JPH06178111A - Image processor - Google Patents

Abstract

PURPOSE:To form a high definition copy image by preventing dense information from being emphasized too much when an original for repeating copies and an original having much ratio of dense information part are copied. CONSTITUTION:After an A/D conversion 30 is performed for the analog image signals converted into RGB electric signals by an image reader control part 1022 and a black correction/white correction 31 is performed for the signals, each signal is inputted in an ND signal generation part 32 and a color detection part 33. Then, a luminance signal Dout is outputted from the generation part 32, color signals Cout of red, green, blue, etc., are outputted from the color detection part 33, and a density correction 36 is performed for the signals. Namely, in this correction part 36, the corrections of density conversions and the gradation of printers are performed by using a look-up table 36. Thus, dense information is prevented from being emphasized too much when the original for repeating copies and the original having much ratio of dense information part are copied and the originals are outputted to a printer control part 1024.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus to which an automatic density conversion method of document information (hereinafter referred to as "AE processing") for optimally reproducing a document is applied.

[0002]

2. Description of the Related Art Generally, in this type of image processing apparatus, a document is read by an image input apparatus, converted into an electric signal, image processing is performed on this signal, and then an image is output by an output apparatus such as a laser printer. It is known to be recorded. At this time, the background part of the document, the part to be emphasized, and the document type are recognized from the histogram of the read document and its feature points, and the background of the document is skipped and the information part is emphasized darkly so that The AE process for outputting a document by the optimum process is realized.

[0003]

[Problems to be Solved by the Invention] However, the above A
In the E processing, since the information portion of the document is emphasized deeply, when generation copy is repeated or when a document having a large proportion of dark information portion is copied, an image that gives an impression of being over-emphasized may be output. It was The present invention solves the above-mentioned problems, and when a document that is repeatedly copied or a document that has a large proportion of dark information portions is copied, a high-quality copy image is obtained by not unnecessarily emphasizing dark information. An object is to provide an image processing apparatus to be obtained.

[0004]

[Means for Solving the Problems]
In order to achieve the object, an image processing apparatus according to the present invention provides a creating unit that creates a histogram for each level of an electric signal obtained by reading a document, and a feature point based on the histogram created by the creating unit. Detecting means for detecting, determining means for determining the original type based on the characteristic points detected by the detecting means, and converting means for applying signal conversion processing according to the original type determined by the determining means to the electric signal With.

[0005]

According to this structure, the creating means creates a histogram for each level of the electric signal obtained by reading the document, and the detecting means detects the characteristic points based on the histogram created by the creating means and makes a determination. The means determines the document type based on the feature points detected by the detection means, and the conversion means performs signal conversion processing according to the document type determined by the determination means on the electric signal.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A preferred embodiment of the present invention will be described in detail below with reference to the accompanying drawings. <Embodiment 1> FIG. 1 is a sectional view showing the structure of an image copying apparatus according to Embodiment 1 of the present invention. In the figure, reference numeral 1 denotes a document feeding device serving as a document feeding means, which feeds the placed documents one by one or continuously to a predetermined position on the platen glass surface 2. Reference numeral 3 is a scanner including a lamp, a scanning mirror 5 and the like.
When it is placed on the image sensor, the main body is reciprocally scanned in a predetermined direction, the reflected light of the document passes through the lens 8 through the scanning mirror 5-7, and is separated into colors by an RGB color separation filter (not shown). An image is formed on the part 9.

Reference numeral 10 denotes an exposure control section composed of a laser scanner, which is an image signal control section 23 of the controller section CONT.
The photoconductor 11 is irradiated with a light beam that is modulated based on the image data output from (see FIG. 2). Reference numerals 12 and 13 denote developing devices that visualize the electrostatic latent image formed on the photoconductor 11 with a developer (toner) of a predetermined color. Denoted at 14 and 15 are transfer paper stacking units, in which standard-size recording media are stacked and housed, and are fed to the registration position by driving the feeding roller, and the image leading edge alignment timing with the image formed on the photoconductor 11 is set. The paper is re-fed in the state where it has been taken.

