JPH04160981A - Image processor for each image area of copying machine - Google Patents

Abstract

PURPOSE:To make it possible to precisely regenerate copies of even many kinds of original pictures by automatically separating an image area of original picture and by making manual adjustment of each processing parameter possible when carrying out image processing. based on an image processing parameter for each separated image area. CONSTITUTION:To automatic image area recognizing circuit 31, a RGB image signal S1 is input from scanner 1, and the character part of an original picture outputs 0, while a variable density image part of the original picture outputs 1 as a recognition signal S21. Image processing circuit 30 adds various image processing operations to RGB image signal S1, and finally the signal S1 is converted to CMYK image signal S2 and outputs the signal S2 to memory unit 3. Image processing treated by processing sections 30a and 30b is variable, and processing parameters can be set in a parameter register group in processing sections 30a and 30b. Processing parameters are set by system controller 5, and when a power supply switch 23 is turned on, a constant data table in a processing program is referred to by the image processor to set up predetermined initial setting parameters.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention provides a method for automatically determining image areas (at least character areas and grayscale image areas) on a document in a copying machine equipped with an image reading means and an image forming means. The present invention relates to an image processing apparatus for each image area of a copying machine in which image processing contents are input independently for each area and a desired copy image is formed on the image area side.

2. Description of the Related Art In recent years, copying machines have become more sophisticated and multifunctional, and various proposals have been made regarding image processing and the like.

For example, according to Japanese Unexamined Patent Publication No. 61-118071, an edge detection means is provided, and smoothing processing is selectively performed from the edge detection output.

Furthermore, according to Japanese Patent Application Laid-open No. 13877/1987, a copy image is created by providing a character area discrimination means and performing halftone processing on the other image areas without performing halftone processing on the character area. I'm trying to play it. In this case, there are three types of character area discrimination means: a so-called automatic discrimination method based on the original image data, an area indication method, and a mark reading method on the original image.
The street is open. Also, JP-A-61-15707
According to the publication No. 2, an area selection means is provided and image processing is performed in the same manner as in the above publication. In addition to automatic region selection based on original image data, region input from the keyboard is also possible.

Furthermore, according to Japanese Patent Application Laid-Open No. 61-161870, in a method for recognizing a halftone dot image area, 1) the recognized halftone dot image is filtered.

Further, according to Japanese Patent Application Laid-Open No. 62-163, a color image detection means is provided, and black characters are output by binary conversion, and color characters are output by dither processing, so that the color information of the original image is preserved to some extent. Also, according to Japanese Patent Application Laid-Open No. 63-184886,
The density of the grayscale image is converted using two values, leaving the text and erasing the non-text.

In addition, in response to the desire to process a specific area of the original image, for example, according to Japanese Patent Application Laid-Open No. 1-192263, the area of the image (characters) marked with a marker is shaded to create an outline. We are trying to perform the following processing. That is, characters to be converted into outline characters or halftone characters are marked, the marked range is automatically detected, and processing is performed only within that area.

Problems that the invention aims to solve By the way, digital copying machines simply sample the original image.
Various problems arise when quantizing and printing out using the dither method or multi-value dither method.

For example, there are problems such as text becoming blurred, black text becoming slightly colored, gradation in photographic areas becoming poor even when the contrast in text areas is sufficient, and moiré occurring in halftone dot originals.

As one solution to this problem, the idea of image processing by image area has been proposed, in which characters and grayscale image parts are separated and optimal image processing is applied to each.

To separate image areas, there are two methods: the so-called area specification method, in which an operator manually specifies the area, and the so-called automatic image area separation method, in which the image area is automatically separated based on image data. It is described in detail in the publication No. 61-13877. Among these, the former method is good when the text and photo areas are separated in large blocks, but it is practically impossible to specify the area when the text is embedded in the photo. There is a problem. On the other hand, according to the latter, such problems can be solved and the operation is simple.

However, although conventional automatic image area separation image processing can statistically perform optimal processing on the text and photographic areas of most manuscripts, it is not necessarily optimal for relatively small numbers of special manuscripts. It does not necessarily mean that it will be processed. For example, an automatic separation means and a spatial filtering means are provided to separate the text part and the halftone image part, and based on the separation results, the text part is subjected to sharpening filter processing, and the halftone image part is subjected to smoothing filter processing. shall be. In such cases, since most halftone originals have a dot density of 150 to 175 lines/inch, the smoothing filter coefficient is set to a value that is expected to prevent moire and minimize image blur. . However, even if it is very rare, there are originals such as newspapers that have roughness of about 85 lines/inch, and when these are copied, problems such as moire may occur with the image processing parameters optimized under the above conditions. In this case, if the degree of smoothing filtering processing is strengthened, the sharpness will decrease slightly, but the moiré will be eliminated and the overall quality of the copy will be improved. Therefore, although the default processing parameters for automatic separation image processing in copiers are fixed, it would be possible to intentionally change each parameter without sacrificing operability and to handle a wider variety of originals. You can say that you get it.

