JPS60187180A - Picture processing system - Google Patents

Abstract

PURPOSE:To enable to reproduce color or monochrome in a short time by providing a binary coding device and multiple coding device of picture signals separately, and reproducing binary signals and multiple signals on print from each processing device. CONSTITUTION:Each color of picture output from a CCD passes through a shading unit 104, is corrected by a gamma correcting unit 105 in gradation, is processed by a masking unit 109 and a UCR processing unit 119 in color and is added to a dither processing unit and multiple coding processing unit of later stage. Here, presence or absence of halftone is decided by a halftone decision circuit 127-3. When halftone is present, dither processing and multiple coding processing are made. When it is decided that there is no halftone, dither processing is omitted and binary coded simply by fixed threshold value level. Multiple coded or binary coded signals are reproduced and printed by a laser scan unit.

Description

[Detailed description of the invention] The present invention relates to an image processing system.

Conventionally, it is known that a color document image is optically read by a photosensor, converted into an electrical signal, processed, and printed out.

In this case, if it is desired to insert characters such as a date into the printed image, it is necessary to read the transparent sheet with the date and the like superimposed on the document when reading the document image.

This is extremely inconvenient.

On the other hand, it is possible to generate a character signal and synthesize it with the document image signal, but when reproducing the halftones of a document image that has halftones, it is necessary to reduce the resolution of the inserted characters. Sometimes I put it away.

Furthermore, when reproducing a document image containing a color image, depending on the undercolor, the inserted characters may become unclear.

Conventionally, it is known to perform binarization processing, such as a dither method, in order to reproduce the halftones of a document image.

This uses nXm pixels to reproduce n×m gradations.

Therefore, the resolution is reduced by 1/(n×m). Therefore, line drawings such as characters included in the document may become unclear, and there is also a risk that synthesized characters may become unclear.

Conventionally, light is applied to a document image on a photosensitive drum for electrophotography through an optical system, and a laser modulated by character signals is applied to the photosensitive drum on all four sides to synthesize the document image and the text image. has been reported. However, this is a complex configuration.

In addition, there is a conventional system that composes images by developing image data and image data to be composited in a memory, but it takes too much time to synthesize the read image data and other data.

Ordinary color copiers or laser beam color printers that use a photoreceptor create a yellow or blue toner image on the photoreceptor and transfer it during transfer, then create a magenta or green toner image and transfer it to the photoreceptor. Transferred to Gion,
Repeat the same operation four times for cyan or red, and then for black. Therefore, even if the original has only one color, operations for four colors are performed, resulting in time loss.

The present invention resides in an image processing system that overcomes the above drawbacks, and the present invention resides in an image processing system that makes it possible to add a character image to a color image without deteriorating the image quality. The present invention provides an image processing system that enables character images such as management data to be added in real time to an optically read image having halftones without deteriorating the image quality; The present invention provides an image processing system that performs halftone processing on a read image signal and outputs a composite digital image signal without impairing the resolution of character images to be added and character images to be added. The present invention provides an image processing system that enables adding a gearactor image to an arbitrary position without degrading the image quality to an image that has halftones. The present invention also resides in an image processing system that prevents the composite image from becoming difficult to see due to the image state of the composite area of the first image when the second image is composited with the first image, The present invention also resides in an image processing system that can print a read image by combining a read image with another image without taking the time required for image combination processing.

The present invention also provides a color image printing system that enables the reproduction of halftones in real time without impairing the resolution of color images having halftones. The case lies in image printing systems that enable color and monochrome reproduction in a short period of time.

FIG. 1 shows an example of an image processing system to which the present invention can be applied, and is a color system that can read color documents and reproduce color images. A document 1 is placed on a transparent plate 2 of a document table and fixed by a document mat 3. The photosensitive drum 24 and the transfer drum 53 rotate in the direction of the arrow to perform a color process. 12 is a dichroic mirror for spectroscopy, 14.16, 18 is a color signal B that senses the spectra,
This is a CCD that generates G and R signals.

The lamp 8 and mirrors 9 and 10 reciprocate to scan the original 1, and simultaneously output color signals B, G, and R from each CCD to create a Y signal for reproduction, and then reciprocate again to output an M signal. , repeat the above scanning four times to sequentially scan Y, M. C, BK
Signals are formed, and the laser is controlled by these signals to sequentially form latent images of each color on the drum 24. The latent colors of each color are sequentially developed by the developing devices 36 to 39, and then transferred to the transfer drum 53.
The developed image is transferred to the upper layer t, and the drum 53 rotates four times to repeatedly transfer the image to the upper layer, thereby obtaining a full-color copy having intermediate tones and intermediate colors.

The optical system emits light from an illumination lamp 5.6 and a reflector 7.
.
1, and then passes through 12 trichrome filters. Here, the separated light of the blue wavelength is separated into light of the natural wavelength, light of the green wavelength, and light of the red wavelength. Among the separated lights, the separated light of the blue wavelength is received by the solid-state image sensor 14 with the blue filter 13. . Similarly, the light at the edge wavelength bends through the green filter 15,
The light is received by the solid-state image sensor 16.