Reference numeral 16 is a transfer separation charger, which transfers the toner image developed on the photoconductor 11 to the transfer paper and then the photoconductor 11
It is further separated and fixed by the fixing unit 17 via the conveyor belt. A paper discharge roller 18 stacks and discharges the transfer-receiving paper on which the image formation is completed on the tray 20. Reference numeral 19 denotes a direction flapper, which switches the carrying direction of the transfer-receiving paper after the image formation to the paper discharge port and the internal carrying path direction to prepare for the multiplex / double-sided image forming process.

Image formation on a recording medium will be described below. The image signal input to the image sensor unit 9, that is, the input signal from the reader 22 described later is processed by the image signal control circuit 23 controlled by the CPU control unit 1025 and reaches the printer unit 24. The signal input to the printer is converted into an optical signal by the exposure control unit 10 and the photoconductor 11 is illuminated according to the image signal. The latent image formed on the photoconductor 11 by the irradiation light is developed by the developing device 12 or the developing device 13. The transfer paper stacking section 14 or the transfer paper stacking section 15 is matched with the latent image timing.
The transfer paper is conveyed, and the developed image is transferred at the transfer unit 16. The transferred image is fixed on the transfer paper by the fixing unit 17, and then discharged from the paper output unit 18 to the outside of the apparatus.

During double-sided recording, after the transfer paper has passed the paper discharge sensor, the paper discharge unit 18 is rotated in the direction opposite to the paper discharge direction. At the same time, the flapper 20 is raised upward to store the copied transfer paper in the intermediate tray 24 via the transport paths 22 and 23. The transfer paper stored in the intermediate tray 24 is fed during the next back surface recording, and the back surface is transferred.

Further, at the time of multiple recording, the flapper 21 is raised and the copied transfer paper is stored in the intermediate tray 24 via the transport paths of the transport paths 22 and 23. The transfer paper stored in the intermediate tray 24 is fed to the next multiplex recording, and the multiplex transfer is performed. FIG. 2 is a block diagram illustrating the configuration of the controller unit CONT shown in FIG.
In the figure, 1025 is a CPU circuit unit, which is the ROM 102.
6. A RAM 1027 is built in, and each unit is comprehensively controlled based on a control program stored in the ROM 1026.

Reference numeral 1021 denotes an automatic document feeder control unit which controls the feeding of the placed originals one by one or two sheets in succession to a predetermined position on the surface of the original glass 2. 10
An image reader control unit 22 is composed of the image sensor unit 9 and the like, and outputs to the image control unit 1023 an analog image signal color-separated by an RGB separation filter (not shown) and photoelectrically converted. A printer control unit 1024 drives the exposure control unit 10 based on a video signal output from the image control unit 1023 to irradiate the photoconductor 11 with a light beam. Reference numeral 1028 denotes an operation unit, which is provided with an operation panel having keys for setting a mode necessary for image formation, a display, and the like. FIG. 3 is a block diagram showing in detail the configuration of the image signal control unit 1023. In the figure, 30 is an A / D converter, 31 is a black correction / white correction unit, 32 is an ND signal generation unit, 33 is a color detection unit, 34 is a scaling unit, 35 is an image processing unit, and 36 is density correction. Reference numeral 37 is a marker area detection unit, and reference numeral 38 is a histogram creation unit.

Next, the operation of the above configuration will be described. The analog image signal converted into the RGB electrical signal by the image reader control unit 1022 is the A / D converter 30.
Are converted into digital signals by. Then, after the black level correction / white level correction unit 31 performs the black level correction and the white level correction (shading correction), the ND signal generation unit 32.
The RGB signals are input to the color detection unit 33. ND
In the signal generation unit 32, the RGB signals are added to obtain 1/3.
And the luminance signal Dout is output.

Dout = (Rin + Gin + Bin) / 3 In the color detecting section 33, for example, red, depending on the RGB signal ratio,
It is classified into green, blue, pink of line markers, yellow, die dye, white, and black, and output as a 3-bit color signal Cout. The luminance signal Dout and the chrominance signal Cout are subjected to scaling in the main scanning direction (CCD line direction) or image moving processing by the scaling section 34 and input to the image processing section 35.