Further, as shown in the above-mentioned Japanese Patent Laid-Open No. 1-192263, there are cases where it is desired to perform various processing on the original characters. However, since this method places marks on the original characters, it damages the original, and it is not able to handle characters embedded in gray scale images in the same way as methods using tablets, etc. . Furthermore, there are many cases where it is desired to perform active image processing only on the shading image portion while leaving the text as is. For example, a map of a mountainous area has text in various colors embedded in a dark brown shading image, but if you copy this using a single-color copying machine, the entire text will be black and the text will be almost unreadable. It is well known that changing the density does not improve the overall contrast. In such cases,
It can be said that an easy-to-read copy can be obtained by lowering the image density of only the shaded image portion while leaving the density of the characters unchanged.

Furthermore, in the upper part, we explained the case where the text is divided into two parts, text and grayscale image, but text is divided into black text and color text, and grayscale image is divided into continuous tone images such as silver halide photographs. The optimal image processing conditions for copy reproduction are different between printed matter and halftone (halftone) images such as those copied by digital copying machines, and the meaning of the information contained in the original image is often different, so image processing is necessary. It would be more convenient to be able to separately set the image processing conditions for processing.

Therefore, the first problem is to automatically separate the image areas of the original image and to perform image processing based on the image processing parameters of each of the separated image areas, making it possible to manually adjust each processing parameter. It is desired to be able to faithfully copy and reproduce even more types of original images, and furthermore to be able to actively process images on the image area side.

As a second problem, it is desired to be able to form copies with different copy magnifications or image positions for character images and grayscale images.

As a third problem, it is desired to be able to selectively convert text portions or grayscale image portions into mosaic images, mirror images, slanting (tilting), shadow images, and outline images.

As a fourth problem, it is desired to be able to selectively smooth the character portion or the grayscale image portion to a desired level or sharpen the image to a desired level.

The fifth issue is selectively blanking out (trimming) the text or dark image area, painting (if applied only to the text, color conversion of only the text will be achieved), contrast conversion, and gradation conversion. It is desired to be able to perform (negative inversion), tone partial inversion (solarization), or tone omission (bostarization or simple binary imaging).

As a sixth problem, it is possible to obtain images in which character parts or grayscale image parts are selectively subjected to specific color correction or under color removal (UCR) processing, color conversion images, monocolor images, or undercolor images. It is requested that the

As a seventh problem, it is desired to obtain an image in which character portions or grayscale image portions are selectively formed with a specific halftone dot shape (dither pattern), size, and directionality.

Furthermore, as the eighth problem, we decided to change the image area of the original image to black (achromatic color).
It is automatically divided into four areas: text area, color (chromatic color) text area, continuous tone (continuous tone) image area, and halftone (halftone) image area. When performing image processing based on image processing parameters, it is desired to be able to manually adjust each processing parameter and to be able to reproduce faithful copies of even more types of original images.

Means for Solving the Problems A copying machine comprising an image reading means for decomposing an original image into pixels and reading it, and an image forming means for forming image data as a visible image on a recording medium, according to claim 1. In the invention, automatic image area recognition means separates and determines at least two areas of an original image, a character area and a grayscale image area, from the image data read by the image reading means, and the automatic image area recognition means separates and determines at least two areas of the original image, a character area and a grayscale image area, from the image data read by the image reading means, and image processing means that selectively performs at least two types of image processing: first image processing on the character area recognized by the automatic image area recognition means and second image processing on the grayscale image area; processing parameter input means for inputting parameters that determine the first image processing content or the second image processing content of the image processing means; and image processing in response to the parameters input from the processing parameter input means. A control means for selectively changing and energizing the first image processing content or the second image processing content of the means is provided.

In this case, the invention according to claims 2 to 8 is configured as follows.

First, in the invention as claimed in claim 2, the image processing means performs at least one image processing of image scaling with a desired scaling factor and image shifting with a desired amount of movement for each automatically separated image area.
A shift means is provided, and at least one parameter of a magnification change rate and an image movement amount for each image area is input and set from a processing parameter input means.

In the invention according to claim 3, the image processing means includes mosaicization of a desired pitch for each automatically separated image area, mirror image/
Image editing means is provided for performing at least one of the following image processing: forming a non-mirror image, slanting the image at a desired angle, shading with a desired length, color, density, and shape, and forming an outline image with a desired line width and color. , mosaic pitch for each image area, mirror selection, oblique angle,
At least one of the shading parameter and the contouring parameter is input and set from a processing parameter input means.

In the invention according to claim 4, the image processing means includes a space using arbitrary filtering coefficients for performing image processing such as smoothing processing to a desired smoothness and sharpening processing to a desired sharpness for each automatically separated image area. Spatial filtering means for performing filtering processing for each color or for all colors in common is provided, and filtering parameters for each image area are input and set from a processing parameter input means.

In the invention as set forth in claim 5, the image processing means includes at least one of blanking, painting with an arbitrary color, contrast conversion, density conversion, gradation inversion, gradation partial inversion, and gradation omission for each automatically separated image area. A gradation processing means that performs one image process for each color or common to all colors is provided, and it can perform blanking selection for each image area, paint color and its density, contrast, density, gradation inversion selection, gradation partial inversion selection, gradation. At least one parameter of tone omission selection and the number of tone levels is input and set from the processing parameter input means.

In the invention according to claim 6, the image processing means includes color correction processing using a desired correction coefficient for each automatically separated image area, removal of an arbitrary amount of undercolor, K plate reprinting processing, density conversion, blanking, and arbitrary A color processing means is provided that performs at least one of the following processes: painting a color, converting any color, converting any color into a monochrome image, and adding an undercolor to any color. Processing parameter input means for at least one of the following parameters: type and degree, density, blanking selection, paint hue and density, original color and converted color for color conversion, hue and density for monochrome conversion, and hue and density for undercolor. The input settings are now set from .