The light with the red wavelength passes through the red filter 17 and is received by the solid-state image sensor 18 . That is, the original 3 is illuminated by the illumination lamp 5.
．． The optical path length is maintained by a moving reflecting mirror 9 that moves together with the moving reflecting mirror 9 and a moving reflecting mirror 10 that moves in the same direction at half the moving speed of the moving reflecting mirror 9. dichroic filter 12
The scanned and color-separated image light is imaged on solid-state image sensors 14, 18, and 16 of each color. The outputs of the solid-state image sensors 14, 16, and 18 are signal-processed through an image processing section 27, which will be described later, and are output as optical output from the semiconductor laser 21 to the polygon mirror 22, which irradiates the photoreceptor.

Since the polygon mirror 22 is rotated by the scanner motor 23, the laser beam is scanned perpendicularly to the rotation direction of the photosensitive drum 24. Further, there is a photosensor 64 at a position just before the laser beam starts running on the drum, and when the laser beam hits this, a BD (beam detection) signal is generated. This is used to determine the timing for outputting one line of M by the HD laser, and also determines the output timing for one line of image data in the line memory.

The photosensitive drum 24 is negatively charged by a negative charger 25 to which a negative high voltage current is supplied from a high voltage source 25. Next, when reaching the exposure section 26, the transparent plate 2 of the document table is exposed.
The original 1 above is illuminated by an illumination lamp 5.6, passes through movable reflection mirrors 9 and 10 and a lens 11, and reaches a dichroic filter 12, where it is decomposed by a blue filter 13, a green filter 15, and a red filter 17, and solid-state imaging is performed. Images are formed on elements (CCU) 14, 16, and 18.

The image output from these CCDs passes through a shading unit 104 for each color in the image processing circuit shown in FIG. The laser beam is processed, subjected to halftone reproduction processing by a dither processing unit 124 and a multivalue processing unit 125, and then outputted from a laser driver unit 126 to a laser 21, and the laser light is imaged on a photosensitive drum 24. There, an electrostatic latent image is formed, and a four-color developer 36.37.3
8.39 and is developed. Here, three colors are separated in one exposure scan and each of the above processes is performed, but each B, G, R,
The output of the UCR corresponding to BK is sequentially selected for each scan of B, G, R, and black BK. 1 in the drawing method processing unit 27 by the timing signal (E signal to each gate corresponding to each UCR output) from the main body control unit 69.
Select color separation light signal. In this case, the corresponding developing device can be selected. The developer selected thereupon performs powder development using a magnetic blade method, and the electrostatic latent image is visualized.

After that, a ghost lamp 40 is used to erase the electrostatic image.
Then, the electrostatic latent image is erased by being negatively charged by the negative post electrode 41 supplied from the negative voltage power source 25.

Next, the upper and lower cassettes 43, 4 selected from the operation section 45
The copy paper 48 fed from one cassette of 4 by the rotation of the paper feed rollers 46 and 47 passes through the upper and lower first registration rollers 49 and 50, and then passes through the second registration roller 52 from the conveyance roller 51. Then, it is wound around the transfer drum 53. Therefore, the toner on the photosensitive drum 24 is transferred to the transfer electrode 5.
4 onto copy paper 48. After the transfer has been completed, the photosensitive drum 24 is charged with a high voltage generator 25, and the copy paper 48 is neutralized by a discharge electrode 55 supplied with a high voltage.

In this way, the printing operation is started almost simultaneously with document scanning, and the printing time is short.

In the case of a normal color original, the above operation is repeated four times for four colors, and the transfer drum is rotated four times to overlap each color. If the original has only one color, black, when one optical movement is completed as described later, if it is detected that the original has only one color black, the process such as G and R scanning, phenomenon, transfer, etc. jump and start copying the black image. In other words, in the case of a color original, the operation time for four colors is required, but in the case of an original of only one color, black, the operation time can be shortened to the operation time for two colors or one color.

The copy paper that has been copied twice or four times is peeled off from the gripper 57, attracted onto the belt 59 by the conveyance fan 58, guided to the fixing section 60, fixed thereon, and sent out of the machine.

2-1 to 2-3 and 3-1.3-2 show image processing circuit diagrams.

The common < ty) part will be explained below. The source of the document separated into three colors by the dichroic filter 12 is the CCD 1.
4,16.18, the output is C for each color.
The data is amplified by CD boards 101, 102, and 103, A/D converted, and 8 bits are sent in parallel to the next shading unit 104 as one pixel data. In order to make the output data for each bit of the CCD equal when the amount of incident light on the CCD is the same (when it is white), the CCD 14 .
It is the shader unit 104 that performs correction to eliminate the variation in 16.18. This is a configuration of a RAM and an arithmetic unit, and the previous 8-bit data is the address of the RAM, and the RAM is accessed using that data, and an appropriate output is produced from the arithmetic unit.

Next, the γ correction unit 105 linearizes the gradation characteristics between input and output, and there is one for each color, and the optimum γ curve can be selected by switching the ROM pattern using switches 106, 107, and 108. I have to. The reason why the upper 6 bits of the 8-bit data are processed and used as output data is that processing in a normal level area is sufficient.

Next, the masking processing unit 109 performs each B, G, R
Color correction is performed by simultaneously processing the signals and changing the mixing ratio of each color component. This allows signal correction to match the color tone of the developed toner.

