, the document data R subjected to the shading correction is referred to as “image data R”. Hereinafter, the document data G subjected to the shading correction is referred to as “image data G”. Further, hereinafter, the document data B subjected to the shading correction is referred to as “image data B”. That is, one pixel px has three types of data, image data R, G, and B.

  The image data R, G, and B described above are input to the RAM 7 and temporarily stored.

  The image data R, G, B temporarily stored in the RAM 7 has different input destinations depending on the on / off setting of the toner save detection mode (flag F). When the toner save detection mode is off (flag F is off), the image data R, G, B temporarily stored in the RAM 7 is supplied to the image processing unit 14. On the other hand, when the toner save detection mode is on (flag F is on), the image data R, G, and B temporarily stored in the RAM 7 are supplied to the lightness value calculation unit 8. Differences in the flow of image data according to the on / off setting of the toner save detection mode will be described in detail with reference to FIGS.

  The lightness value calculation unit 8 uses the input image data R, G, and B to calculate the lightness value for each pixel and the average value of the lightness values of the white background. The lightness value calculation unit 8 performs a calculation process based on, for example, the following expressions (1), (2), and (3).

  The following expression (1) is an expression for calculating the brightness value Vpx for each pixel using the image data R, G, and B. Note that the maximum value of the brightness value Vpx for each pixel is 255, and the minimum value is 0. That is, when Vpx reaches the maximum value (255), the density of the pixel becomes the minimum value (0). Further, when Vpx becomes the minimum value (0), the density of the pixel becomes the maximum value (255).

In the following equation (1), “Max image data (R, G, B)” indicates that the largest value among the image data R, image data G, and image data B of the pixel is output. And
Lightness value for each pixel Vpx = Max image data (R, G, B) (1)

The following expression (2) is an expression for calculating the average value “VLave” of the lightness values of all the pixels in one reading line of the main scanning X. It is assumed that one reading line of the main scanning X is composed of m pixels (1st to mth pixels). In the following equation (2), Vpx1 is the lightness value of the first pixel (the lightness value of the first pixel calculated by the equation (1)), and Vpxm is the lightness value of the mth pixel (the equation (1)). (Brightness value of the mth pixel calculated by). In equation (2), m is the number of pixels in one line. The brightness value calculation unit 8 calculates VLave for all the reading lines (first to nth lines) of the main scanning X.
Average value of brightness values of all pixels in one reading line of main scanning X VLave = (Vpx1 +... + Vpxm) / m (2)

The following expression (3) is an expression for calculating the average value Vave related to the lightness value of the read n line in the sub-scanning Y ′. VL1ave in the following equation (3) is the average value of the brightness values of all the pixels in the first scanning line of the main scanning X (the VLave of the first line calculated by the above equation (2)). In addition, VLnave in the following equation (3) is a lightness value of all the pixels on the read nth line in the main scanning X (VLave on the nth line calculated by the above equation (2)).
Average value of brightness values of n lines read in sub-scanning Y ′ Vave = (VL1ave +... + VLnave) / n (3)

  As described above, the lightness value calculation unit 8 can calculate the average value Vave of the lightness values of the white background of the document P using the equations (1) to (3).

  Hereinafter, the brightness value of the white pixel of the document P calculated by the above equation (1) is simply referred to as “Vpx”. Hereinafter, the brightness value of pixels other than the white background of the document P is referred to as “V′px”.

  In this embodiment, an example in which the lightness value calculation unit 8 reads a white background of the document P is within 5 mm from the front end (the front end portion in the document transport direction) of the document P will be described. Hereinafter, the area (front end) regarded as the white background is referred to as “front end white background”. Usually, the front end portion of the document P is white without printing. The detailed configuration of the lightness value calculation unit 8 will be described in detail with reference to FIG. The lightness value Vave of the front end white background calculated by the lightness value calculation unit 8 is supplied to the threshold value calculation unit 9.

The threshold calculation unit 9 calculates a threshold th0 for binarizing V′px of the document P based on the input brightness value Vave. For example, the threshold value calculation unit 9 may calculate the threshold value th0 by multiplying the input brightness value Vave by a certain value (for example, + 10% density). The threshold calculation unit 9 performs processing based on the following equation (4), for example. Then, the threshold calculation unit 9 supplies the calculated threshold th0 to the region cutout unit 10.
Threshold th0 = Vave × (1-10%) = 0.9 × Vave (4)

  The area cutout unit 10 uses the input threshold th0 and the lightness value V′px for each pixel to form an image area (hereinafter also referred to as “character area”) from the document P, or an image area of a figure or a table. (Hereinafter also referred to as “chart area”) and a photographic image area (hereinafter also referred to as “photo area”) are cut out. Each region cut out by the region cutout unit 10 is expressed by binary data (1/0) obtained by V′px. In this embodiment, an image of a figure or table is a figure or table drawn by a line (a frame line of a figure or table, a colored portion in the frame line, etc., and an image of characters in the frame line) Is not included).

  Hereinafter, the binarized data displaying the character area is hereinafter referred to as “V1′z”. Hereinafter, the binarized data displaying the chart area will be referred to as “V2′z”. Further, hereinafter, the binarized data displaying the photographic area is represented as “V3′z”.

  V1'z, V2'z, and V3'z are supplied to the RAM 7, respectively.

  Hereinafter, V′px for each pixel with respect to V1′z is referred to as “V1′px”. Hereinafter, V′px for each pixel with respect to V2′z is referred to as “V2′px”. Furthermore, in the following, V′px for each pixel with respect to V3′z is referred to as “V3′px”. The configuration of the area cutout unit 10 will be described in detail with reference to FIG.

  The edge detection unit 11 cuts out an edge of each region using the input V1′z, V2′z, and V3′z. Edge data of each region detected by the edge detection unit 11 is supplied to the RAM 7. The configuration of the edge detection unit 11 will be described in detail with reference to FIGS.

  The toner save detection unit 12 detects whether or not the toner is saved for each pixel by using V1′px / V2′px / V3′px. When the toner save detection unit 12 detects an image in which toner is saved from the character area and the chart area, the toner save ratio is also calculated. Information on the toner save detected by the toner save detection unit 12 is supplied to the correction unit 13. The configuration of the toner save detection unit 12 will be described in detail with reference to FIGS.

  The correction unit 13 corrects the image data in the area where the toner is saved and acquires correction data.

  Note that the correction unit 13 does not perform correction on an area where toner is not saved. Correction data R, G, and B corrected by the correction unit 13 are supplied to the RAM 7. The configuration of the correction unit 13 will be described in detail with reference to FIGS.

  The image processing unit 14 performs image processing such as gamma correction, color conversion, and compression processing on the image data R, G, B input from the RAM 7 or the correction data R, G, B. The image data / correction data processed by the image processing unit 14 is stored again in the RAM 7.

  The I / F unit 146 has a communication function with the printer unit 18.

  The CPU 5 transmits a command or image data / correction data to the printer unit 18 via the I / F unit 146 or receives a command from the printer unit 18.

  The I / F unit 146 can be configured by, for example, a communication transceiver IC and a receiver IC. The transceiver IC can be composed of, for example, an LVDS (Low Voltage Differential Signaling) transceiver and its communication bus. The receiver IC can be composed of, for example, a USB (Universal Serial Bus) chip and its communication bus.

  The printer unit 18 has a function of printing image data / correction data received from the I / F unit 146 or print data received from an external personal computer (hereinafter referred to as “PC”) 19.

  The printer unit 18 mainly includes an I / F unit 158 for communication with the I / F unit 146, an I / F unit 159 for communication with the PC 19, an engine control unit 144, an image processing unit 160, A high-voltage power supply 142 and a low-voltage power supply 161 are included.

  The I / F unit 159 can be composed of a communication chip and its communication bus. As a communication chip constituting the I / F unit 159, for example, a USB chip or a LAN (Local Area Network) chip can be applied.

  The engine control unit 144 has a function of controlling the hopping motor 95, the registration motor 105, the belt motor 130, the fixing motor 114, the lift-up motor 141 (not shown), and the LED heads 131, 132, 133, and 134 described in FIG. ing.

  The image processing unit 160 has a function of performing image processing on image data / correction data / print data input from the image reading apparatus 2 or the PC 19.

  The image processing unit 160 performs image processing such as expansion / compression, color conversion, binarization conversion, and gamma correction.

  The PC 19 is connected to the I / F unit 159 via the interface cable 145.

  As the interface cable 145, for example, a communication cable such as LAN / USB can be applied.

  Next, the flow of data when toner save detection is off (flag F is off) in the multifunction apparatus 1 will be described with reference to FIG.

  As shown in FIG. 4, document data R, G, and B are supplied from the RAM 7 to the shading unit 157. Then, the image data R, G, and B are supplied from the shading unit 157 to the RAM 7. Then, the image data R, G, and B are supplied from the RAM 7 to the image processing unit 14. Then, the image data R, G, and B processed by the image processing unit 14 are re-input to the RAM 7.

  Next, a data flow when toner save detection is turned on (flag F is turned on) in the composite apparatus 1 will be described with reference to FIG.

  As shown in FIG. 5, document data R, G, and B are supplied from the RAM 7 to the shading unit 157. Image data R, G, and B are supplied from the shading unit 157 to the RAM 7. Then, the image data R, G, and B of all pages (including the front end white background) of the document P are supplied from the RAM 7 to the brightness value calculation unit 8. Then, the lightness value calculation unit 8 supplies the lightness value Vave of the front end white background to the threshold value calculation unit 9.