In the invention as set forth in claim 7, the image processing means includes halftone processing such as halftone dot shape processing, halftone dot size, halftone dot direction, etc. for expressing a desired pseudo halftone for each automatically separated image area. A halftone processing means is provided for performing the process for each color or for all colors in common, and halftone processing parameters for each image area are input and set from a processing parameter input means.

In the invention according to claim 8, the automatic image area recognition means outputs different signals for black characters and color characters in the character area of the original image, and outputs different signals for the continuous tone image and the halftone image in the grayscale image part of the original image. The image processing means performs different image processing on black characters, color characters, continuous tone images, and halftone images, and at least one of the processing parameters is set as ξλ input by the processing parameter input means. I did it like that.

According to the invention described in claim 1, image processing parameters can be input by the processing parameter input means, and when these parameters are input, the processing parameters of the image processing means are changed and set by the control means. , processing parameters can be set arbitrarily, increasing the degree of freedom in image processing.

In the invention as set forth in claim 1, according to the invention as set forth in claim 2, since the image processing means has a scaling/shifting means, it is possible to make copies with different copying magnifications or image positions for the character image and the grayscale image. It becomes something that can be formed.

Further, according to the invention as claimed in claim 3, since the image processing means has the image editing means, the text portion or the grayscale image portion is selectively converted into a mosaic image or a mirror image for each image area selected for image processing. , slanting, and shading can be used to create a copy that has been converted into an image or a contour image.

According to the invention as claimed in claim 4, since the image processing means has a spatial filtering means for each RGB or CMYK color or common to all colors, the character part or the grayscale image part can be selectively smoothed or sharpened to a desired level. It becomes possible to create a copy of the

According to the invention as set forth in claim 5, since the image processing means has the gradation processing means, the text portion or the grayscale image portion can be selectively blanked out (trimmed), painted, contrast converted, gradation converted, and the gradation portion This makes it possible to create a copy with inversion or gradation omitted.

According to the invention set forth in claim 6, since the image processing means has the color processing means, an image, a color-converted image, or a monochrome image in which a text portion or a grayscale image portion is selectively subjected to specific color correction or undercolor removal processing is provided. Copy images with color images and undercolor images can be formed.

According to the invention set forth in claim 7, since the image processing means has a halftone processing means, a specific halftone dot shape, halftone dot size, halftone dot directionality, etc. can be selectively applied to a text area or a grayscale image area. You can obtain a copy image that has been subjected to halftone processing as specified.

Furthermore, according to the invention as set forth in claim 8, the image area of the original image is divided into four areas, a black character area, a color character area, a continuous tone image area, and a halftone image area, by the automatic image area recognition means and separated. When image processing is performed on an image area based on image processing parameters specific to each image area, each processing parameter can be manually adjusted, and faithful copies of even more types of original images can be reproduced.

Example An embodiment of the present invention will be described based on the drawings.

[Overall Configuration] FIG. 2 shows a system block diagram of a digital color copying machine as an example to which the present invention is applied, and FIG. 3 shows its arrangement structure. Generally, a scanner unit (image reading means) 1 and an image processor (image processing means)
2, memory unit 3, and printer (image forming means)
4, a system controller (control means) 5, a console unit (image area designation means & processing parameter input means) 6, a digitizer unit 7, a sorter unit 8, and an AD
It consists of an F unit 9, an external device interface terminal 10, and a magnetic disk unit 11.

The general functions of each of these units will be described below.

System Controller 5> This is a stored program type 32-bit microcomputer system that controls the entire copying machine system.

In addition to the CPU, program memory, and work memory, it has interface means for communicating with external units such as a scanner 1, image processor 2, printer 4, and console unit 6, and an interrupt controller for processing hardware interrupts. . This system controller 5 monitors the status of other units and also
The operation specifications of each unit that should function are determined according to the various copy modes input from the console unit 6, and operation parameters are sent to each unit before the copy process starts, and processing start signals and processing It has the role of supplying various real-time signals necessary during the process to other units.

In addition, the work memory stores important information such as billing management information, so the power supply is always backed up.

Scanner 1> Document 13 on platen 12 is scanned by CCD 14. l, 14, .
14e, the digital image signal S1 is separated into RGB, sampled at a sampling density of 400 dpi, quantized to a quantization level of 8 bits, and the digital image signal S1 is sent to the image processor 2.
Alternatively, it is supplied to the external device connection interface terminal 10.

Image processor 2> Performs various image processing such as color correction and dither processing on the RGB original image signal S1 supplied from the scanner 1 or the external device connection interface terminal 10, and finally outputs the CMYK image signal S3, which is a print signal. It has an image processing function that converts images into images. It also has an image detection function that detects the document size and the color of a specific part and provides this information to the system controller 5.

Furthermore, it has an image generation function that generates various patterns and number patterns. It is possible to operate these three functions at the same time, and for example, an image in which a numerical pattern is synthesized into an image of the scanner 1 can be sent to the next stage memory unit 3, and finally a synthesized image can be formed in the printer 4. become.