This operation is performed by a coefficient multiplication ROM and an addition/subtraction ROM.

The mixing ratio of each color is determined by changing the values (coefficients) of switches 110 to 118. Note that the reason why the calculated value is set to the upper 4 bits is because it is limited to a significant area level. Each ROM is addressed by input data and outputs data as a result of operation. Each ROM simultaneously outputs each color.

Next, in the UCR processing unit 119, each converter COMP logically compares each output signal to each gate M.
The minimum value signals of B, G, and R are determined by the logic of IN. The value obtained by multiplying the steep A/H output from the MiN by an arbitrary coefficient depending on the value of the switch 120 is set as a black level signal. This value, which becomes the output of UCR BK, is subtracted from each color signal in each UCR circuit.This allows black to be processed separately, and the black signal can be removed from B, G, and R, resulting in clear color reproduction. The signals are sent to the select signals 121, 122, 12 from the control section 69 in the gate circuit.
3 in synchronization with the color output timing, and a signal of one of these colors is sent to the dither processing unit 124. Each color signal is sequentially sent to 124 for each document scan.

In the dither processing unit 124, each signal is a deep signal, for example, a 6-bit signal per pixel, and the dither ROM is accessed so as to refer to a table, and the input signal is converted into a binary ratio signal of 0 or 1 per pixel. Alternatively, as shown in FIG. 3, the input data is compared pixel by pixel by comparators 138 to 140 with the data in the dither ROMs 135 to 137, which store the data of a dither pattern in a 4 x 4 matrix, for example, to obtain data of one bit per pixel. Convert to 1 or 0 digital signal, 4×4
Expressing intermediate tones using pixels. This makes it easier to modulate the laser. The patterns of the dither ROMs 135 to 137 are 9. as shown in Figure 3-1. -9, you can select carefully. Further, the dither processing can be omitted using the selector as shown in FIG. 3-2.

Next, send this signal to the read-write line memory,
Pixels for each line of the document and one line of reprint are stored and output in synchronization with UB.

Thereafter, the multi-value processing unit 125 performs multi-value processing, and the Rayleigh driver unit 126 drives the laser 21. The dither processing unit has a ROM with a low threshold level array, a ROM with a high threshold level, and an intermediate ROM,
The input signal is compared with these ROM outputs at the same time, and the output from each comparator is put into the line memory 141 and latched, and one pixel is divided into three equal parts. In other words, one pixel data output by R0M1 to 3 is separated by pulses with different widths of φ1 to φ4 (FIG. 8) at the AND gate 142, and outputted by the OR gate 143 in correspondence with ROM and ~3 pulses, and then the OR gate 143 outputs riJ. Outputs 1 pixel data with different values. As a result, one pixel can be represented by modulating the beam in four ways with the output converted to 41 (K).Thereby, one pixel can represent the middle, -A, 0, 2nd, 3rd, etc.
The processing of the figure is X.

This is done in real time (real time) almost simultaneously with the input of Y and Z. In other words, printing can start almost at the same time as document scanning, and color printing does not take much time. By the way, halftone reproduction by binarization dither can reproduce 16 different gradations in the case of a 4×4 matrix. Therefore, there are 4 intermediate gradations due to pulse width modulation, so a total of 6
Four gradations can be reproduced.

On the other hand, a part of the output of each UCR ROM corresponding to B, G, and R is input to the black No. number determination circuit 127-1 shown in FIG. 2-2. Note that 127-1 is 1 via u, v, w in Figure 2-1.
He was placed at 27-2. 6BI is input here
These are the top 4 BiTs among T.

This is because color signals with low density are ignored, so 6Bi
You can leave it as T.

In the memory 128-1, θ is stored at address θθθ, and F is stored at all other addresses. Therefore, if there is no UCR signal color signal input to 127-1, θ is changed to F.
is output, latched by the latch circuit 129-1, and input to the hold circuit 13 (31) in synchronization with the clock.

This output S.

is determined by a program by the CPU of the control unit 69 and contributes to sequence control. This will be explained using the flowchart of FIG. 4-1.

This flow is programmed into the microcomputer of the control unit CPU (69 in FIG. 1), which first outputs the RESET signal Sr immediately before optical scanning for document scanning, and outputs Q1 to Q4 of the hold circuit 130-1. (Step 1).

If there is a color signal even once until the first optical scan is completed, the hold circuit 130-1 outputs an FF, and as a result, the output S0 of the OR gate 131' becomes H level.

The control circuit CPU checks this signal (4) immediately after completing the optical scan (3), and if it is an H level signal, performs the normal full color copying operation (routine 5).

If the gate 131 remains at the L level, it is determined that the document is completely black, and the B. A sequence selection signal is output in order to omit the G and R processing and complete the process with only the black copying operation (6). Therefore, a sequence controller (not shown) moves only the black developing device to form and develop a latent image, and when the transfer drum rotates once, the gripper 57 is released and the transfer paper is discharged.

In this case, when scanning and developing are performed in the order of B, G, R, and black BK, the process rotation for U and R is omitted, so the processing time for two colors is sufficient.

Further, since the original is preliminarily scanned and the color can be determined at the end, the development time for one color of black is required.

Also, if black, B, G, and R are processed in this order, the formation of a black latent image is light at the time of color judgment (at the end of scanning), so by blocking the subsequent processing, one color can be processed. The processing time is only 1 minute.