In the image processing section 35, a patterning process for converting the color information into a single color pattern, a masking process, a trimming process, a black-and-white inversion process, etc. are carried out in the image processing section 35. Thereafter, the density correction unit 36 performs brightness-density conversion and density correction in the printer, and the printer control unit 10 of the laser printer.
Sent to 24. ND signal generator 32 and color detector 33
The histogram creating unit 38 creates a histogram of the brightness signal Dout and the color signal Cout output from the brightness signal. Color signal information is added to this histogram as needed.

Further, the color signal Cout is the marker area detection unit 3
7, the signal of the area designated by the marker is detected on the original to obtain the area of the marker, which is sent to the image processing section 35 as a processing area signal, and processing such as black and white inversion inside and outside the area and shading is executed. FIG. 4 is a block diagram showing the configuration of the histogram creation unit 38. The whole is HSYNC, HVALi
It is controlled by an internal timing generator based on the synchronizing signals of D and CLK. Further, it can be controlled by a signal from the CPU.

FIG. 5 shows the synchronizing signal HSYNC and the operating state of the histogram creating section 38. Control signal C from CPU
PAL is synchronized by HSYNC to produce the TSEL signal. The luminance signal Dout from the ND signal generator 32 is written in the memory described later while the TSEL signal is at the L level. While the TSEL signal is at the H level, the CPU reads the contents of the memory and creates a histogram for one line in the RAM in the CPU.

In FIG. 4, reference numeral 50 denotes a writable memory such as a RAM having a capacity capable of storing one line of image information read by the image reader 22.
Reference numeral 51 is an output controllable buffer, and when the TSEL signal is at L level, the luminance signal Dout from the ND signal generator 32 is sent to the data input of the memory 50. Data selectors 52 and 53 are control signals (address, / OE, /) generated by the timing generator 54 by the TSEL signal.
WR, / CS) and control signals (address bus, / MRD, / MWR, / MCS) in the CPU are selected and given to the memory 50.

Reference numeral 54 is a timing generator, which is CLK, HV.
A control signal is generated from the synchronizing signals of ALID and HSYNC.
Reference numeral 55 is an output controllable buffer, which is a negative logic input NAN.
/ TSEL signal and / M input to D gate 57
The output is controlled by the WR signal. When the NAND gate 57 becomes L level, the data from the CPU data bus is sent to the data input of the memory 50. Reference numeral 56 is an output controllable buffer, which is input to the negative logic input NAND 58 /
The output is controlled by the MCS and / MRD signals. When the NAND gate 58 is at L level, the buffer 56 sends the data read from the memory 50 to the CPU data bus. 59
Is a D-type flip-flop, which produces a TSEL signal by synchronizing the control signal CPAL from the CPU with the 1-line synchronization signal HSYNC.

FIG. 6 shows timings at the time of writing and reading in the memory 50 inside the histogram creating section 38. 5, (a) shows the memory writing timing during the writing period of the luminance signal in the memory 50 in FIG. 5, which is created by the timing generating section 54. The address counter (not shown) in the timing generator 54 is initialized by HSYNC.
The RS signal becomes zero. The address counter is an up counter and counts CLK, which is a synchronizing signal of one pixel of image information, when the HVALID signal is at the H level, and ADRS
Generate a signal. Memory write signal / W accordingly
When R rises from the L level to the H level, the luminance signal is written in a predetermined address ADRS.

FIG. 5B shows the timing of reading from the memory 50 by the CPU in FIG. 5 and the memory reading timing from the CPU during the histogram creation period. When the memory selection signal / MCS from the CPU is at L level, reading from the memory 50 is permitted. C
The address signal output from the PU to the address bus is given to the address input terminal of the memory 50, and when the memory read signal / MRD of the CPU is at the L level, the memory contents are read and the data bus of the CPU is read. Is output.