Memory unit 3> A memory unit 3 which supplies CMYK four-color image signals S to the printer 4 by delaying C (cyan), M (line fi), and Y (yellow) data by a predetermined time with respect to K (black) data. 1
A second operation mode stores CMY image data S, and reads out the image data stored in the second operation mode by delaying each of the C, M, and Y data by a predetermined time with respect to the data. It is an image memory means that is provided with three modes, S, and a third operation mode for supplying the image to the printer 4, and can selectively activate any one of the operation modes. By operating the second operation mode a plurality of times, it is possible to insert characters generated by the image processor 2 into the copied image by performing image synthesis or partial rewriting.

Magnetic Disk Unit 11> This is a large-capacity magnetic disk drive that can store application programs for the system controller 5 and image data for multiple pages, and the drives include a floppy disk drive (a, ) and a hard disk drive (b).
It consists of two sets.

Printer unit 4> This is a laser printer having a recording station for four CMYK colors, and based on the data supplied from the image processor 2 and the CMY image data S supplied from the memory unit 3, the photoconductors 15BK, 15c, 1
1 to the laser scanning means 168 on 5M and 15Y;
An electrostatic latent image is formed by the action of 6C, 16M, and 16Y,
These latent images are transferred to developing units 17BK, 17C, and so on.

17M and 17Y to visualize the image, and by the action of transfer chargers 18BK, 18c, 18M, and 18Y, a full color print image is printed on the transfer paper 21 fed from the paper feed tray 19 and conveyed by the transfer belt 20. is transferred, fixed by the fixing device 22, and discharged.

Console unit 6> Consists of a 512 x 512 dot matrix liquid crystal display, 128 x 128 matrix transparent touch switches mounted on the display, and buttons such as a 10 key and a start button. Arbitrary figures can be used as display means,
Since characters can be displayed, the operator can obtain output information from the copying machine, and can also provide desired operating specifications to the copying machine by touching a switch on a predetermined icon display.

Digitizer unit 7> A means for obtaining pen input coordinate position information at intervals of 0.2+mn, and used by the operator to specify a specific part of a document or input the insertion position when inserting a character string into a copied image. Becomes the property of

Sorter unit 8> This unit is for sorting the copied transfer paper 21.

(ADF unit 9) This unit is for automatically feeding the original 13 onto the platen 12.

External device interface terminal 10) For example, it is a multi-pin connection terminal for supplying image data to an external device such as a general-purpose computer, receiving image data from an external device, and exchanging various information.

Note that the structure, function, and operation of the scanner 1, memory unit 3, printer 4, and console unit 6 may be those disclosed in, for example, Japanese Patent Application Laid-Open No. 64-25673, and the details will be omitted here. Furthermore, the ADF unit 9, sorter unit 8, digitizer unit 7, magnetic disk unit 11, and system controller 5 are also units constructed using conventionally well-known techniques, and their details will also be omitted.

Below, various embodiments of the image processor 2 and related units will be mainly described.

L 1st Example]...See Figure 1 In this example, image processing is switched between the text area and the grayscale image area, and these image processing parameters can be input separately from the console unit 6. This makes it possible to obtain copies with independent image processing effects for different parts.

In FIG. 1, the circuit included in the image processor 2 is as follows:
Image processing circuit 30 and automatic image area recognition circuit (means) 31
It is.

The automatic image area recognition circuit 31 receives RGB from the scanner l.
The image signal S1 is input, and the recognition signal 8 is "0" for the character part of the original picture and "rl" for the grayscale image part of the original picture.
．． Output as 1. Recognition is performed pixel by pixel, and the output of the recognition signal S I l is also switched pixel by pixel. The image processing circuit 3o performs various image processing operations on the RGB image signal S1 sent from the scanner 1, and finally converts it into a C
It is converted into a MYK image signal S and sent to the memory unit 3.

The image processing circuit 30 includes a first processing section 30a that performs first image processing and a second processing section 30b that performs second image processing.
The RGB image signal S from the scanner I is commonly input to the image processing units 30a and 30b. Which image processing to perform is determined by the recognition signal S, 1 of the automatic image area recognition circuit 3I, and when the value of the recognition signal S m + is "0", the image processing by the first processing unit 30a is selected, When it is "1", image processing by the second processing section 30b is selected. Processing results by these processing units 30a or 30b are output as CMYK image signals S. Therefore, when the recognition signal S a I is rOJ, the processing result by the first processing unit 30a becomes the CMYK image signal S, and when it is “1”, the processing result by the second processing unit 3ob becomes the CMYK image signal S. The image processing performed by the processing units 30a and 30b is variable, and the processing parameters are determined by these processing units 30a.

It can be set in the parameter register group inside 30b.

The processing parameters are set by the system controller 5, and the moving direction of this setting data S s −+ Sab is indicated by a broken line arrow. Processing parameters can be set arbitrarily in this way, but defaults are also determined, and when the system power switch 23 is turned on, the system controller 5
refers to the constant data table in the program and sets predetermined initial setting parameters. As for the initial setting parameters, processing parameters suitable for characters are selected for the internal register group of the first processing section 30a, and processing parameters suitable for grayscale images are selected for the internal register group of the second processing section 30b.

These processing parameters are changed when copying a special original or when it is desired to obtain a copy with aggressive image processing. By opening a processing parameter change screen on the console unit 6 and touching icons and various buttons on the screen, the operator can input desired character processing parameters and grayscale image processing parameters independently. In response to an operator's input operation, the console unit 6 issues an operation input signal S, which the system controller 5 decodes and sends predetermined processing parameters to the processing unit 30.
a and 30b, respectively.