Further, even if it is determined that the input signal is a monochromatic image of any one of B, G, R, Y, M, and C other than black, sequence processing and signal processing can be omitted in the same way.

This determination can be made by independently monitoring the outputs B, G, and R of the UCR and detecting that there is almost no output for any of the colors (described later).

127-2 in Figure 2-1 is for single-color determination; B, G,
A part of the output of the ROM of each R UCR is input to the monochrome signal determination circuit 127-2. 6BI is input here
These are the top 4 BiTs among T.

This is because color signals with low density are ignored, so 6Bi
You can leave it as T.

OR gate 128-2 receives the UC input to 127-2.
If there is no color signal in the R signal, L is output, and if there is, H is output, this is latched by latch circuit 129-2, and input to hold circuit 130-2 in synchronization with the clock. This output is subjected to software judgment by the control unit CPU and is applied to sequence control. This will be explained using the flowchart shown in FIG. 4-2.

This flow is programmed into the microcomputer of the control unit CPU (69 in Figure 1), and first, just before optical scanning for document scanning, a RESET signal is output, and the outputs O1 to q of the hold circuit 130-2 are reset. (Step 1
00). Next, perform an optical preliminary scan and expose the original (
Step 101).

The optical preliminary scan is completed (step 102), and if there is a color signal even at 11 times in the trench, the hold circuit 130-
The H signal is output to the output terminal corresponding to the two color signals. For example, in the case of a B color original, Q1 becomes H and Q2 to Q4 become L.

The control circuit (69 in FIG. 1) checks this signal just before starting the optical scan for reproduction of each color (step 10).
3) If the signal is H, reproduction processing for that color is performed (step 104). For example, when the original is of blue B color, a sequence selection signal is output to omit the green G, red R, and black reproduction processing and perform only the B reproduction processing. Therefore, the sequence controller (not shown) is used only for the blue developer. A blue latent image is formed, the latent image is developed, and the blue image is transferred to the transfer paper of the transfer drum. When one rotation for this purpose is completed, the gripper 57 is released and the transfer paper is discharged.

In this case, scan with B, G, R, and black values, tB,
When processing images, the process rotation for G, R, and black is omitted, and the processing time for one color is sufficient.

In this case, the main scan may be performed in steps 101 and 102, and when a single blue image is found at the end of the scan, the formation of the blue latent image has been completed, so by blocking the subsequent processing, one color image can be scanned. processing time.

In the case of a document with only two colors, for example, in the case of an original with only B and G, the reproduction processing of red R and black is similarly omitted.

In the case of full color, Q1 to Q4 are all H, so all steps 104 to 107 are executed.

In addition, in the case of a type in which each color is reproduced on four photosensitive drums, a resist is taken on a sheet of paper, and the process is sequentially transferred, once the process for a specific color is completed, the feeding speed can be increased and the time can be shortened.

The present invention is effective even if the input signals B, G, and R in FIG. 2 are from the host computer, and the
, Z can be switched between the host and the CCD reader as necessary.

In this case, if a monochrome command signal or a monochrome command signal is attached to the beginning of the transmission signal from the host, this can be determined and considered as a black image or monochrome. Even if the printer is of the 1-pixel, 4-dot type, it is effective in reducing the time required for monochrome processes, single-color processes, and full-color processes. Furthermore, since the full-color signal processing step can be omitted, black and other single-color images can be reproduced with good quality.

Furthermore, if a monochromatic image (one color each of B, G, R, black, etc.) is determined, it can be recognized as a character image and outputted with the dither unit off, so that the resolution is not impaired. In addition, in this case, in order to reproduce some intermediate tones by using the pulse width modulation of the laser drive signal using the four values mentioned above in Fig. 3-1,
Buzzer ROMI-a for static area 1 shift (3rd level)
The above-mentioned pulse width modulation or 41ff modulation can be performed by setting the signals a, .about.a.

Furthermore, by determining the output from the hold circuit 130 for each l line and applying a reset, the single color "r(
It can be used as a selector for signal processing such as sequential buzzers.

Further, it is possible to make a determination every few pixels, and to perform the above-mentioned partial selection control with accurate synchronization.

As described above, by determining a specific color such as black in a color system such as a color abdominal copying machine, it is possible to reduce the copying time from approximately % to K for a monochrome original. Further, signal processing can be performed to improve the resolution of characters and the like. Since unnecessary color signal processing is not performed, the quality of a specific color will not deteriorate.
Since processes such as transfer and cleaning are prohibited, unnecessary fatigue is not caused and the life of the machine can be extended.

In addition, since the image processing is selectively controlled by color determination, the image quality is not impaired. In addition, black judgment is determined by whether all the outputs after each color UCR processing are at the peak level or not, or the maximum value of the input B, G, R, signal (signal after r conversion) or Y after masking supplementation
, M, C, (B, G.

R) Determination is made based on whether the minimum value of the signal exceeds a predetermined level. Furthermore, it is also possible to determine that the image is a line image and to turn off dither processing so as not to impair resolution. In addition, it is also possible to further determine whether the black is a line drawing or a gradation based on the black determination, and to perform dither processing for the latter black with a matrix pattern different from that for color.