  Then, V ′ px of the document P is supplied to the RAM 7. Then, the threshold value th 0 is supplied from the threshold value calculation unit 9 to the region cutout unit 10. Then, V′px is also supplied from the RAM 7 to the area cutout unit 10. Then, V1′z, V2′z, and V3′z are supplied from the region cutout unit 10 to the RAM 7, respectively. Then, V1′z, V2′z, and V3′z are supplied from the RAM 7 to the edge detection unit 11, respectively. Then, edge data of each region is supplied from the edge detection unit 11 to the RAM 7. Then, V 1 ′ px / V 2 ′ px / V 3 ′ px is supplied from the RAM 7 to the toner save detection unit 12. Then, the presence / absence information of the image data saved with toner is supplied from the toner save detection unit 12 to the correction unit 13.

  When the toner save information from the toner save detection unit 12 is input to the correction unit 13, the image data R, G, and B are also supplied from the RAM 7 to the correction unit 13. In this case, correction data R, G, and B are supplied from the correction unit 13 to the RAM 7. Then, the correction data R, G, and B are supplied from the RAM 7 to the image processing unit 14. Then, the image data R, G, and B processed by the image processing unit 14 are re-input to the RAM 7. The image data R, G, and B of the image that has not been toner saved is supplied from the RAM 7 to the image processing unit 14 as in FIG. 4 described above.

  FIG. 6 is a block diagram showing the internal configuration of the lightness value calculation unit 8.

  The lightness value calculation unit 8 includes a pixel lightness value calculation unit 154, a line lightness value calculation unit 155, and a multiple line lightness value calculation unit 156.

  The pixel brightness value calculation unit 154 performs processing based on the above equation (1), and can calculate a brightness value for each pixel. The brightness value Vpx calculated by the pixel brightness value calculation unit 154 is supplied to the line brightness value calculation unit 155. The brightness value V′px calculated by the pixel brightness value calculation unit 154 is input to the RAM 7 and is used to cut out each area from the document P.

  The line brightness value calculation unit 155 performs processing based on the above equation (2), and calculates the average value VLave of the brightness values of all the pixels for one line constituting the front end white background.

  The multi-line lightness value calculation unit 156 performs processing based on the above equation (3), and calculates an average value Vave of lightness values of n lines on the front edge white background.

  FIG. 7 is a block diagram showing an internal configuration of the area cutout unit 10.

  The region cutout unit 10 includes a binarization unit 22, a region matching unit 23, and an identification unit 26.

  The binarization unit 22 binarizes V′px as follows by comparing it with V′px of the document P using the input threshold th0.

  When V′px <th0, the binarization unit 22 sets the result of binarization of the V′px to “1”. When V′px ≧ th0, the binarization unit 22 sets the result of binarization of the V′px to 0. The binarizing unit 22 supplies binarized binarized data (1/0) for each pixel to the region matching unit 23. In this embodiment, the value of V′px binarized for each pixel is also simply referred to as “binarized data”.

  The area matching unit 23 matches the input binarized data (1/0) for each pixel for each m ′ × m ′ matrix, character area, chart area, and photograph / chart area (colored) ). That is, the area matching unit 23 matches all areas of the document P in an m ′ × m ′ matrix according to the reading order of the document P.

  Since the result of matching the photographic area and the colored chart area by the area matching unit 23 is the same, the photograph / graphic area (colored) is set. The above-described two types of areas (photograph area and chart area) are separately identified by the identification unit 26 described later. Information on the result of matching by the region matching unit 23 is supplied to the character region storage unit 24, the photo / chart region storage unit 25, and the chart region storage unit 28 in the RAM 7, respectively.

  Although the set value of m ′ × m ′ is not limited, for example, the value of m ′ may be set so as to correspond to main scanning 10 mm × sub-scanning 10 mm. The configuration of the region matching unit 23 will be described in detail with reference to FIG.

  The identification unit 26 separately identifies a photographic region and a diagram region (colored) using information stored in the photograph / chart region storage unit 25. The identification unit 26 supplies information of the identified photographic area and chart area (with coloring) to the photographic area storage section 27 and the chart area storage section 28, respectively. The identification unit 26 will be described in detail later with reference to FIG.

  The character area storage unit 24, the chart area storage unit 28, and the photo area storage unit 27 store (hold) each area information and V1'z, V2'z, and V3'z for each pixel.

  FIG. 8 is a block diagram showing an internal configuration of the region matching unit 23.

  The region matching unit 23 includes a counting unit 29, a subtracter 30, a comparison unit 31, a counting unit 32, and a region determination unit 33.

  The counter 29 counts 1 and 0 for each m ′ × m ′ matrix with respect to the input information (1/0) for each pixel. The count numbers of 1 and 0 counted by the counting unit 29 (the count number obtained by counting the number of 1 and the count number obtained by counting the number of 0) are respectively supplied to the subtractor 30.

The subtracter 30 subtracts the number of 1 (the count number of 1) from the number of 0 (the count number of 0) by using the numbers of 1 and 0 (counter numbers) input for each m ′ × m ′ matrix. . The subtracter 30 performs processing based on the following equation (5). Then, the subtractor 30 supplies the calculated result X ′ to the comparison unit 31.
Calculation result X ′ = (count number of 0) − (count number of 1) (5)

  The comparison unit 31 compares the input X ′ with threshold values th1, th2, th3, and th4, respectively. The thresholds th1, th2, th3, th4 are stored in the ROM 15 in advance. Note that the thresholds th1, th2, th3, and th4 have a relationship of th4> th3> th1> th2. Also, it is assumed that the total number of pixels in the m ′ × m ′ matrix is set in th4. Information on the result of comparison by the comparison unit 31 is supplied to the area determination unit 33 and the RAM 7. The format of the comparison result output by the comparison unit 31 is not limited. For example, when th3 <X ′ <th4, a signal indicating the result is output. Note that the determination result of the comparison unit 31 is any one of “th1 <X ′ <th3”, “th2 <X ′ <th1”, “X ′ = th4”, “th3 <X ′ <th4”. .

  The comparison unit 31 supplies information of the comparison result to the counting unit 32 when th3 <X ′ <th4. Then, the counting unit 32 includes four m ′ × m ′ matrices (hereinafter, “ Refer to the comparison result of the “connected matrix”. Then, the counting unit 32 counts the number of the four connected matrices connected to the attention matrix that have the same comparison result (th3 <X ′ <th4). For example, when the comparison result between the upper and lower connected matrices of the attention matrix is “th3 <X ′ <th4”, the counting unit 32 sets the counting result related to the attention matrix to “2”. Then, the information of the result counted by the counting unit 32 is supplied to the area determination unit 33. The counting unit 32 obtains information on the comparison result of the connected matrix from the RAM 7.

  The area determination unit 33 determines the area of the document P (attribute of each m ′ × m ′ matrix) based on the information supplied from the comparison unit 31 and the counting unit 32.

  FIG. 9 is an explanatory diagram illustrating an example of determination conditions and determination results of the region determination unit 33.

  As shown in FIG. 9, the region determination unit 33 of this embodiment performs determination based on the four types of comparison results of the comparison unit 31 and the count results of the counting unit 32.

  As shown in FIG. 9, for the m ′ × m ′ matrix in which the comparison result is “th1 <X ′ <th3”, “th2 <X ′ <th1”, or “X ′ = th4”, The counting process in the counting unit 32 is not performed. On the other hand, for the m ′ × m ′ matrix in which the comparison result is a predetermined result (“th3 <X ′ <th4”), the counting unit 32 performs the counting process, so that the region determination unit 33 uses the counting unit. Thirty-two counting results are considered.

  As illustrated in FIG. 9, the area determination unit 33 determines that the m ′ × m ′ matrix whose comparison result is “th1 <X ′ <th3” is “character area”.

  Further, as shown in FIG. 9, the area determination unit 33 sets “photo / chart area (colored)” for the m ′ × m ′ matrix whose comparison result is “th2 <X ′ <th1”. judge.

  Further, the area determination unit 33 determines that the m ′ × m ′ matrix whose comparison result is “X ′ = th4” is a “blank area”.

  Furthermore, the area determination unit 33 performs determination in consideration of the counting result of the counting unit 32 for the m ′ × m ′ matrix in which the comparison result is “th3 <X ′ <th4”. For the m ′ × m ′ matrix in which the comparison result is “th3 <X ′ <th4” and the counting result of the counting unit 32 is “1” or more, the region determination unit 33 displays “the chart region (colored unpainted)”. ) ”. For the m ′ × m ′ matrix in which the comparison result is “th3 <X ′ <th4” and the counting result of the counting unit 32 is “0”, the region determination unit 33 reads “character region (symbol)”. judge. The symbol in this case represents a symbol such as “,” “,” “;”.

  When the determination result other than “blank area” is obtained, the area determination unit 33 temporarily stores the determination result of the m ′ × m ′ matrix in the RAM 7.

  FIG. 10 is a block diagram showing the internal configuration of the identification unit 26.

  The identification unit 26 includes a comparison counting unit 34 and a determination unit 36.