FIG. 6A, which is provided to the memory 50,
The timing signal shown in (b) is selected by the TSEL signal and given. FIG. 9 shows a flowchart of the AE processing in this embodiment. First, a histogram is created (step 91), then characteristic points of the histogram are detected (step 92), and a conversion table is created based on this data (step 93). Finally, the γ table is created including this conversion table (step 9
4) It is written in the density correction unit 36 of the image signal control unit 23. Hereinafter, the details of each step will be described in order. (Step 1: Creation of Histogram) Creation of the histogram is performed in the following order.

Prior to reading a document, a luminance signal is input and a prescan (preliminary scan) is performed to create a histogram. Although all pixels may be input as the sampling of the luminance signal, sampling is performed by thinning out so that the characteristics of the original histogram are not disturbed. For example, about 1 mm. Input of luminance signal for one line All pixel data for one line is written in the memory 50 while the TSEL signal in FIG. When the TSEL signal is L level, the output of the buffer 51 is enabled and the luminance signal Dout from the ND signal generator 32 is stored in the memory 5.
Given to 0. Further, in the data selectors 52 and 53, the select S becomes L level, the A input is selected, and the control signal (address, / OE,
/ WR, / CS) is provided to the memory 50. The write timing is as shown in FIG. Reading of Memory by CPU In FIG. 5, the memory contents written during the period when the TSEL signal is H are read by the CPU. TSEL signal is CPU
The CPU reads the data for one line immediately before the TSEL signal goes to the H level from the memory. When the TSEL signal is at the H level, the output of the buffer 51, which is disabled, becomes high impedance. Further, in the data selectors 52 and 53, the select S becomes H level, the B input is selected, and the control signal (CPU address, / MR
D, / MWR, / MCS) is provided to the memory 50.
Also, the buffer 56 is / MCS and / MRD from the CPU.
When the signals become L level at the same time, the output is enabled and the data read from the memory 50 is output to the data bus of the CPU. Buffer 55 is / TSEL and / MW
When R is L level at the same time, output is enabled and CPU
Data is sent to the memory 50. Here, if the normal reading resolution is 400 dots / inch, 1 mm is 16 dots, so data may be read from the CPU at every 16 addresses (main scanning direction). For example, the address is changed to 1,17,33,49,65. The read timing is as shown in FIG. 6 (b). Creation of Histogram The histogram is created by adding the frequencies of the brightness signals read from the memory 50 for each same level. 1
Process sampling data for lines and CP the result
It is stored in the memory 50 inside the U.

In the present embodiment, since the luminance signal has 8 bits, 0 to 255 levels are added. Also,
Assuming that one level is represented by 16 bits, the maximum frequency is about 65,000, and this data can be stored. That is, a memory capacity of 256 words (512 bytes) is required to store the histogram data. , Are repeated only within a predetermined range.

Since the sampling interval is 1 mm even in the sub-scanning direction, the reading resolution is 400 dpi / i.
In the case of nch, it is sufficient to write the luminance signal in the memory every 16 lines. Since this time is determined by the control of the CPAL signal from the CPU, the CPAL signal is set to the H level and the CPAL signal is set to the L level after the histogram data for one line is created every time corresponding to the time of 16 lines.

FIGS. 7 and 8 show the relationship between the sampling and histogram creation ranges for a document. FIG. 7 shows the histogram creation range. When the number of bits of the memory for storing the histogram is 16 bits by sampling every 1 mm, the maximum frequency of about 65,000 can be stored, so that the histogram creation range is A4 size (210 mm × 297 mm).

FIG. 8 shows sampling intervals. Data is sampled every 16 dots in the main scanning direction and every 16 lines in the sub scanning direction. Here, the prescan (preliminary scanning) speed is the same as the normal reading speed (equal magnification), so the sampled data corresponds to one pixel for reading. (Step 2: Detection of characteristic points of histogram) By repeating the above processing, a histogram as shown in FIG. 11 is created. FIG. 11 is a diagram showing a histogram of a normal document and a conversion table. This is a histogram that is considered to be the most common in ordinary manuscripts, and the manuscript has a background (called the background) of almost the same density in a wide range, and characters and the like are written on it in a darker density than the background. . The horizontal axis represents the signal level, the left corresponds to 0 level (dark), and the right corresponds to 255 level (bright). The vertical axis represents the frequency, which is usually considered as a percentage (%) of the total frequency.