Next, with reference to FIGS. 2-3 and 3-2, the determination of halftones and the control of halftone processing based on the judgment will be described.

Other signals branched after the masking process are sent to the halftone judgment circuit 1.
Sent to 27-3. Memory 128-3128-4.1
Each of 28-5 has θ from address 00 to address 02, and 1
Bal is stored in advance from address 0 to address 2E, and θ is stored in advance from address 2F to address 3F. This is because when the bit to which l is set among the 6 bits of data to 127 is in the middle, l is output to indicate the presence of an intermediate tone. Therefore, a memory that can be accessed with 6 bits is provided, and
Addresses that can be accessed with bits are 64 from θθ to 3F.
Therefore, address this memory with 6-bit data and set the data level to upper 11''1.Divide into middle and lower to determine the presence or absence of halftone for each color.Memory Bx2
Depending on the strength of the light input manually to the 6BIT input to 8-1c (No. H is CCI), the difference between θθ when there is a lot of light (the density of the original is low) and 3F-J when there is little light (the density of the original is high)
The number changes to 64 letters.

As an example, when the input data to 127 is a signal from θθ to θF, it is a low concentration, and when it is a signal from 10 to 2E, it is an intermediate concentration, and from 2F to 3F.
When the signal is, it is considered a high concentration signal.

For example, when an intermediate density signal is input to memory B, it outputs 1, and otherwise outputs θ. This is the latch circuit 129-
3 to latch and hold circuit 13 in synchronization with the pixel clock.
Enter 0-3. This hold circuit holds data until a reset signal is input. Therefore, when data between 1θ and 2E exists, the OR gate 131 outputs 1 (H). When the microcomputer of the control circuit 69 determines this, it performs the dithering process shown in FIG. 3-2.
If H) is not output, it is determined that the dither processing is stopped and binarization is performed at the fixed 2-pin threshold level.

This operation will be explained using the flowchart of FIG. This flow is executed by the microcomputer of the control unit 69.
It is programmed into M.

Immediately before optical scanning, a RESET signal Sr is output (step 200) to reset the outputs Q1 and Q2 of the hold circuit. The operation of the mirror system is started in order to start the first optical scan. Intermediate # even once during that scan
If there is a signal, the hold circuit 130-3 latches 1 and outputs it, and the output S1 of the OR gate 131-3 becomes "H". The control circuit 69 (FIG. 1) receives this signal.

When the completion of the optical scan is identified (step 202), a check is made to confirm whether the optical scan is completed (step 203). If the signal is "H", the control circuit 69 outputs a switching signal Sl=0 to the selector, and the selector 141- 143 in Dili ROM1
Switch to the 35-137 side (step 204) and dither processing! make them do If 8. If the signal is L IT times Sr
Output = 1 and set selectors 141 to 143 to fixed data 1F
is switched to the generation circuit side (step 205), and the dither processing is omitted.

Takumi does not apply dither to character images such as letters that do not have halftones, so resolution is not compromised. Also, if halftones are determined for all color components and even one component has halftones,
The quality of color reproduction is good because it is dithered.

As this first optical scan, the M image is not scanned directly after forming a reconstructed image, but is scanned at a high level (without forming a reconstructed image), and the selector is controlled in advance to select dither and fixed values. You can also keep it under control.

Further, the range to be determined as intermediate density is 1θ to 2 in the memory 128.
Not only the parts corresponding to EK, but also l and θ between θθ and 3rd
By switching the memory of the table that changes the memory pattern, you can freely decide.

Circuits such as those shown in FIG. 5 may be added to or replaced with x, y, and z in FIGS. 2-3. This is done by inputting the BD signal from the beam detector 64 shown in FIG.
If so, a RESET signal is output to the hold circuit 130 at 4.

In that case, by directly inputting the signal S of the OR gate 131 to the switching signal 144 as the switching signal Sl shown in FIG.
If each line has an intermediate density portion, dither processing can be performed sequentially. Therefore, it is possible to selectively control the dither threshold value and the fixed threshold value for each area of every four lines. This second Q match
A buffer capable of storing four lines of gate output data shown in the figure is provided, and the output of this buffer is input to a dither circuit for binarization processing using dither or fixed threshold values. This allows you to distinguish between midtone areas and text areas while scanning and printing documents. In addition, in order to perform dither processing on other lines at high speed while performing halftone determination, two 4-line buffers may be installed in parallel and used for both determination and processing.

In this way, with a simple circuit, it is possible to perform dither processing on originals or their parts with intermediate density to obtain high-gradation prints, and to obtain high-resolution prints without performing dither processing on documents without intermediate density. can.

In addition, B in Figure 2. Even if the G and R input signals are from the host computer, the present invention is a single right-hand drive, and it is possible to switch between the host and the CCD reader at the connection points of X, Y, and Z as necessary. can. In this case, if a command signal without halftone is attached to the beginning of the transmission signal from the host, the selectors 132 to 134 can be controlled to determine this and memit the dither processing. Also, 1 pixel 4 bit type printer, thermal printer,
This example can also be applied to inkjet printers.