  The comparison counting unit 34 performs a process of comparing V′px for each pixel of each m ′ × m ′ matrix. Then, the comparison counting unit 34 counts the number of “differences” and “same values” of the V′px values in each m ′ × m ′ matrix. Then, the comparison counting unit 34 supplies the number of “difference numbers” and “same values” of the V′px values in each m ′ × m ′ matrix to the determination unit 36.

  For example, the comparison counting unit 34 may acquire the number of V′px values having the largest number in the m ′ × m ′ matrix as the number of “same values”. For example, the comparison counting unit 34 may acquire the number of V′px having a value different from the value determined to be the same value as the number of “differences”.

  The determination unit 36 compares the number of the same values as the difference number of the input V′px values with a preset threshold value. Then, the determination unit 36 determines whether the m ′ × m ′ matrix is a photographic area or a chart area (colored) based on the comparison result.

  FIG. 11 is an explanatory diagram illustrating an example of determination results according to the number of differences and the same number in the determination unit 36. In FIG. 11, the total number of pixels in the m ′ × m ′ matrix is represented as “Di”.

  As shown in FIG. 11, the determination unit 36 has an m ′ × m ′ matrix in which the number of differences in V′px values is one and the number of V′px having the same value is 0.8 Di or more. , It is determined to be a chart area (colored). Further, as shown in FIG. 11, the determination unit 36 determines that “the number of differences in the V ′ px value is two and the number of V ′ px having the same value is 0.4 Di or more”, “the V ′ px value is “The number of differences is 3 and the number of V′px having the same value is 0.26 Di or more”, “The number of differences in the V′px value is 4 and the number of V′px having the same value is 0. 2Di or more "," V'px value difference number is 5 and the same value V'px number is 0.16 Di or more "," V'px value difference number is 6 and the same The value of V′px of the value is 0.13 Di or more ”,“ the number of differences in the V′px value is 7 and the number of V′px of the same value is 0.11 Di or more ” The m ′ × m ′ matrix is determined to be a chart area (colored). Furthermore, the determination unit 36 determines that an m ′ × m ′ matrix having “the number of differences in V′px values is 10 or more and the number of V′px having the same value is 0.1 Di or less” is a photographic region. To do.

  As described above, when the number of differences in V′px values is equal to or less than a predetermined value (7 or less), the determination unit 36 determines that the same number of V′px values corresponding to each number of differences is equal to or greater than a threshold (ratio to Di The m ′ × m ′ matrix is determined to be a chart area (colored) on the condition that the threshold value is greater than or equal to the threshold value represented by (1). For the same number of thresholds corresponding to the number of differences, for example, a suitable value may be obtained in advance by experiments or the like and recorded in the ROM 15 or the like.

  FIG. 12 is a block diagram illustrating an internal configuration of the edge detection unit 11.

  The edge detection unit 11 includes a character edge detection unit 37, a chart edge detection unit 38, and a photograph edge detection unit 39.

  The character edge detection unit 37 cuts out (extracts) an edge portion (hereinafter referred to as “character edge”) for each character region using V1′z of each pixel of the character region stored in the character region storage unit 24. To do). The range in which the character edge detection unit 37 matches the character edge is the character region and the surrounding pixels. The configuration of the character edge detector 37 will be described in detail with reference to FIG.

  The chart edge detection unit 38 uses the V2′z of each pixel in the chart area stored in the chart area storage unit 28 to cut out a graphic and a table edge portion (hereinafter referred to as “chart edge”). The range in which the chart edge detection unit 38 matches the chart edge is the chart area and its periphery. The configuration of the chart edge detector 38 will be described in detail with reference to FIG.

  The photo edge detection unit 39 cuts out an edge portion of the photo (hereinafter referred to as “photo edge”) using V3′z of each pixel in the photo area stored in the photo area storage unit 27. The range in which the photograph edge detection unit 39 matches the photograph edge is the photograph area and its periphery. The configuration of the photograph edge detection unit 39 will be described in detail later with reference to FIG.

  FIG. 13 is an explanatory diagram showing the internal configuration of the character edge detector 37 of FIG.

  The character edge detection unit 37 includes a character matching unit 40 and a character edge determination unit 42.

  The character matching unit 40 uses V1′z for each pixel stored in the character region storage unit 24 and binary data (1/0) for each pixel around the character region as a character edge detection matrix ( For example, matching is performed using a 3 × 3 matrix. Although the size of the detection matrix is not limited, the following description will be made assuming that the detection matrix is a matrix of 3 vertical pixels × 3 horizontal pixels (hereinafter also referred to as “3 × 3 matrix”).

  The character matching unit 40 supplies information (1/0) of each pixel (target pixel I) matched by a 3 × 3 matrix (inspection matrix) to the character edge determination unit 42. The matching range of the character matching unit 40 will be described in detail with reference to FIG.

  The character edge determination unit 42 performs character alignment on the target pixel I with a character edge matching pattern set in advance using information on the target pixel I in the input 3 × 3 matrix (inspection matrix). It is determined whether or not it is an edge or “white spot (temporary)” on a character edge.

  “White blank (temporary)” represents the result of determining that “white blank” due to toner saving has occurred for the pixel. When a character related to a pixel determined to be blank (provisional) is finally detected as a character in which toner is saved by the toner save detection unit 12, “white” It is assumed that the review to “missing (true)” is performed (determining that the pixel is a blank pixel).

  Moreover, the character edge determination part 42 cuts out one character assembled by the pixels of the determined character edge. When the character edge determination unit 42 cuts out one character described above, the character edge determination unit 42 detects, as one character, an area in which the periphery of the character (periphery of the character edge) is entirely surrounded by zero values. Note that in the image reading apparatus 2, the image inside the edge cut out by the character edge determination unit 42 is also applied to a large or bold image (for example, a character drawn with a line having a thickness of 3 pixels or more). It can be processed as a part of pixels of a character image (region). The character edge matching pattern will be described in detail later with reference to FIG. Further, the matching pattern of white spots on the character edge (temporary) will be described in detail with reference to FIG.

  FIG. 14 is an explanatory diagram showing a matching range of character edges by the character matching unit 40.

  As shown in FIG. 14, the range in which the character edge is matched by the character matching unit 40 is a character region and its surrounding pixels (pixels adjacent to the pixel in the character region). In FIG. 14, a range surrounded by a dotted line is illustrated as a range in which character edges are matched by the character matching unit 40.

  FIG. 15 is an explanatory diagram showing an example of a matching pattern for character edges performed by the character matching unit 40.

  Each matching pattern MP shown in FIG. 15 has a 3 × 3 matrix (inspection matrix) format. In FIG. 15, three groups of matching patterns MP are illustrated depending on the number of pixels whose binarized data is 1.

  In FIG. 15, 16 matching patterns MP101 to MP116 are illustrated as matching patterns including three pixels whose binarized data is 1. In FIG. 15, twelve matching patterns MP117 to MP128 are illustrated as matching patterns including four pixels whose binarized data is 1. Further, in FIG. 15, twelve matching patterns MP129 to MP140 are illustrated as matching patterns including five pixels whose binarized data is 1.

  Hereinafter, the pixel of interest I whose binarized data is 1 in the above-described matching patterns MP101 to MP140 is referred to as “edge pixel of interest I”.

  When the 3 × 3 matrix similar to any one of the above 40 matching patterns MP101 to MP140 is input from the character matching unit 40, the character edge determination unit 42 includes the 3 × 3 matrix. Assume that an edge-of-edge pixel I whose binarized data is 1 is determined to be a character edge.

  FIG. 16 is an explanatory diagram illustrating an example of processing for detecting white spots (temporary) on a character edge.

  In the following, it is assumed that an identifier (number) of each pixel px of the 3 × 3 matrix is assigned as shown in FIG. In FIG. 16G, numbers 1 to 3 are assigned in order from the left to the upper pixel px of the 3 × 3 matrix. In FIG. 16G, the pixels px in the middle stage of the 3 × 3 matrix are numbered 4 to 6 in order from the left. Further, in FIG. 16G, the numbers 7 to 9 are assigned in order from the left to the lower pixel px of the 3 × 3 matrix. Hereinafter, the pixels px having the numbers 1 to 9 are referred to as px1 to px9, respectively.

  In the following, in FIG. 16, pixels detected as white spots (temporary) in the matrix of interest are shown hatched (hatched).

  As shown in FIG. 16 (f), FIG. 16 shows an example in which white spot (temporary) is detected from the matrix of interest (3 × 3 matrix) with the matching pattern MP 101 of FIG. 15. In the matching pattern MP101, the pixels px4, px5, and px6 are 1 and are determined to be character edges.

  When the attention matrix (3 × 3 matrix) has the contents as shown in FIGS. 16A to 16C, the character edge determination unit 42 detects white spots (temporary).

  The attention matrix shown in FIGS. 16A to 16C is one pixel (out of the pixels px4 to px6) out of the three pixels whose binarized data is 1 as compared with the matching pattern MP101. Only one of the pixels) is 0. Therefore, as shown in FIGS. 16A to 16C, the character edge determination unit 42 removes the pixel (the pixel whose binarized data is 0 compared to the matching pattern MP101). It is detected as a (provisional) pixel. Similarly, the character edge determination unit 42 can detect white spots (temporary) as shown in FIGS. 16A to 16C even in any one of the other 39 matching patterns of FIG. Is possible.