[Second example] See FIG. 4 In this example, the content of image processing can be made different between the character part and the grayscale image part within the area input by the tablet.

The circuits included in the image processor 2 in FIG. 4 include an image processing circuit 30, an automatic image area identification circuit 31, an image area designation circuit (means) 32, and an image processing selection circuit (means).
33.

The automatic image area recognition circuit 30 receives the input of the RGB image signal S1 from the scanner 1 in the same manner as described above, and outputs a recognition signal S II + which is "0" for the character part of the original image and "IJ" for the grayscale image part. It is output, and is done pixel by pixel.

The operator uses the digitizer unit 7 which is an area input means.
You can input a specific region shape from . The system controller 5 generates region shape data from the coordinate data S from the digitizer unit 7, and also automatically assigns a region number to each region in order to identify which region it is in case there are multiple regions. to be given. Thereafter, the system controller 5 loads the region shape data and the region number into the image region designation circuit 32. This setting process is performed by the system controller 5 using the signal S.

Do it through. When document scanning is started, a synchronizing signal LSYNC for each scanning line and a video synchronizing signal VCLK for each pixel are input to the image area specifying circuit 32, and the image area specifying circuit 32 counts these signals to determine the current The original image scanning position is specified, and the previously set area identification number corresponding to the coordinates is output as a signal Sam.

The image processing selection circuit 33 receives the signal S s + + S a
This is a combinational logic circuit that takes @ as an input and outputs a 1-bit signal Sss. Therefore, the image processing selection circuit 33 can be realized with a simple look-up table, and the signal S, .

A single output is determined by the combination of Sl. The table is a so-called RAM, and the contents of the table can be rewritten by the system controller 5.

During the copy operation, the signal S a + t S s is connected to the address input of the table RAM, and the RAM is accessed in a continuous data output mode. In addition, when it is necessary to rewrite table data, it is placed in a copy waiting state, and both the address and data lines are connected to switch means (not shown).
All you have to do is switch to the system controller 5 side and access in write mode.

The configuration and operation of the image processing circuit 30 are the same as those shown in FIG. However, the first and second processing units 30a.

30b to perform the image processing is determined by the output signal S1.30b of the image processing selection circuit 33. This signal S
When as is rQJ, first processing is performed by the first processing unit 30a, and when as is rl, second processing is performed by the second processing unit 30b, and the respective selected processing results are output as CMYK image signals S1. Therefore, for example, the image processing selection circuit 33
The total number of input bits is "2", and the upper bits have a signal S
Assuming that Wl is input and signal S is input to the lower bit, and the identification number of the specific area input by the operator is "1" and the remaining area is rQJ, the character part in the area is If we want to perform image processing only by the first processing unit 30a, and to perform image processing by the second processing unit 30b on the grayscale image parts outside the area and inside the area, the table data of the image processing selection circuit 33 is "01". It may be set so that "0" is output for inputs and "1" is output for other combination inputs.

[Third example]...See Figure 5 In this example, similar to the previous example, the image processing contents are different depending on the text part and the grayscale image part in the area input with the tablet.
The scope of the image processing contents to be changed at that time can be finely adjusted.

The basic configuration is similar to that of the second example, but in this example, the image processing circuit 30 is divided into a plurality of processing steps (n) according to the processing content, as shown in FIG. ．． ．． 30. , ~,30°. Although these image processing circuits 301, 30, . ). Therefore, the output signal is also S * + t S * *
It is output as +~esmn. Further, the output of the image processing selection circuit 33 is also increased in accordance with the number of steps n of the image processing circuit 30, and the output signal S@#l corresponds to the input signal S*+tSas to the image processing selection circuit 33. l S * m * l~.

S. It is possible to obtain anything you want.

[Fourth example] See FIG. 6 This example is a combination of the contents of FIGS. 4 and 5. That is, it is possible to change the image processing content depending on the character part and the grayscale image part within the area inputted with the tablet, and to finely adjust the range of the image processing content to be changed. Furthermore, the processing content of each category can be specified from the console unit 6. The system controller 5 includes an image processing circuit 30. , 30. , ~. 30. , and n
Two sets of image processing parameters for each processing step can be set using signals S and e.

[Fifth Example] See FIG. 7 This example includes all the functions described in the first to fourth examples, and is a more specific configuration of the image processor 2.

First, a plurality of image processing circuits 34 to 41 of different categories are provided in series between the input and output of the signals S, , S, so that the image data can be subjected to pipeline processing. Each of the processing circuits 34 to 41 has the ability to execute multiple types of processing in parallel. In addition, in this example, area specification processing is possible, so for example, by applying different processing to only a certain shaped area, it is possible to partially obtain a special image processing effect, even if the area shape is like a letter. For example, even if such characters do not exist in the original, it is possible to form an image on the copy that has a visual effect similar to that of characters. Furthermore, since an automatic image area recognition circuit 31 is provided, it is possible to perform different image processing for text areas and gray scale image areas, and to select image processing based on the combination of specified areas and automatically recognized image areas. You can also. Furthermore, it is also possible to have a structure in which these various editing processes are performed in series instead of in parallel.