On the water side, when dither processing is omitted, the pulse width modulation of the laser drive signal is performed using the four values described above, so it is possible to reproduce some halftones, and low-level halftones (sometimes dither omit is done at the end) becomes possible to reproduce. It is also possible to produce halftones with a single pixel, allowing for accurate synchronization and the aforementioned selective production.

FIG. 6 is a circuit diagram for inserting letters, numbers, etc. into the above-mentioned color image.

200 is a code generation means that generates characters and numerical values as code data (for example, ASCII code), M is a buffer memory that stores the code data at the same time as code 46, and ADC1 is an address for storing and reading the memory. CG is the memory M,
M2 is a well-known character generator that outputs characters and numerical values as dot pattern image data based on the code data read from the CG.

In the backup memory that is stored at the same time as output, it is stored so that one pixel of image data corresponds to one dot from CG.
In other words, the dot pattern (bit pattern) for multiple characters and rectangular digits indicated by 0.1 is a set of each character and each numerical value, with a predetermined interval similar to the bit serial data of the reproduced image. Stored. ADC2 is an address counter that controls the address for writing and reading data in the memory M2. Moreover, it determines the readout start timing in synchronization with the processing progress of color image data. The synthesis position can be determined. 201 is a signal manual gap that presets the timing;
When the image processing according to the preset coordinates X, Y construction Figure 2 is reached, reading from the memory M2 is started, and characters are output in synchronization with the color reproduction output corresponding to the above position after dither processing, and then they are synthesized. be.

205 is a controller that outputs the CG dot output as H'' (black) when the output of the comparator 206 is L'' (white, halftone);
204 is an inverter that outputs L (white) when the output of the comparator is °'H'' (black).

The comparator 206 outputs 11'' when the output level of the black component shown in FIG. 2 is above a certain level L1, and outputs L'' when it is below a certain level. Therefore, when the image has a dark background color, a signal at the above level is output as the character image signal B in order to change the character image from the CG from a dark background color to white, and when the image has a bright tone, to make the character image sharp. This signal B is input to the OR gate 210 in FIG. 3-2, and is synthesized in an overlapping manner with the dithered image data. In this case, since the inserted characters are not dithered, there is no loss of resolution.

The R/W signal is a write/read signal, and the continuation signal of the memory M2 is applied to the black scan and flick processes in synchronization with the plaque processing process, and is outputted as a seed forming black as a - character. If the color image is a monochromatic blue image and you want to insert characters in red, perform the red processing step: When executed, the successive signals are processed in red so that the B signal is output only during that step. It is assigned to memory M in synchronization with .

The address counter ADO is used to determine the timing to start reading the memory dot (CLK) for image data processing.
) and lines are counted, and when the count value reaches Brisera LX, Y by 201, reading from the memory M is started in synchronization with the clocks CI and KK. The count of dots (pixels) is bits (=CLi()
, the 1 line end signal starts the busy count. Line count t, l An end signal obtained by counting bits for one line minus the beam detection signal in the laser scanner indicating the end of line scanning is started at 1σ when color data processing starts. The processing shown in FIG. 5.6 is also performed in real time. In other words, document scanning and printing can be performed almost simultaneously, while character separation and character composition can be performed.

As shown in Fig. 7, when the r1984J code is stored in the memory M1 by the key or transmission line attached to the vibrator shown in Fig. 1, the CG converts it into a dot pattern and prints the memo v.
Store in h. After this is completed, the aforementioned color data processing becomes possible (in this example, original scanning using an optical system becomes possible).

Unless a command indicating that there is no inserted data is input, document scanning Vi is prohibited until then. ) Color data processing starts, and when the number of dots and lines reaches the preset coordinates X, Y of 201 in step 4M of black scan (fourth document scan), memory M,
fy't, starts outputting, sequentially outputs the character B signal in synchronization with CI, enters the gate 21o in Figure 3-2, and synthesizes it with the data before multilevel processing after dither processing. A latent image of black text is formed on the drum, which is transferred and composited with the previous color image to form a color plate with text.

You get lint. Red as described above, if necessary.
It can be entered with the pond color letters such as blue.

However, when characters are added to a position where the background is partly a bright color (black), as shown in ■ in Figure 7, the background is determined by the signal person in the automatic fishing system as described above, and the white text is signaled. It is completely formed and output as signal B. This procedure relies on a circuit configuration with precise synchronization relationships for image processing to be achieved in real time. The preset data 201 is made possible by the code data transmitted from the copying machine key or strip shown in FIG. The memory ROM 200 may be a memory ROM storing a format having characters, lines, etc.

By the way, in cases where the ground color repeats in a short range, such as in a striped pattern, it may be rather unsightly to treat the white areas accordingly, so to eliminate this inconvenience, a delay circuit is provided at point W in Figure 6. Therefore, whitening can be performed only when the background in a predetermined range is dark.

By the way, when adding all characters in white, it is necessary to determine the readout timing of the memory M2 for each color so that the process can be performed in the color image processing process for each of the four colors at that time.

In FIG. 9, the composite image data after binarization processing is transmitted to another printer without being multivalued.

Selector 132 in the figure. Dither ROM135. Comparator 138. OR gate 21G, latch, line memory 1
41 is the same as in FIG. 3-2.