  As described above, the character edge determination unit 42 compares the attention matrix (3 × 3 matrix) with the edge attention pixel I of each matching pattern MP, and when there is a pixel having a different value by one pixel (only one pixel). In the case where a pixel having a value of 0 exists, the pixel (a pixel that is 0 compared to the matching pattern MP) is determined to be blank (temporary).

  As described above, as shown in FIGS. 16A to 16C, characters having white (provisional) pixels are detected by the toner save detection unit 12 as toner-saved characters. It is assumed that the white spot (temporary) is reexamined as a white spot (true) (a process for confirming that it is a white spot pixel).

  In addition, as shown in FIGS. 16D and 16E, the character edge determination unit 42 considers the matching result of a matrix adjacent to the matrix of interest (hereinafter referred to as “adjacent matrix”). White pixels (temporary) in the matrix are detected.

  In FIG. 16D and FIG. 16E, an example of detecting white spots (temporary) in the adjacent matrix M102 using the matching result between the attention matrix M101 and the adjacent matrix M102 (matching result with the matching pattern MP101). Show.

  As shown in FIGS. 16D and 16E, the character edge determination unit 42 compares the attention matrix M101 with the matching pattern MP101, and as a result, the edge attention pixel I whose data is 0 is 2 When there are two pixels (when there are two or more pixels having different data), the adjacent matrix M102 is compared with the matching pattern MP101. Then, as a result of comparing the adjacent matrix M102 and the matching pattern MP101, the character edge determining unit 42 determines that there are two or more matching edge target pixels I (pixels whose data is 1) in the target matrix M101. The two edge target pixels I whose data is 0 (pixels whose data is 0) are detected as white spots (temporary).

  In the example of FIG. 16D, when the attention matrix M101 and the matching pattern MP101 are compared, the data of the pixels px5 and px6 among the edge attention pixels I of the matching pattern MP101 are 0. Then, when the character edge determination unit 42 compares the adjacent matrix M102 and the matching pattern MP101, there are three matching edge target pixels I (pixels whose data is 1) (pixels px4, px5, and px6). I can recognize that. Therefore, the character edge determination unit 42 detects two px5 and px6 whose data is 0 in the attention matrix M101 as white spots (temporary).

  Similarly, in the example of FIG. 16E, the pixels px4 and px5 of the attention matrix M101 are set as white spots (temporary) based on the matching result of the attention matrix M101 and the adjacent matrix M102 (matching result with the matching pattern MP101). Detected. Similarly, the character edge determination unit 42 can detect white spots (temporary) as shown in FIGS. 16D and 16E even in any one of the other 39 matching patterns of FIG. Is possible.

  Note that, as described above, as shown in FIGS. 16D and 16E, characters having blank (provisional) pixels are detected as toner-saved characters by the toner save detection unit 12. In this case, it is assumed that the review of white spot (provisional) to white spot (true) is reviewed (determining whether the pixel is a white spot).

  FIG. 17A and FIG. 17B are diagrams showing an image of the character “bright” cut out by the character edge detection unit 37. FIG. 17A shows an image when there is no white spot (temporary), and FIG. 17B shows an image when there is white spot (temporary).

  In FIG. 17A, each pixel constituting “bright” is assembled by a pixel whose data is 1 (edge attention pixel). In FIG. 17A, the periphery of the edge of the image constituting “bright” is all solidified by pixels with 0 data. In FIG. 17B, pixels whose data is 0 (empty (provisional) pixels) are included on the edge of the image forming “bright”.

  FIG. 18 is a block diagram showing the internal configuration of the chart edge detector 38.

  The chart edge detection unit 38 includes a chart matching unit 44 and a chart edge determination unit 46.

  The chart matching unit 44 uses the V2′z for each pixel stored in the chart area storage unit 28 and the binarized data (1/0) for each pixel around the chart area as a matrix for detecting a chart edge. Match with. In this embodiment, description will be made assuming that a matrix for detecting a chart edge uses a matrix larger than a 3 × 3 matrix.

  In the example of this embodiment, as shown in FIG. 19, the chart matching unit 44 is configured by a matching pattern MP201 configured by a matrix of horizontal 6 pixels × vertical 3 pixels and a matrix of horizontal 3 pixels × vertical 6 pixels. It is assumed that the matching process using the matching pattern MP202 and the matching patterns MP203 and MP204 configured by a matrix of 6 horizontal pixels × 6 vertical pixels is performed. In the matching patterns MP201 to MP204, the target pixel II whose data is 1 is hereinafter referred to as “edge target pixel II”.

  In the following, the matrix for which the chart matching unit 44 detects the chart edge is also referred to as a “chart matrix”.

  The chart matching unit 44 matches each pixel (target pixel II) of the input chart matrix with each matching pattern MP201 to MP204, and supplies the result to the chart edge determination unit 46.

  The chart edge determination unit 46 uses a preset matching pattern (in this embodiment, the matching patterns MP201 to MP204) for the information (binary data) of the target pixel II in the input chart matrix. It is determined whether or not it is an edge, or whether or not there is a white spot (temporary) on the chart edge. The chart edge determination unit 46 also performs a process of cutting out one figure / table assembled by the pixels of the determined chart edge. In the chart edge determination unit 46, the number of matching patterns used for matching and the size of each matching pattern are not limited.

  When the figure / edge determination unit 46 cuts out one figure / table, the figure / table is surrounded by a value of 0 around the figure / table (around the figure / table edge). Detect as. In addition, in the image reading apparatus 2, for a figure / table drawn with a thick line (for example, a line having a thickness of 3 pixels or more), the inside (for example, the edge of the edge cut out by the chart edge determination unit 46) A region constituting a thick line) can also be processed as a part of pixels of the graphic / table image (region).

  When the information (chart matrix) that matches any one of the matching patterns MP201 to MP204 is input from the chart matching unit 44, the chart edge determination unit 46 has data 1 in the chart matrix. The edge attention pixel II is determined to be a chart edge.

  FIG. 20 is an explanatory diagram showing processing for detecting white spots (temporary) on the chart edge performed by the chart edge determination unit 46. FIG. 20 shows an example of a white (temporary) detection process using a 6 × 3 matrix matching pattern MP201.

  As shown in FIG. 20, the chart edge determination unit 46 compares the chart matrix with the edge target pixel II of each matching pattern MP, and when there are three or less different pixels (the data of the edge target pixel II has 0 data). When the number of pixels is 3 or less), the pixel (the pixel whose data is 0 compared to the edge target pixel II of the matching pattern MP) is determined to be blank (temporary).

  In the example of FIGS. 20A to 20D, pixels having different data are in the range of 1 to 3 compared to the chart matrix and the edge target pixel II of the matching pattern MP201. Therefore, the chart edge determination unit 46 has different data in the chart matrix shown in FIGS. 20A to 20D compared to the edge target pixel II of the matching pattern MP201 (data becomes 0). Pixels) are detected as white spots (temporary).

  FIG. 21 is a block diagram illustrating an internal configuration of the photograph edge detection unit 39.

  The photo edge detection unit 39 includes a photo matching unit 48 and a photo edge determination unit 49.

  The photo matching unit 48 uses the V3′z for each pixel stored in the photo area storage unit 27 and the binarized data (1/0) for each pixel around the photo area as a matrix for detecting a photo edge. Match with.

  The photo edge detection matrix used in the photo matching unit 48 is the same as the character edge detection matrix, and a 3 × 3 matrix (a matrix of 3 horizontal pixels × 3 vertical pixels) is used. In the following, the matrix that is subject to photo edge matching in the photo matching unit 48 is referred to as a “photo matrix”. In the following, the pixel of interest III whose data is 1 in the matching pattern for detecting a photo edge will be referred to as “edge pixel of interest III” below.

  The photo matching unit 48 matches each pixel (target pixel III) of the input photo matrix with a matching pattern for detecting a photo edge, and supplies the result to the photo edge determining unit 49.

  The photo edge determination unit 49 uses the information of the pixel of interest III in the input photo matrix to determine whether or not it is a photo edge with respect to the pixel of interest III in a preset photo edge matching pattern, or Then, it is determined whether or not there is a white spot (temporary) on the photo edge. In addition, the photograph edge determination unit 49 cuts out a photograph assembled by the pixels of the determined photograph edge. Note that in the image reading device 2, when the periphery of the edge of the photograph is all set to 0, it is detected as a photograph.

  FIG. 22 is an explanatory diagram showing an example of a matching pattern for a photographic edge used in the photographic edge determination unit 49. In FIG. 22, 16 matching patterns MP301 to MP316 used for photo edge matching are illustrated.

  The process for detecting white spots (temporary) on the photo edge performed by the photo edge determining unit 49 is the same as the process for detecting the white spots (temporary) of the characters described above (the same process as the character edge determining unit 42). Can be detected. This is because, as described above, the photographic edge determination unit 49 performs processing in units of 3 × 3 matrix in the same manner as the character edge determination unit 42.

  FIG. 23 is a block diagram showing the internal configuration of the toner save detection unit 12.

  The toner save detection unit 12 includes a character toner save detection unit 50, a chart toner save detection unit 51, and a photographic toner save detection unit 52.

  The character toner save detection unit 50 uses the V1 ′ px of each pixel in the character area stored in the RAM 7 to detect whether or not toner is saved for each character. The character toner save detection unit 50 supplies toner save information of the detected character to the character correction unit 69 of the correction unit 13. The configuration of the character toner save detection unit 50 will be described in detail with reference to FIG.