First, the processing circuit 34 is a magnification/shifting means that simultaneously moves the image by a plurality of types of movement amounts and changes the magnification by a plurality of magnification ratios. It is possible to perform enlargement/reduction processing with different image shift amounts and image magnification ratios for characters and grayscale images inside and outside a designated area, or a combination thereof.

The processing circuit 35 serves as an image editing means, and can perform image mirroring, tilting at multiple angles, mosaic at multiple pitches,
Editing processes such as shadow casting in multiple forms and contour imaging in multiple forms can be performed in parallel. Like the previous processing circuit 34, this can selectively perform these processes depending on the pixel position of the original image.

The processing circuit 36 has a plurality of spatial filters arranged in parallel and serves as a spatial filtering means, and processes the RGB original image signal S.
Add various filtering operations to 1. Since each filter coefficient is loaded with an arbitrary value from the system controller 5, processing such as image smoothing, averaging, sharpening, first-order partial imaging, and second-order differential imaging can be performed. Furthermore, it is also possible to perform image density conversion by coefficient manipulation.

The processing circuit 37 constitutes a gradation processing means that simultaneously performs a plurality of gradation conversion processes arranged in parallel, and the conversion method is a table lookup method using a RAM.

Therefore, by rewriting this table, the system controller 5 can perform gradation correction according to the original image signal, negative/positive inversion, density conversion, contrast conversion, gradation omission such as bossarization, partial gradation inversion called solarization, etc. Operations such as blanking by setting all table values to rQJ and painting obtained by setting all table values to constant values other than rQJ can be performed.

The processing circuit 38 constitutes a color processing means that simultaneously executes a plurality of types of color correction processing, and converts the RGB image signal S1 into a CMYK image signal S. Since the conversion calculation parameters can be set arbitrarily by the system controller 5, various image processing processes such as color correction, UCR processing, color conversion, monochrome processing, blanking, brightness conversion, color painting, and undercolor can be performed. can be administered.

The processing circuit 39 is a CMYK gamma correction circuit, which constitutes a part of the gradation processing means, and corrects γ to be suitable for the density gradation characteristics of the printer 4. The internal circuits and functions are similar to those of the processing circuit 37.

The processing circuit 40 is a CMYK spatial filter circuit and constitutes a part of the spatial filtering means.

Apply color-based filtering processing to Y. Functionally,
It is similar to the processing circuit 36.

The processing circuit 41 is a dither processing circuit that simultaneously performs a plurality of dither processes, and constitutes halftone processing means. The dither pattern can be arbitrarily set, and halftone processing can be performed with various halftone dot densities, halftone dot shapes, halftone dot sizes, and screen angles.

Each of the processing circuits 34 to 41 that perform image processing steps in different categories has the ability to perform multiple different processes in parallel, and only the signal resulting from one of the multiple processes is sent to the next processing circuit. Sent. Furthermore, the processing contents of each of the plurality of processes are variable, and processing parameters are downloaded from the system controller 5 to each processing circuit 34 to 41 through a path line 42 connected to the system controller 5. For example, the processing circuit 35 for image editing can perform two types of shading processing in different colors, two types of mosaic processing, two types of contouring processing, one type of mirroring processing, and no processing processing in parallel. , the pitch dimension of the mosaic, the width of the shadow, and the color are downloaded as operating parameters from the system controller 5 to this processing circuit 35 prior to document scanning, and when the document is scanned, only one of the eight types of processing results is used in the next stage of spatial filtering. The data will be sent to the processing circuit 36 for use.

This mechanism will be explained with reference to FIG. FIG. 8 shows the configuration common to all of the processing circuits 34 to 41 as a conceptual model, and 1ndata is image data sent to the processing circuit from the preceding processing circuit, etc. If the circuit is any one of the processing circuits 34 to 38, it is RGB image data, and if the circuit is any one of the processing circuits 39 to 41, it is CMYK image data. First, pO~
pm is (m+1) parallel processing circuits, to which the same image data 1ndata is input. The outputs of these parallel processing circuits pOA-pm are represented by pout O to poutm. For each parallel process, there are two cases: a case in which the processing parameters are predetermined, and a case in which they are variable. In the case of variable processing, operating parameters are loaded into internal registers through an internal bus directly connected to the system controller and controller 5 before starting the image processing operation. Generally, there are multiple internal registers, so these selections can be achieved by decoding part of the address signal in the bus signal using an address decoder, and connecting each of the decoded signal lines to each internal register. can.

Such parallel processing results are input to input ports INO to INm of the multiplexer mupx, respectively. The multiplexer mupx selectively outputs only one set of input data from the output port OUT. Which input data to select depends on the input code to the SEL terminal; rQJ is lN01, INI is "1", ~
, rm", the input corresponding to INm is output from the output port OUT.

adec is an image processing selection means, which corresponds to the image processing selection circuit 33 shown in FIG. However, although the image processing selection circuit 33 in the example shown in FIG.
It is said to have a format for every 41 to 41. The mechanism is a lookup table circuit. Physically, a RAM is used, and the input signal is connected to the address line of this RAM.
The read output data of M is used as Dout.

That is, a predetermined value is output from the Dout terminal to the SEL terminal of the multiplexer mupx for 8-bit input values from ao (lsb) to c (msb). The output value ranges from "O" to r m J. The contents of the lookup table can be arbitrarily rewritten by the system controller 5 through an internal bus directly connected to the bus 42.