However, the dither ROM 135 has a dither pattern different from that shown in FIG. 3-2. It is selected by signal a. a is generated by a data transmission command from the key human power section of FIG. The dither pattern determined by a does not have a pattern exclusively for multilevel conversion, but is a pattern exclusively for binary conversion, and even without processing by dither ROMB6.137,
This pattern is designed to reproduce halftones.

Therefore, by transmitting only the composite data of the output of the comparator 138 and the character signal B, it is possible to sufficiently transmit and reproduce a halftone color composite image.

In FIG. 9, in FIG. 3-2, the dither R0M 135 is set as described above for the line memory 141 that stores data with a high density level.

A transmission circuit is added.

150 is a switch that changes the data sending for print/print and transmission, and is switched in the direction of the dotted line by the above transmission command (-g No. a). 151 is the number of times 1" in the image data repeats, ; 152 is a run length counter that counts the number of times the image data is encoded;
51.152 is a well-known compressor that compresses image data to reduce the number of bits. 0153 is a switch that is switched sequentially in synchronization with the document scan shown in Figure 1 and the end of encoding by 151.152. Due to the end of encoding of image data (controlled by No. 8 b).

154 is a memory for storing each color component data encoded in correspondence with the switch 153 or data of each color component sent; B, G, R, 2% each having a storage capacity of one page of a document; BK has four memories (-). 155 transmits the data of memo IJ154 to the transmission unit MU.
This switch is used to select whether to send the data to D or to the print section, and the switch changes to the dotted line when data is received.

MOD 156 is a well-known high frequency modulator for transmitting data over long distances. DEMOD 157 is a well-known high-frequency demodulator that extracts data from the high-frequency waves sent to it. 158
is a separator that determines the type of data sent and outputs it to line MH if it is an MH code, and outputs it to line As if it is an ASCII code (16, Atsushi code). or,
Even in the case of the MH code, it is determined whether or not it is a color B, G, R, or BK component, and each is output to the corresponding line. This separator is
Since command data indicating the type of data is attached to the beginning of the sent data, the output line is selected by determining this. ASCII code is character data such as letters, and is character data when halftone images and character images are sent serially at different times. Furthermore, when the code corresponding to MHH code B is sent, the code corresponding to G is sent, and the data of each color component is sent serially. Each MH data is stored in memory 154. 159-161 are M in Figure 7
1. CG, a character image generator similar to M2,
Bit dot character image data C is output from the character code by a character generator.

This character data C is processed by the OR gate 162.
It is synthesized in synchronization with the sent halftone data as shown in FIG. 7 and sent to the print section. 163 and 164 are well-known MH decoders and run length counters that convert sent MH code data into bit image data. Reference numeral 163 is a selector that determines whether the data to be sent to the print section is GA coded data or document scanned data, and is switched to the received data based on the received signal CK.

To explain the operation, in the transmission mode, first, during the first scan of the document, the color processing ID number corresponding to the blue color is subjected to binary rhi processing in the /izer circuit 124 and converted into data of 1 pixel and 1 bit. The composite data of this data and character data is input to a MH encoded open circuit consisting of a counter 151 and a decoder 152 via a switch 150, and is input to a maximum of 3
The data is converted into a 6-bit MHH code and stored in memory B of the memory 154 via a switch 153. The high frequency is then modulated by the modulator 156 using the data in memory B and transmitted. When the transmission of the composite data for blue is completed, the document is scanned a second time, color processed for red, binarized and composited as described above, and stored in memory. , to perform transmission.

As described above, the composite color data of the characters and the scanned color image is sequentially created in B, G, R, BK for each scan.
are sent in this order. In this case, images that do not have halftones, such as characters in the scanned image, are selected using the selector 1 as described above.
Since the signal is not dithered by 32, it can be transmitted while maintaining the resolution provided by the CCD. In addition, the data of Memo IJ 1.54 is sequentially filed by color on an optical disk, and multiple pages of the document 5. Around the time you can file the next file.

Note that if it is determined in Figure 2-1.2-2 that the document image has one component such as a black component, the determination (
Scanning of documents after the determination is made is prohibited. Therefore, only the composite data of the 11i data and the character data is stored in the memory 154 and transmitted.

In addition, the dither ROi\4135 to 137 is calculated for each color component (rC even if the dither pattern is different, so as not to impair the color quality.It is -%2-1.2-3 gate 1 in the figure).
Control signals to the 21-1230E boat 13. G
, it, and BK, and by selecting a dither pattern using a 2-bit code signal K indicating a color component.

Next, the received signal is processed by the data separator 158.
Data lines pc for each color component are sent separately and stored in a memory corresponding to each color.

This data is sent to the decoder via switch 155.

The code data is sent to a counter and converted into 7 real dot bit data. This bit data is combined with a character as necessary and sent to the selector 163, and stored in the line memory 141 for printing.After that, the radar print is performed in the same manner as described above.Additional character image C is generated by the key operation in the Taroki Hakuho 1 system shown in Figure 1. Also, the character image C to be added is a combination that is sent as an Assen code separately from the MH image data. , the code data is separated into an MH image by the separator 15B, and converted into a dot-bit image by the key Y lacta generator CG and V via the line A8.Since the character is transmitted as a code, the transmission efficiency is high and the time is shortened. This character data and the character data in Figure 7 are the management information of the 0 provisional fold, which includes the chapter information, date, and timing information created by word processing by key personnel. The output timing of the embedded information is preset in the address counter AD2 (FIG. 7).