  The chart toner save detection unit 51 uses the V2′px of each pixel in the chart area stored in the RAM 7 to detect whether or not toner is saved for each chart or chart. The chart toner save detection unit 51 supplies the detected toner save information of the chart to the chart correction unit 70 of the correction unit 13. The configuration of the chart toner save detector 51 will be described in detail with reference to FIG.

  The photographic toner save detection unit 52 detects whether or not toner is saved in the photograph using the V7 'px of each pixel in the photographic area from the RAM 7. The detected toner save information of the photograph is supplied to the photograph correcting section 71 of the correcting section 13. The configuration of the photographic toner save detection unit 52 will be described in detail with reference to FIG.

  FIG. 24 is a block diagram showing the internal configuration of the character toner save detection unit 50.

  The character toner save detection unit 50 includes a character histogram unit 53, a character toner save detection unit 54, and a character toner save rate calculation unit 55.

  The character histogram unit 53 uses the V1 ′ px of all pixels related to a single character edge to create a histogram related to the character. The configuration of the character histogram unit 53 will be described in detail with reference to FIG.

  The character toner save detection unit 54 detects whether or not toner is saved for each character image based on the information supplied from the character histogram unit 53. The condition for detecting whether or not the toner is saved by the character toner save detection unit 54 is shown in FIG.

  When the character toner save rate calculation unit 55 detects that the character toner save detection unit 54 detects that the character is saved in the character, the character toner save rate calculation unit 55 calculates an average value of the character toner save rate and the character toner save rate. The character toner save rate calculation unit 55 performs processing based on the following equations (6) and (7).

The character toner save rate calculation unit 55 calculates the toner save rate per1 of one character by using the following equation (6) in all pixels of the edge of one character where the toner is saved. In the following equation (6), “sum1” indicates the sum of the number of pixels in each density section that is less than the maximum density section in the pixel of the character that is the target of calculation of the toner save rate per1. In the following equation (6), “coefficient 1” is assumed to be “the number of all pixels of the edge for each 1 / character”.
1 character toner save rate per1 = sum1X coefficient 1 (6)

In the following equation (7), “the sum of N-character toner save rates” is the sum of toner save rates per1 for N characters (N characters). The value of N is not limited, but may be N = 10, for example. “Coefficient 2” in the following equation (7) is assumed to be 1 / N. For example, if N = 10, “coefficient 2 = 1/10 = 0.1”.
Toner save rate average value S1av
= (Sum of N-shaped toner save rates) x (coefficient 2) (7)

  The character toner save rate calculation unit 55 supplies the calculated S1av to the character correction unit 69. The configuration of the character toner save rate calculation unit 55 will be described in detail with reference to FIG.

  FIG. 25 is a block diagram showing the internal configuration of the character histogram unit 53.

  The character histogram unit 53 includes a counting unit 57, a threshold calculation unit 56, a comparison unit 58, and a counting unit 59.

  The counting unit 57 counts the number of pixels in each preset density interval using V1′px of all the pixels of the edge for each character. The counting unit 57 supplies the counted number of pixels in each density interval to the comparison unit 58.

The threshold value calculation unit 56 calculates the threshold value th7 by multiplying the data (value) of all the pixels of the edge for each character by a certain value (for example, 5%). For example, the threshold value calculation unit 56 performs processing based on the following equation (8). The threshold calculation unit 56 also supplies the calculated th7 to the comparison unit 58.
th7 = number of all pixels of one-character edge X0.05 (8)

  The comparison unit 58 compares the number of input pixels in each density section with th7. The comparison unit 58 supplies information on the comparison result to the counting unit 59.

  The counting unit 59 counts the number of density intervals with the number of pixels equal to or greater than th7 using the input comparison result information, and supplies it to the character toner save detection unit 54.

  The character toner save detection unit 54 uses a histogram based on the result supplied from the counting unit 59 to detect whether or not toner is saved for each character.

  FIG. 26 is an explanatory diagram showing an example of a histogram used in the character toner save detection unit 54 and its analysis conditions. FIG. 26A shows an example of a histogram relating to an edge of one character (for example, black character) in which toner is saved. FIG. 26B is an explanatory diagram showing conditions for the character toner save detection unit 54 to detect whether or not toner is saved for each character.

  Then, the character toner save detection unit 54 analyzes the histogram as shown in FIG. 26A according to the rule of FIG. 26B to detect whether the toner is saved for each character.

  In the histogram of FIG. 26A, the horizontal axis is “V1′px density interval”, and the vertical axis is “number of pixels px for the density interval”. In the histogram of FIG. 26A, the th7 level is indicated by a dotted line. In the histogram shown in FIG. 26A, there are five density sections with the number of pixels equal to or greater than th7.

  In other words, the character toner save detection unit 54 of this embodiment detects whether or not toner is saved for each character by analyzing the above-described histogram according to the condition shown in FIG.

  In FIG. 26B, two rules are described. In FIG. 26B, a rule is described in which, when there is one density section with the number of pixels equal to or greater than th7 per character, the character is detected as a character that is not toner-saved. In FIG. 26 (b), there is a rule that when there are two “density areas with the number of pixels of th7 or more” in one character, this character is detected as a toner-save character. Yes.

  That is, the character toner save detection unit 54 determines whether or not each character is a character whose toner has been saved by using “the number of density intervals with the number of pixels equal to or greater than th7”.

  FIG. 27 is a block diagram showing the internal configuration of the character toner save rate calculation unit 55.

  The character toner save rate calculation unit 55 includes an adder 60, a multiplier 61, an adder 173, and a multiplier 176.

  The adder 60 is an adder for calculating the sum sum1 of “the number of pixels in each density section less than the maximum density section” in the above equation (6), and the input “number of pixels in each density section less than the maximum density section”. ”To calculate sum1. The number of pixels counted by the counting unit 57 can be applied to the “number of pixels in each density section less than the maximum density section” described above.

  The multiplier 61 calculates a toner save rate for each character by multiplying “sum1” input from the adder 60 by “coefficient 1”. The multiplier 61 supplies the calculated toner save rate for each character to the adder 173.

  The adder 173 is an adder that calculates the “sum of N-character toner save rates” in the equation (7).

  The multiplier 176 calculates S1av by multiplying the “sum of toner save rates” input from the adder 173 by “coefficient 2”. The multiplier 176 inputs the calculated S1av to the character correction unit 69.

  FIG. 28 is a block diagram showing an internal configuration of the chart toner save detection unit 51.

  The chart toner save detection unit 51 includes a chart histogram unit 62, a chart toner save detection unit 63, and a chart toner save rate calculation unit 64.

  The chart / histogram section 62 creates a histogram of this chart / table by using V2'px of all pixels related to the edge of one chart / table. The configuration (processing contents) of the chart histogram unit 62 is the same as the configuration of the character histogram unit 53 described above, and thus detailed description thereof is omitted.

  The chart / toner save detection unit 63 detects whether or not each chart / table is toner-saved based on the information input from the chart histogram unit 62.

  Since the two conditions shown in FIG. 26 (b) can be applied to the condition that the chart / toner save detection unit 63 detects whether or not the toner / save is saved for each figure / table, detailed description thereof is omitted. To do.

  The chart / toner save rate calculator 64 calculates the toner / save ratio of the chart / table in which the chart / save save detector 63 detects that the toner is saved in the chart / table.

The chart toner save rate calculation unit 64 performs processing based on the following equation (9). In the following expression (9), “per2” indicates the toner save rate of one figure / table. In the following formula (9), “sum2” represents the sum of the number of pixels in each density section less than the maximum density section. Furthermore, in the following equation (9), “coefficient 3” can be obtained by “1 / number of all pixels of the edge for each chart”.
Toner save rate per2 = sum2 × factor 3 in the figure / table (9)

  The chart / toner save rate calculation unit 64 supplies the chart / table toner save rate per2 calculated based on the formula (9) to the chart correction unit 70.

  FIG. 29 is a block diagram showing the internal configuration of the photographic toner save detection unit 52.

  The photographic toner save detection unit 52 includes a photographic histogram unit 65, a filter unit 66, and a photographic toner save detection unit 68.

  The photo histogram unit 65 creates a photo histogram using V3'px of all pixels in the photo area. That is, the photograph histogram unit 65 performs a process of counting the number of pixels with respect to V3′px.

  The filter unit 66 excludes isolated data that is present in the darkest part and the brightest part from the photograph histogram created by the photograph histogram part 65 and data that has a small number of appearances (data whose number of appearances is not more than a predetermined value). Process. This is because the above-described data is likely to be noise in a photographic image. FIG. 30 is an explanatory diagram showing an example of a photographic histogram. FIG. 30 shows an image of data excluded from the photo histogram (isolated data in the darkest part and the brightest part, and data with a small number of appearances). In the histogram shown in FIG. 30, the horizontal axis of the photographic histogram indicates V3 'px (density), and the vertical axis indicates the number of pixels px. In the histogram shown in FIG. 30, the thick curve indicates the photographic histogram created by the photographic histogram unit 65. In the histogram shown in FIG. 30, two parts of data surrounded by a dotted line (ellipse) (data isolated from the thick curve) are excluded by the filter unit 66, so that the photographic toner save detection unit 68 is not input. That is, as shown in FIG. 30, the filter unit 66 supplies the photographic toner save detection unit 68 with the above-described data (data estimated to be noise) excluded from the photographic histogram.