In order to analyze the shape of the histogram in detail, all peaks of the histogram are obtained. The outline of how to find the peak is to check sequentially from 0 level to 255 level, and when the frequency of the checked level is equal to or higher than the peak judgment reference value YLIM, and when this frequency is higher than the frequency of the preceding and following levels, the peak Recognize that. YLIM is, for example, about 0.03 of the total frequency.

The characteristic points of the histogram and the preset data are shown below. ILIM… Threshold of dark and light areas (eg 130) FLIM… Limit value of the level to be emphasized (eg 10) FMIN… Minimum frequency required to limit emphasis (eg 0.1% of the total) Idark… Darkest level of signal level Ilight… Brightest level of signal level lpeak… Darkest peak in dark area Jlevel… Darkest level recognized as background area rpeak… Inside bright area of bright area For the detection of the darkest level Idark, the frequency from 0 level to 255 level is checked in order, and the judgment reference frequency is LJU first.
Adopt a frequency level that exceeds G. The judgment reference frequency LJUG eliminates a judgment error due to noise or the like when the histogram is created, and is set to about 0.01% of the whole numerical value. For example, if the total frequency is 65000, L
JUG is 65, and a level having a frequency of 65 or more is detected.

Similarly for the brightest level llight, the frequencies from the 255th level to the 0th level are checked, and the level with the frequency exceeding LJUG is adopted first. lpeak is the darkest peak that appears in the bright area. Among peaks appearing in the bright part, rpeak is the maximum nth brightest peak level, assuming that it is recognized as, for example, the nth background peak. Jlevel is rpeak minus the offset value. By using this offset value (for example, 16), a portion darker than rpeak also becomes a background region, and the background is easily skipped. Further, when there is no peak in the bright part, Jlevel is obtained by subtracting the offset value from Ilight. (Step 3: Creation of conversion table) The conversion table is for obtaining the corrected output luminance signal from the input signal. When the input level is I in and the output level is I out, it is expressed by the following equation and becomes as shown in FIG. .

When Iin <black, Iout = 0 black ≦ Iin ≦ white, Iout = (255 / (white-black)) * (Iin-black) Iin> white, Iout = 255 Next, How to find black and white in the formula is described.

By applying the above formula with white = Jlevel black = Idark, it is possible to create a conversion table for emphasizing the character part by skipping the background as shown in FIG. 11 (if black = lpeak, it is possible to further embody the conversion table). ). At this time, if lpeak is equal to or less than FLIM and the frequency of FLIM level is equal to or higher than FMIL as shown in FIG. 12, it is considered that the document has a large proportion of a dark information portion, so black = 0 is set. FIG. 12 is a diagram showing a histogram and a conversion table of a document which is considered to be a document having many dark information portions. By doing so, the inclination of the conversion table becomes small, and it serves as a limiter for preventing the document from becoming darker than necessary. (Step 4: Creation of γ Table) Based on the conversion table obtained in the processing of steps 1 to 3 above, the final γ
Create a table.

The density correction unit 36 in FIG. 3 performs density conversion and gradation correction for correcting the gradation of the printer using an LUT (look-up table). First, as a density conversion process, the read luminance signal is converted into a density signal, which is generally called log conversion. The log conversion table is calculated from the following equation. Dout = -255 / DMAX * LOG (Din / 255) Next, the gradation correction table will be described. The gradation correction table is for correcting the gradation characteristics of the printer. For example, the gradation characteristics of the electrophotographic printer are shown in FIG. The characteristics of the correction table for this are shown in FIG.

Correction data = gradation correction (-255 / D _max * LO
It can be obtained from a formula such as G (Din / 255). The conversion table for density conversion and gradation correction is stored as a table in, for example, a ROM in the CPU, and optimum data is selected. Furthermore A
The final table is created by combining the conversion tables of the luminance signals obtained by the E processing. These processes are CPU
The program is done.