Here, the 4 bits aC)-a3 are the image area designation circuit 3.
The signal S s from 2 is inside. Also, C,...
CG, H, and P are recognition signals S from the automatic image area recognition circuit 31, and C is rl when it is recognized as a black character part.
J, and the other values are rQJ. Similarly, CC is recognized when characters in colors other than black are recognized, P is recognized when continuous tone images such as silver halide photographs are recognized, and H is recognized when halftone images are recognized as in printing or copying machines. At times, each has a value of "1", and at other times, it has a value of rQJ. Therefore, if you set the contents of the lookup table adec to appropriate values,
Parallel processing circuit pO”- according to the type of original image and specified area
The processing results at pm can be selectively sent to the next stage image processing circuit.

Since the input is 8 bits, there are 256 combinations. Also, the number of parallel processing circuits pc)-pm (m+1)
is less than 256, the table data is set to a value smaller than m. In such a case, there will naturally be cases where the output is the same for a given input, and (m+1)
= 256, the same data may be set.

For example, if the input binary value of the image processing selection means adec is rl 0OOXXXX (X is 14 is O), the table data is set as rQJ. If the data in rXXXXOOOIJ is set as 5, the processing selection signal will be "1" regardless of the type of original image.
At this time, the processing result of the parallel processing circuit p5 is selected.

Selection of parallel processing results is possible on a pixel-by-pixel basis.

That is, it is possible to perform any processing on any part of one document 13.

Here, as shown in FIG. 9, the automatic image area recognition circuit 31 determines, for each pixel of the original RGB signal S1, which of the four types of black text, color text, photograph, and halftone image the original picture belongs to. Recognize the signal S I l as C, CC, H,
"1" and "O" are output to P as appropriate. The recognition algorithm is well known and will be omitted, but it includes the differential value of the image density, the degree of continuity of the uniform density part, the periodicity of the density distribution,
The continuity of the lowest density portion, the spatial frequency characteristics, the connection state of pixels with the same density, etc. are analyzed for each RGB color, and a comprehensive judgment is made.

Further, the image area designation circuit 32 outputs a 32-bit processing selection signal S mt to the processing circuits 34 to 41. However, l bit (lowest bit bO) in this
is also supplied to the memory unit 3 as a memory override signal. The image area specifying circuit 32 has a function of issuing an output signal to selectively switch image processing to a desired portion of the original image. Switching over whether to override or not can be done pixel by pixel. Here, the image area designation circuit 32 is used for two purposes. One is when the operator uses the digitizer unit 7 to apply special image processing to a desired area, which is conventionally well known as area designation processing for copying machines. The other method is that the system controller 5 automatically generates area data based on internal memory data, detection output values of the detection circuit, and character code information input by the operator, without the operator specifying the area. When this area data is downloaded to the main image area specifying circuit 32 and special image processing such as painting the area shape or blanking is performed on the transfer paper 21 to form a pattern image that looks like characters, figures, or graphs. be. Therefore, if the area shape is made into an alphabet string, a number string, or a pictogram, meaningful information to the operator can be provided from the copying machine to the operator in the form of a hard copy. Details of the circuit configuration will be described later.

Further, in FIG. 7, an original size detection circuit 43 and an original color detection circuit 44 are provided in the image processor 2 of this example. That is, the detection results obtained by scanning the original are stored in internal registers (static registers) belonging to each circuit, and the system controller 5
It is configured such that the contents of these registers can be referenced at any time through. Furthermore, a data compression circuit 45 is also provided. This data compression circuit 45 compresses the RGB image data $1 and converts the compression result into data S 41
The data is sent to the magnetic disk unit 11 as a. Conversely, the compressed image data S sent from the magnetic disk unit 11
A data decompression circuit 46 is provided for restoring the I+ to the original RGB image data, and the restored image data S4
．． is the original image data S.

Similarly, it is input to the magnification/shift circuit 34.

Each of the above-mentioned processing circuits has, for each circuit, a group of registers that determine the operation of each circuit (these are referred to as command registers, parameter registers, etc. for convenience) and a group of registers that store information on the operation results of each circuit (these are called registers for convenience). , referred to as a static register for convenience.

The configuration of the bus 42 connected to the system controller 5 is similar to that of a general 32-bit microprocessor. That is, the data bus width and address bus width are each 32 bits, and there are basically 66 signal lines in total, including control signals such as read signals and write signals. Address bus signals are decoded and used to select processing circuits and internal registers. The upper 3 bits of the address bus are decoded by a decoder (not shown) in the image processor 2 and used to select processing circuits 34 to 41 and circuits 45 and 46. The lower 9 bits of the address bus are used to select the processing circuits 34 to 41 and 45 to 46. It is decoded by each decoder and used to select one register among the command register group and static register group in each circuit. That is, each circuit can have a maximum of 5 and 2 registers. This is the same method as general chip selection of peripheral elements and in-chip register selection.Also, a circuit selection decoder may be provided on the system controller 5 side.

Therefore, the system controller 5 can selectively freely access one of the registers inside each processing circuit in the image processor 2 through the bus 42. In other words, from the perspective of the system controller 5, each processing circuit and internal register can be regarded as the same as a memory connected to the CPU bus. Therefore, the system controller 5 can execute the process of downloading operating parameters to each circuit command register and obtaining the success/failure of processing results (error information) and size and color detection results from the static register at extremely high speed. .