  The photographic toner save detection unit 68 uses the photographic histogram data supplied from the filter unit 66 and a preset threshold th8 to detect whether or not toner is saved for each photo image. The configuration of the photographic toner save detection unit 68 will be described in detail with reference to FIG.

  FIG. 31 is a block diagram showing the internal configuration of the photographic toner save detection unit 68.

  The photographic toner save detection unit 68 includes a split rate calculation unit 67, a detection unit 162, and a determination unit 163.

  The split ratio calculation unit 67 is the number of pixels in the density interval from gray (for example, the color of “127” data) to light (for example, the color of “255” (maximum value 255)) of the photo histogram. The "division rate" ("ratio"; "ratio") of the sum of

The split rate calculation unit 67 performs calculation based on the following equation (10). In the following formula (10), “sum3” represents the sum of the number of pixels in the density section. In the following equation (10), “sum3” may be obtained by calculation using an adder (not shown) in the split rate calculation unit 67. In the following equation (10), “1 / number of all pixels of the photo histogram” can be applied to the “coefficient 4”.
Split rate = sum3 × coefficient 4 (10)

  The split rate calculation unit 67 supplies the calculated “split rate” to the determination unit 163.

  The detection unit 162 performs a process of comparing the sum “sum4” of the number of pixels in the dark density section on the photograph histogram with the “threshold th8”. The threshold th8 may be set in advance in the ROM 15 or the like, for example.

  For example, the dark density section of the photograph histogram may be a section from density O to 20 (density section of density 20 or less). “Sum4” is calculated by the adder 167 in the detection unit 162.

  The detection unit 162 supplies (outputs) information on the result of comparison (any one of the following two results) to the determination unit 163. If “sum4 ≧ th8”, the detection unit 162 determines that “the number of pixels is in the dark density section of the photo histogram”. In addition, when “sum4 <th8”, the detection unit 162 determines that “the number of pixels is not in the dark density section of the photo histogram”. For example, the detection unit 162 may output “1” as the detection result when “sum4 ≧ th8”, and output “0” as the detection result when “sum4 <th8”.

  The determination unit 163 determines whether or not the toner of each image is saved based on the information supplied from the split rate calculation unit 67 and the detection unit 162.

  FIG. 32 is an explanatory diagram illustrating determination conditions and determination results of determination performed by the determination unit 163.

  As shown in FIG. 32, the determination unit 163 determines whether the data from the split rate calculation unit 67 satisfies the condition that “the split rate is 60% or more” and the data from the detection unit 162 satisfies “sum4 <th8”. It is determined that the toner is saved. As shown in FIG. 32, the determination unit 163 determines that toner saving is not performed on a photographic image that does not meet the above-described conditions.

  The determination unit 163 supplies the determination result (information on whether toner is saved for each photo) to the photo correction unit 71.

  The photo correction unit 71 corrects the photo histogram of each photo based on the determination result supplied from the determination unit 163.

  FIG. 33 is an explanatory diagram for verifying the determination process (determination process based on FIG. 32 described above) performed by the determination unit 163.

  33A to 33D, respectively, a photo histogram based on a photo that has not been toner-saved, and a photo histogram based on an image when the photo is toner-saved (a photo based on a photo that has been toner-saved). An example of a histogram is shown. In FIG. 33A to FIG. 33D, a photo histogram in which toner is not saved is shown by a thick solid line, and a histogram in which toner is saved is shown by a thick dotted line.

  The photograph histogram (thick solid line histogram) in which toner is not saved in FIG. 33A is a distribution with a good density balance (a normal distribution having a density near the center (127) as a vertex).

  In the photo histogram (thick solid line histogram) in FIG. 33 (b) where toner is not saved, the ratio of dark density sections (the ratio of dark sections from the center (127)) is large.

  In the photo histogram (thick solid line histogram) in FIG. 33 (c) where toner is not saved, the ratio of gray sections (ratio near the center (127)) is large.

  In FIG. 33 (d), the ratio of the bright section (the ratio of the section brighter than the center (127)) is large in the photo histogram (thick solid line histogram) where toner is not saved.

  As described above, FIGS. 33 (a) to 33 (d) show the non-toner-saved photo histogram and the toner-save photo histogram for the photos with different density balances. As a result of the applicant's experiment, in each graph of FIG. 33A to FIG. 33D, when a photo histogram based on a photo not saved with toner is compared with a photo histogram based on a photo saved with toner. It has been found that all the conditions for determining whether or not the toner has been saved described with reference to FIG. 32 are satisfied. That is, it has been found that the condition described with reference to FIG. 32 contributes to the detection of whether or not the toner is saved regardless of the balance of the density of the photo to be detected.

  FIG. 34 is a block diagram showing the internal configuration of the correction unit 13.

  The correction unit 13 includes a character correction unit 69, a chart correction unit 70, and a photo correction unit 71.

  The character correction unit 69 corrects the image data of the characters saved with toner. The character correction unit 69 supplies character correction data to the RAM 7 for recording. The configuration of the character correction unit 69 will be described in detail with reference to FIG.

  The chart correction unit 70 corrects the image data of the chart saved with toner. The chart correction unit 70 supplies the chart correction data to the RAM 7 for recording. The configuration of the chart correction unit 70 will be described in detail with reference to FIG.

  The photo correction unit 71 corrects image data of a photo saved with toner. The photographic correction unit 71 supplies photographic correction data to the RAM 7 for recording. The configuration of the photo correction unit 71 will be described in detail with reference to FIG.

  FIG. 35 is a block diagram showing the internal configuration of the character correction unit 69.

  The character correcting unit 69 includes a minimum value calculating unit 170, a correction rate calculating unit 168, an R color character correcting unit 72, a G color character correcting unit 73, and a B color character correcting unit 74.

  The minimum value calculation unit 170 calculates a minimum value (maximum density value) from V1′px of all pixels in the edge portion of the character (hereinafter referred to as “character edge”).

The correction rate calculation unit 168 calculates a correction rate for each pixel of the character edge (hereinafter referred to as “correction rate I”). The correction factor calculation unit 168 performs processing for calculating the correction factor I based on the following equation (11). In the following equation (11), the value calculated by the minimum value calculation unit 170 can be applied as the “maximum density value”.
Correction factor I
= (V1′px of the pixel−maximum density value) / V1′px of the pixel (11)

  The maximum density value calculated by the minimum value calculation unit 170, the correction rate I calculated by the correction rate calculation unit 168, and S1av calculated by the character toner save rate calculation unit 55 are all R color character correction unit 72 /. It is supplied to the G color correction unit 73 / B color correction unit 74. Then, the R color correction unit 72 / G color correction unit 73 / B color correction unit 74 corrects the image data R / G / B of each pixel constituting each character (image correction processing). )I do. Specifically, the R color correction unit 72 / G color correction unit 73 / B color correction unit 74 performs predetermined correction processing on the image data R / G / B of each pixel constituting each character. (For example, correction processing defined in FIG. 36 described later) is executed.

  FIG. 36 is an explanatory diagram showing an example of correction processing performed by the R color correction unit 72 / G color correction unit 73 / B color correction unit 74. In the following description, it is assumed that “V1′px = image data G” is set in advance in the lightness value calculation unit 8 and the character correction unit 69. Note that the reference color in the lightness value calculation unit 8 and the character correction unit 69 is not limited to G, and may be another color. In other words, in the character correction unit 69, it is assumed that the color to be corrected to the maximum density value (hereinafter also referred to as “first color”) is set to G. Note that the color to be corrected to the maximum density value may be another color.

  As shown in FIG. 36, the R color correction unit 72 / G color correction unit 73 / B color correction unit 74 applies the following to image data R, G, B of one pixel of the character edge as follows. Perform image correction.

  When the image data G of the pixel at the edge of the character is less than the maximum density value (including the case where the pixel is blank (true)), the G color correction unit 73 converts the data of the image data G to the maximum density. Perform correction to change to value. Hereinafter, the corrected image data G is also referred to as “correction data G”.

  The R color correction unit 72 performs processing for correcting the data of the image data R of the pixel at the character edge to “image data R × (1−correction rate I of this pixel) of this pixel”.

  The B color correction unit 74 performs processing for correcting the data of the image data B of the pixel of the character edge to “image data B × (1−correction rate I of this pixel)”.

  As shown in FIG. 36, the R color correction unit 72 / G color correction unit 73 / B color correction unit 74 are image data of each pixel in an internal part of the character (hereinafter referred to as “character internal”). Image correction is performed on R, G, and B as follows. The R color correction unit 72 performs a process of correcting the image data R data of the pixels inside the character into “image data R × (1-S1av) of this pixel”. The G color correction unit 73 performs processing for correcting the image data G of the pixels inside the character to “image data G × (1-S1av) of this pixel”. The B color correction unit 74 performs processing for correcting the data of the image data B of the pixels inside the character into “image data B × (1-S1av) of this pixel”.

  FIG. 37 is a block diagram showing the internal configuration of the chart correction unit 70.

  The chart corrector 70 includes a minimum value calculator 172, a correction rate calculator 171, an R color chart corrector 77, a G color chart corrector 78, and a B color chart corrector 79.

  The minimum value calculation unit 172 calculates the minimum value (maximum density value) by V2′px of all pixels at the edge of each figure / table.