The density correction unit 36 is composed of a writable storage element such as a RAM, and the obtained γ table data is written from the CPU. This data is calculated and written in the density correction unit 36 each time the original is replaced. As described above, according to the first embodiment, even when a document having a large proportion of dark information portions is AE processed,
Since the copy image having the optimum density is obtained, for example, even when the generational copy is repeated, it is possible to obtain a high-quality copy that is not emphasized too deeply. <Example 2> Instead of the method of obtaining black used in Example 1, the darkest level value becomes the same level value after conversion,
Moreover, in order to skip the bright part from Jlevel, the black is calculated as follows.

Black = idark * (255-Jlevel) / (255-idark) By doing this, it is possible to prevent the document from becoming dark while maintaining the darkest level as shown in FIG. FIG. 13 is a diagram showing a histogram and a conversion table of a document which is considered to be a document having many dark information portions in the second embodiment. <Embodiment 3> Instead of the method used in Embodiments 1 and 2 to determine whether or not a document has a large amount of dark information, a level darker than lpeak is added to the total frequency of all levels in the histogram. If the total ratio of the frequencies of the histogram is above a certain value, it may be determined that the document has many dark information portions.

The present invention may be applied to a system composed of a plurality of devices or an apparatus composed of a single device. Further, it goes without saying that the present invention can be applied to the case where it is achieved by supplying a program to a system or an apparatus.

[0038]

As described above, according to the present invention,
Even when an original having a large proportion of dark information portions is subjected to AE processing, a copy image with optimum density can be obtained.
Even when the generational copying is repeated, it is possible to obtain a high-quality copy that is not over-emphasized.

[Brief description of drawings]

FIG. 1 is a sectional view showing the structure of an image processing apparatus according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a COUNT of the controller unit of FIG.

FIG. 3 is a block diagram showing in detail a configuration of an image signal control unit in FIG.

FIG. 4 is a block diagram showing a configuration of a histogram creation unit 38.

FIG. 5 is a timing chart showing an operation state when creating a histogram.

FIG. 6 is a timing chart showing the read / write timing of the internal memory of the histogram creation unit.

FIG. 7 is a diagram illustrating a histogram creation range when creating a histogram.

FIG. 8 is a diagram illustrating a sampling interval when creating a histogram.

FIG. 9 is a flowchart illustrating an operation procedure of AE processing according to the first exemplary embodiment.

FIG. 10 is a diagram showing an example of a conversion table.

FIG. 11 is a diagram showing an example of a histogram of a normal document and a conversion table.

FIG. 12 is a diagram showing an example of a histogram and a conversion table of a document considered to be a document having many dark information portions.

FIG. 13 is a diagram showing an example of a histogram and a conversion table of a document which is considered to be a document having many dark information portions in the second embodiment.

FIG. 14 is a diagram showing a gradation correction table.

[Explanation of symbols]

 1 Document Feeding Device 2 Document Platen Glass Surface 3 Lamp 5 Scanning Mirror 8 Lens 9 Image Sensor Section 10 Exposure Control Section 11 Photosensitive Body 12, 13 Developing Device 14, 15 Transferred Paper Loading Section 16 Transfer Separation Charger 17 Fixing Section 18 Paper ejection roller 19-direction flapper 20 Tray 22 Reader 24 Intermediate tray 1021 Automatic document feeder control unit 1022 Image reader control unit 1023 Image control unit 1024 Printer control unit 1025 CPU circuit unit 1026 ROM 1027 RAM 1028 Operation unit

Claims (3)

[Claims] 1. A creating unit that creates a histogram for each level of an electric signal obtained by reading an original, a detecting unit that detects a feature point based on the histogram created by the creating unit, and a detecting unit that detects the characteristic points. Image processing characterized by comprising: a determination unit that determines a document type based on the detected feature points; and a conversion unit that performs signal conversion processing according to the document type determined by the determination unit on the electrical signal. apparatus. 2. The image processing apparatus according to claim 1, wherein the conversion unit has a conversion table. 3. The image processing apparatus according to claim 1, wherein the characteristic points include a brightest level, a darkest level, and a level recognized as a peak.