The received 4If information represents halftones in 8 bits.
Separate +58i for +i+j element 8-bit element upper bit data
jb is determined by the command, sent to the buffer 170, and the upper 6 bits of data are inputted into the box (■) in FIG. 9, and the dither time #! 1124. Therefore, as described in 711, each dither 1 (, OM 135 to 1
37, the data is binarized and sent to each line memory 141 as bit-serial image data. In this case, the selector 163 is set in the direction of sending the scan image data to the print section, and the character data C of this combination is synthesized via the 0 or 210. In the case of halftone data, pulse rIJ modulation is also performed to reproduce halftones from both digital and analog sources.

When the character data C and the received image are in full color, the address of the memory M4 is controlled so that each color component code data is output in synchronization with the decoding operation button. If you want to make character data of a specific color, write a memo IJM4 in synchronization only with the decoding of the code of the specific component.
output from.

The MOD 156 includes a converter that converts 8-bit parallel data into 1-bit serial data for transmission over a data line or lineless.

The DEMOD 157 has a converter that converts the received 1-bit serial data into 8-bit parallel data.

In FIG. 1θ, the bit number of the character image to be synthesized is added to +iiJ of the halftone judgment in FIGS. 2-3. This can be achieved by inserting the circuit of FIG. 9 at point P in FIGS. 2-3.

This is to insert the upper 2 bit line (corresponding to 2F to 3FK) of the 6-bit image signal from the masking circuit with the higher white density (corresponding to 2F to 3FK) K character signal B (FIG. 7). As described above, the halftone judgment circuit 127-3 judges whether there is data 1 in the middle bits 1θ to 2E among the 6 bits. Therefore, the galactor signal inserted in 2F to 3PK is not a halftone. Since it is not considered, selector 132-1
34, each pixel is binarized not at the dither pattern level but at a constant threshold level. Therefore, this character (No. 1) is not dithered. That is, the bit character signal B is applied to the upper two bit data lines through the OR gates 301 and 302 in FIG. Since this character is not subjected to dither intermediate processing, the resolution is not increased.In addition, each latch in Figures z-x and 2-3 delays the data in J by about 1 bit to synchronize image processing. Also, B, O, and X7, which follow point P, are latch circuits that similarly apply a series of about one line from the second bit.

In FIG. 11, a bit signal of a character image is added after the multi-value processing shown in FIG. 2-12. 3rd-
This can be achieved by inserting the circuit shown in Figure 1O at point Q in Figure 2. This adds bit data B of the character image to the multivalued output, that is, the pixel data (dot data) pulse width modulated by the pulses φ, .about.φ in FIG. At this time, when the B signal is combined with the image signal at φ1 and synchronization 1~, % dark characters are added. That is, when switch 310 in the figure is turned on, φ.

A pulse of this width is output from the AND gate 303 with H, 310, and the gate 306. This is added to the image signal line via 307. Also, when the switch 308 is turned on, the character (
The g number B is attached to the image signal line. Therefore, the pixel becomes a pixel with a width of 3, and a character with a density of b is added. In this way, the density of the inserted character can be selected by selecting the switches 308-310. With this method, 1
Since the pulse width in the pixel is changed, the resolution of the character image is not impaired. Also, as mentioned above, the black component VC is run at the same time,
Alternatively, since character data can be added in synchronization with each color component, monochromatic characters such as black characters or blue characters can be inserted, and their shading can be controlled.

In each of the above figures, a part of the scanned image can be canceled and a character image can be inserted into the canceled part. This can be achieved by providing a data selector in place of each OR gate 210, corresponding to the synthesis section, and controlling the time of the selector in synchronization with the address counter. In this case, if the scanned image is in full color, the cancellation part can be made white and black or monochrome characters can be formed on the back.

Also, in one of these examples, as shown in Figure 2-3, halftone judgment and character composition can be performed before the document scan is completed, and the glint can also be revealed before the scan is completed, so the playback time of the composite image can be shortened. . This is particularly convenient for color image reproduction. Also, because it uses real-time color processing, it requires less memory.

[Brief explanation of drawings]

Fig. 1 is a sectional view of a color copying machine to which the present invention can be applied, Figs. 2-1 to 2-3, Figs. 3-1, 3-2, Fig. 5, and Fig. 6.
The figure is a circuit diagram of the surface image processing section, Figures 4-1 to 4-3 are flowcharts for black determination, single color determination, halftone determination and processing accordingly, and Figure 7 is a diagram explaining the reproduced image. Yes, 8th
The figure is a pulse waveform diagram for pulse width modulation of a laser beam, and FIGS. 9 to 11 are other image processing circuit diagrams. In the figure, 101-103 is a CCD board, 104 is a shading correction circuit f4, 105 is an r correction circuit, 109 is a masking circuit, 119 is a background color removal circuit, 127-1, 127-
2 is a monochrome detection circuit, and 127-3 is a halftone detection circuit.

Claims (1)

[Claims] means for binarizing the image signal; means for performing multi-value processing on the image signal; means for transmitting the binary image signal from the binarization processing means; and multi-value processing for the multi-value processing means. An image processing system comprising means for printing and reproducing image signals.