The correction rate calculation unit 171 calculates a correction rate for each pixel at the edge of each figure / table (hereinafter referred to as “correction rate II”). The correction factor calculation unit 171 can obtain the correction factor II for each pixel at the edge of each figure / table by calculation based on the following equation (12). Note that the value calculated by the minimum value calculation unit 172 can be applied to the “maximum density value” in the following equation (12).
Correction factor II
= (V2′px−maximum density value of this pixel) / V2′px of this pixel (12)

  The maximum density value calculated by the minimum value calculation unit 172, the correction rate II calculated by the correction rate calculation unit 171 and the per2 calculated by the chart toner save rate calculation unit 64 are all R color chart correction unit 77 /. This is supplied to the G color chart correction unit 78 / B color chart correction unit 79. The R color chart corrector 77 / G color chart corrector 78 / B color chart corrector 79 performs a process (image correction process) for correcting the image data R / G / B of each pixel in each figure / table. To do.

  Specifically, the R color chart correction unit 77 / G color chart correction unit 78 / B color chart correction unit 79 performs predetermined processing on the image data R / G / B of each pixel constituting each figure / table. Correction processing (for example, correction processing defined in FIG. 38 described later) is executed.

  FIG. 38 is an explanatory diagram showing an example of correction processing performed by the R color chart correction unit 77 / G color chart correction unit 78 / B color chart correction unit 79. In the following description, it is assumed that “V2′px = image data G” is set in advance in the lightness value calculation unit 8 and the chart correction unit 70. In other words, in the chart correction unit 70, it is assumed that the color to be corrected to the maximum density value (hereinafter also referred to as “first color”) is set to G. Note that the color to be corrected to the maximum density value may be another color.

  As shown in FIG. 38, the R color chart correction unit 77 / G color chart correction unit 78 / B color chart correction unit 79 corrects image data R, G, and B of one pixel at the chart edge as follows. Processing is performed to obtain correction data R, G, and B.

  When the image data G of the pixel at the chart edge is less than the maximum density value (including the case where the pixel is blank (true)), the G color chart correction unit 78 converts the data of the image data G to the maximum density. Perform correction to change to value.

  The R color chart correction unit 77 performs a process of correcting the image data R data of the pixel at the chart edge to “image data R × (1−correction rate II of this pixel)”.

  The B color chart correction unit 79 performs processing for correcting the data of the image data B of the pixel at the chart edge to “image data B × (1−correction rate II of this pixel)”.

  As shown in FIG. 38, the R color chart correction section 77 / G color chart correction section 78 / B color chart correction section 79 is a portion inside the chart edge (including the paint color area (solid-filled area)). Image correction is performed on the image data R, G, and B of each pixel (hereinafter referred to as “the inside of the chart edge”) as follows. The R color chart correction unit 77 performs processing for correcting the data of the image data R of the pixels inside the chart edge to “image data R × (1-per2) of this pixel”. The G color chart correction unit 78 performs processing for correcting the data of the image data G of the pixels inside the chart edge to “image data G × (1-per2) of this pixel”. The B color chart correction unit 79 performs a process of correcting the data of the image data B of the pixels inside the chart edge to “image data B × (1-per2) of this pixel”.

  FIG. 39 is a block diagram showing the internal configuration of the photo correction unit 71.

  The photo correction unit 71 includes an R color photo correction unit 80, a G color photo correction unit 81, and a B color photo correction unit 82.

  The R-color photo correction unit 80 performs correction processing based on preset data (a “correction tone curve” to be described later) on the image data R of each pixel constituting the toner-saved photo. The G color photo correction unit 81 performs a correction process based on preset data (a “correction tone curve” to be described later) on the image data G of each pixel constituting the toner-saved photo. The B-color photo correction unit 82 performs correction processing based on preset data (a “correction tone curve” to be described later) on the image data B of each pixel constituting the toner-saved photo.

  Specifically, the photo correction unit 71 performs correction processing for further improving the density of the image data R, G, and B of the photo saved with toner, using a correction tone curve for increasing the density.

  FIG. 40 is an explanatory diagram showing an example of a correction tone curve used in the R color photographic correction unit 80 / G color photographic correction unit 81 / B color photographic correction unit 82.

  In FIG. 40, γ represents an example of a correction tone curve. In FIG. 40, γ 'is a tone curve in which toner is saved. The image data R, G, and B of the photograph in which the toner is saved are corrected to values converted by the correction tone curve γ. As shown in FIG. 40, when the input values are the same, γ has a larger output value as a density than γ ′.

(A-2) Operation | movement of embodiment Next, operation | movement of the composite apparatus 1 of embodiment which has the above structures is demonstrated.

  FIG. 41 is a flowchart showing toner save detection and correction operations when color copying is performed by the multifunction apparatus 1 of FIG.

  When the user places the document P on the document table 84, sets the toner save detection mode on the operation panel 16 to ON, and presses a color copy button 177 (not shown) on the operation panel 16, the following operation starts.

  First, the image sensor unit 87 reaches below the white reference plate 85. Then, the image sensor unit 87 reads the white reference plate 85 to acquire white plate data R, G, and B. Then, the image sensor unit 87 reaches under the white background of the front end of the document P. Then, the image sensor unit 87 reads the document P and acquires document data R, G, and B of the document P including the front end white background (S1).

  Next, the image sensor unit 87 determines whether or not the reading of the original P has been completed (S2). When the reading of the document P is completed, the multifunction apparatus 1 starts from step S3 to be described later. When the reading is not completed, the multifunction apparatus 1 returns to the above-described step S1 and continues reading the document P.

  When the reading of the document P is completed, in the multifunction apparatus 1, the shading unit 157 performs shading correction on the document data R, G, and B, and acquires the image data R, G, and B (S3).

  Next, the threshold value calculation unit 9 calculates the threshold value th0 (S4).

  Next, a character area, a chart area, and a photograph area are cut out from the original P by the area cutout unit 10 (S5).

  Next, the character edge detector 37 cuts out an edge for each character from each character region. Further, the chart / edge detector 38 cuts out the edge of each chart / table from each chart area. Furthermore, the edge of each photograph is cut out from each photograph area by the photograph edge detection unit 39 (S6).

  Next, the toner save detection unit 12 checks whether toner is saved in each image (character, chart, or photo image) (S7). Specifically, the character toner save detection unit 50 detects whether or not each character is toner-saved. In addition, the chart / toner save detection unit 51 detects whether each chart / table is toner-saved. Further, the photograph toner save detection unit 52 detects whether or not each photograph is saved with toner.

  When the toner save detection unit 12 detects toner save in any of the images (characters, charts, or photographic images), the correction unit 13 performs processing for correcting each image in which the toner save is detected. Perform (S8, S9). Specifically, the character correction unit 69 corrects the character image data R, G, B for which toner save is detected, and acquires character correction data R, G, B. Further, the chart correction unit 70 corrects the image data R, G, and B of the chart / table where the toner save is detected, and acquires the chart / table correction data R, G, and B. Further, the photo correction unit 71 corrects the image data R, G, and B of the photo in which toner save is detected, and acquires the photo correction data R, G, and B.

  Next, the image processing unit 14 performs image processing (images that can be supplied to the printer unit 18) image data R, G, and B (including correction data R, G, and B of each region that has been toner saved) on each region. Converted into data) (S10).

  Next, the printer unit 18 prints an image based on the image data R, G, B (including correction data R, G, B after image processing) of each region supplied from the image processing unit 14 on the sheet 93. (S11) A series of copy processing ends. Specifically, the printer unit 18 converts the image data R, G, and B of the document P (including the correction data R, G, and B image-processed in step S10 described above) to the print data K, Y, M, and Convert to C. The printer unit 18 inputs the printing data K, Y, M, and C to the LED heads 131, 132, 133, and 134, respectively, and causes the image forming units 200K, 200Y, 200M, and 200C to form toner images. Then, the printer unit 18 transfers the toner images formed by the image forming units 200K, 200Y, 200M, and 200C to the sheet 93, respectively. Then, the printer unit 18 controls the fixing unit 115 to fix the color toner image on the sheet 93 and then conveys the sheet 93 to the stacker 121.

(A-3) Effects of Embodiment According to this embodiment, the following effects can be achieved.

  In the multi-function apparatus 1 of this embodiment, when the toner save detection mode is turned on when the color original P is read in the color mode (for example, the flag F is turned on by operating the operation panel 16), the area cutting unit 10 A character area, a chart area, and a photograph area are cut out from the image. Then, the character edge detection unit 37 extracts edge data for each character from the cut character region. Then, the character toner save detection unit 50 creates a histogram for each character using the extracted edge data for each character, and counts the number of density sections having the number of pixels equal to or greater than the threshold th7 in the created histogram. Then, the character toner save detection unit 50 detects whether toner is saved for each character. Then, the character correction unit 69 performs correction processing on the detected image data of the character whose toner has been saved. Further, in the composite apparatus 1 of this embodiment, toner save detection and correction processing is executed for the cut chart area and photograph area.

  As described above, in the composite apparatus 1 according to this embodiment, correction is performed by detecting whether or not toner is saved for each character, chart, and photograph regardless of color (for example, correction such as increasing the density). )I do. As a result, in the multifunction apparatus 1 of this embodiment, even when an image of a document that has been saved with toner is read, an image having a quality equivalent to that when a document that has not been saved with toner is read is acquired. Can do.

(B) Other Embodiments The present invention is not limited to the above-described embodiments, and may include modified embodiments as exemplified below.

  (B-1) In the above embodiment, the image sensor unit of the image reading apparatus 2 included in the composite apparatus 1 has been described as being of the CIS type, but the specific reading method is not limited. For example, a CCD (Charge Coupled Device) type may be applied as the image sensor unit of the image reading apparatus 2 included in the composite apparatus 1.

  (B-2) The image reading apparatus 2 constituting the composite apparatus 1 of the above embodiment has been described as including the FB. However, the image reading apparatus 2 may further include an automatic document feeder (Automatic Document Feeder).

  (B-3) In the above embodiment, the example in which the image processing apparatus, the image reading apparatus, and the image forming apparatus of the present invention are applied to the composite apparatus 1 including the image reading apparatus 2 and the image forming apparatus 3 has been described. The present invention can be applied to an apparatus (for example, an apparatus such as a FAX or a copying machine) provided with an image processing apparatus that processes an image obtained by reading another document. Further, the image processing apparatus and the image reading apparatus of the present invention may be applied to a simple image reading apparatus (for example, a scanner) such as the image reading apparatus 2.

  DESCRIPTION OF SYMBOLS 1 ... Composite apparatus, 2 ... Image reading apparatus, 3 ... Image forming apparatus, 4 ... FB, 5 ... CPU, 6 ... Reading part, 7 ... RAM, 8 ... Lightness value calculation part, 9 ... Threshold value calculation part, 10 ... Area | region Extraction unit, 11 ... edge detection unit, 12 ... toner save detection unit, 13 ... correction unit, 14 ... image processing unit, 15 ... ROM, 16 ... operation panel, 17 ... bus, 18 ... printer unit, 19 ... PC, DESCRIPTION OF SYMBOLS 23 ... Area matching part, 24 ... Character area memory | storage part, 25 ... Chart area memory | storage part, 26 ... Identification part, 27 ... Photo area memory | storage part, 28 ... Chart area memory | storage part, 29 ... Counting part, 30 ... Subtractor, 31 ... Comparison part, 32 ... Counter part, 33 ... Area determination part, 34 ... Comparison counting part, 36 ... Determination part, 37 ... Character edge detection part, 38 ... Chart edge detection part, 39 ... Photo edge detection part, 40 ... Character Matching unit, 42 ... character edge determining unit, 44 ... chart matching unit 46 ... Chart edge determination unit, 48 ... Photo matching unit, 49 ... Photo edge determination unit, 50 ... Character toner save detection unit, 51 ... Chart toner save detection unit, 52 ... Photo toner save detection unit, 53 ... Character histogram unit, 54 ... Character toner save detection unit, 55 ... Character toner save rate calculation unit, 56 ... Threshold calculation unit, 57 ... Counter, 58 ... Comparator, 59 ... Counter, 60 ... Adder, 61 ... Multiplier, 62 ... Chart histogram section, 63 ... Chart toner save detection section, 64 ... Chart toner save ratio calculation section, 65 ... Photo histogram section, 66 ... Filter section, 67 ... Split ratio calculation section, 68 ... Photo toner save detection section, 69 ... Character Correction unit, 70 ... Chart correction unit, 71 ... Photo correction unit, 72 ... R color correction unit, 73 ... G color correction unit, 74 ... B color correction unit, 77 ... R color chart correction unit, 78 ... G Color chart correction 79 ... B color chart correction unit, 80 ... R color photo correction unit, 81 ... G color photo correction unit, 82 ... B color photo correction unit, 83 ... Top cover, 84 ... Document platen, 85 ... White reference plate, 86 Reference glass, 87 ... Image sensor unit, 88 ... Home sensor, 89 ... Pulley, 90 ... Conveyor belt, 91 ... Stepping motor, 92 ... Paper storage cassette, 93 ... Paper, 94 ... Hopping roller, 95 ... Hopping motor, 96 ... Pinch roller, 97 ... Registration roller, 98 ... Paper transport path, 99 ... Sensor, 100 ... Pinch roller, 101 ... Registration roller, 102 ... Sensor, 103 ... Pinch roller, 104 ... Registration roller, 105 ... Registration motor, 106 ... Sensor 107... Toner cartridge 108. Drum cartridge 109 109 Photosensitive drum 110. Transfer roller 111... Driven roller 112... Driving roller 113 113 transport belt 114 fixing motor 115 fixing unit 116 pressure roller 117 heat roller 118 sensor 119 pinch roller 120 Ejection roller, 121 ... Stacker, 122 ... Transfer roller, 123 ... Transfer roller, 124 ... Transfer roller, 125 ... Thermistor, 126 ... Cleaning blade, 127 ... Waste toner tank, 128 ... Concentration sensor, 130 ... Belt motor, 131, 132 133, 134 ... LED head, 135 ... charging roller, 136 ... developing roller, 137 ... toner supply roller, 142 ... high voltage power supply, 143 ... heater, 144 ... engine control unit, 145 ... interface cable, 146 ... I / F unit 148 ... Rod lens, 149 ... Light guide plate DESCRIPTION OF SYMBOLS 150 ... LED light source, 151 ... CIS control part, 152 ... Light source control part, 154 ... Pixel brightness value calculation part, 155 ... Line brightness value calculation part, 156 ... Multiple line brightness value calculation part, 157 ... Shading part, 158 ... I / F unit, 159 ... I / F unit, 160 ... image processing unit, 161 ... low voltage power supply, 162 ... detection unit, 163 ... determination unit, 167 ... adder, 168 ... correction rate calculation unit, 170 ... minimum value calculation unit 171 ... Correction rate calculation unit, 172 ... Minimum value calculation unit, 173 ... Adder, 176 ... Multiplier, 177 ... Color copy button, 200, 200C, 200K, 200M, 200Y ... Image forming unit.

Claims (11)

Detecting means for detecting whether the image of the document is toner-save;
An image processing apparatus comprising: correction means for performing correction processing on an image for which toner save has been detected by the detection means. The detection means includes
A character edge detection unit that extracts an edge for each character from the image of the original;
Using the density data of each pixel of the edge of the character cut out by the character edge detection unit, create a histogram of the character, and count the number of density sections in which the pixel having a threshold value or more exists in the created histogram, The image processing apparatus according to claim 1, further comprising: a character toner save detection unit that detects whether toner is saved for the character by using the count result. The correction means includes
With respect to the first color image data indicated by the brightness of the first color that constitutes the pixel of the edge of the character that is saved by toner, the data that is less than the maximum density value is corrected to the maximum density value, thereby configuring the pixel. A character correction unit that corrects the second color image data indicated by the brightness of a color other than the first color based on a correction rate calculated using the first color image data. The image processing apparatus according to claim 2. The detection means further includes a character toner save rate calculation unit that calculates a toner save rate of the character determined to be toner saved by the character toner save detection unit,
The correction means is based on the average value of the toner save rate calculated by the character toner save rate calculation unit for the image data indicated by the lightness of the color constituting the pixels inside the toner saved character. The image processing apparatus according to claim 3, wherein correction is performed. The detection means includes
A chart edge detector that cuts out an edge for each figure or chart from the image of the document;
Using the density data of each pixel at the edge of the figure or table cut out by the chart edge detection unit, the histogram of the figure or table is created, and the number of density sections in which the pixels above the threshold exist in the created histogram The image processing apparatus according to claim 1, further comprising: a chart toner save detection unit that detects whether or not toner is saved for the chart or table using the count result. The correction means includes
For the first color image data indicated by the brightness of the first color constituting the pixel of the edge of the chart or table where the toner is saved, data less than the maximum density value is corrected to the maximum density value, and the pixel is A chart correction unit that corrects the second color image data indicated by the lightness of the color other than the first color, based on a correction rate calculated using the first color image data. The image processing apparatus according to claim 5. The detection means further includes a chart toner save rate calculation unit that calculates a toner save rate of the chart or table determined to be toner saved by the chart toner save detection unit,
The chart correction unit is an average value of the toner save rate calculated by the chart toner save rate calculation unit for the image data indicated by the brightness of the color that constitutes the pixel in the chart or table in which the toner is saved The image processing apparatus according to claim 6, wherein correction is performed based on the correction. The detection means includes
A photo edge detection unit for cutting out an edge for each photo from the image of the original;
A histogram of the photograph is created using the density data of each pixel in the photograph area including the edge cut out by the photograph edge detection unit, and the histogram for the number of pixels of all pixels in the photograph area is created using the created histogram. Using the ratio of the sum of the number of pixels in a density section brighter than one density and the sum of the number of pixels in a density section darker than a second density different from the first density, the photographic area is toner-saved. The image processing apparatus according to claim 1, further comprising: a photographic toner detection unit configured to detect whether or not the toner image is detected. The correction means includes
The image processing apparatus according to claim 8, further comprising: a photographic correction unit that performs a tone curve correction process for increasing a density of a photographic area where toner is saved.   10. An image reading apparatus comprising: a document reading unit that reads a document and generates an image of the document; and an image processing unit that processes an image generated by the document reading unit. An image reading apparatus to which the image processing apparatus according to claim 1 is applied.   An image forming apparatus comprising: an original reading unit that reads an original and generates an image of the original; an image processing unit that processes an image generated by the original reading unit; and an image forming unit that forms an image on a medium. An image forming apparatus to which the image processing apparatus according to claim 1 is applied as a processing unit.

Images