[Detailed Explanation 1 of Image Area Designation Circuit 32 A detailed circuit configuration of the image area designation circuit 32 is shown in FIG. This circuit is provided with an internal pass line 50,
It is directly connected to the pass line 42 of the image processor 2. Further, there are a plurality of internal registers, and an address decoder is provided for selecting these, but illustration thereof is omitted.

First, a region register group 51 consisting of four registers indicated by a, b, c, and d is provided. These four registers all have the same structure, and the configuration of one of them is shown in FIG. 11. That is, this register has a word length of 32 bits and has the role of holding area processing selection data when one copy image is divided into a plurality of areas and different image processing is applied to each area. To be exact,
As described above, this is selection data when one of the parallel processing results in each of the processing circuits 34 to 41 is selectively sent to the next stage circuit. Functionally, this register is divided into 4 bits each, and each processing circuit 34 to 41 is divided into 4 bits.
Contains the selection number for selecting the process. For example, bits 12 through 15 are connected to color processing circuit 38.

Note that only bit O is special, and this signal is connected to the dither processing circuit 41 and is also output to the memory unit 3 at the same time. When the memory unit 3 is in the second operation mode (storage mode), if the signal of this bit O is rQJ, it is not overridden, and if it is "1", it is overridden.

That is, a process of partially rewriting the image data in the memory is performed.

One of the four area registers is selected for each pixel, and the register data is output to each of the processing circuits 34 to 41, thereby making it possible to perform image processing for each area.

Also provided is a multiplexer 52 with 32-bit input and 2-bit output. A 2-bit output from multiplexer 52 is used as a register selection signal for one register in area register group 51.

Furthermore, toggle memories 53 and 54 are provided which have the function of storing area register selection data for all pixels of one scanning line. Each toggle memory 53.54 is 297 words x 32
It has a bit structure, and one word holds register selection information for 16 pixels. In other words, the total memory amount is 400dpi (
This corresponds to the memory size (297 x 16 x 2) for one scanning line of 297 mm when each pixel has 2 bits of information at a pixel density of approximately 16 dots/value). The two toggle memories 53 and 54 operate in a toggle manner, and when one is performing a write operation, the other one is performing a read operation. This toggle switching is performed in units of l scanning lines. These toggle memories 53.54
A write/read controller 55 is connected to. That is, write data for 297 words from the bus 50 is written to the toggle memory 53 or 54 from the A port in synchronization with the bus cycle, and the memory 5 that is not being written is written to the toggle memory 53 or 54 from the A port.
4 or 53 has the function of reading data and outputting it from the B boat. A flip-flop 56 is connected to this write/read controller 55. This flip-flop 56 repeats inversion with a line synchronization pulse LSYNC output once for each scanning line, and its output is X of the write/read controller 55.

It is used as a Y write/read switch, that is, a toggle signal.

In addition, a 1/16 frequency dividing circuit 57 and a resettable counter 5
8 is provided, and a memory address Ba is provided to the B boat of the write/read controller 55 for each 16 pixels.
dd is generated. In other words, the resettable counter 58
is cleared by the line synchronization pulse LSYNC that is emitted prior to scanning one scanning line, and is counted up every time the video synchronization signal VCLK, which is output once for each pixel, is input by 16, and counts from "○" to r296J. . That is,
A memory read access is performed from word 0 to the 296th word, and memory data is provided to the multiplexer 52 via the data line B data of the B boat.

Further, a hexadecimal counter 59 is provided.

This hexadecimal counter 59 is initially incremented every pulse of the video synchronization signal VCLK, and is incremented by 1
When it counts up to 5, it returns to 0 again with the next pulse. The output value of the counter 59 is used as the selection information SEL signal for selecting two consecutive bits in the one word 32-bit data output by the multiplexer 52. Simply put, 32 bits are divided into 2 bits each, 16,
It is divided into sections, and the divided two bits are sequentially supplied to the area register group 51 in synchronization with the pixel clock VCLK.

Here, the area register group 51 is connected to the multiplexer 5.
When the signal supplied from 2 is rOJO, register a, “1
”, register B, and r2J, register C1 “3”
In this case, each register d is selected, and the data in the selected register is used as a selection signal S I t to each processing circuit 34 to 4.
Send to 1.

FIG. 12 is an explanatory diagram illustrating the operation of such image area designation circuit 32. 11 shown on the transfer paper 21. l
2. ~． 1m, . . . (11-1), 11 is a scanning line. The scanning line is actually A3 size, 6720
There are many books, but here I will illustrate with a small number of books. Further, on the transfer paper 21, the "4" shaped part is one designated area, and the area other than this area (the part excluding the "4" shape) is another designated area. The former is area 1, the latter is area O
Name it. In scan lines 11 and 12, only area 0 exists, but in scan line 13, pixels from xO to xl are in area 0.
．． The pixels from XI to X2 are in area 1, and the pixels from X2 to X3 are in area 0. Pixels from X3 to X4 are area 1,
Pixels 4 to Xn belong to area O. After this, the same area switching continues in the main scanning direction for a while during sub-scanning, and when scanning line 14 is reached, pixels from XO to xl are in area 0, pixels from XI to X5 are in area 1, and x5.
Pixels from to Xn belong to area O, and so on. Here, a number such as area 0.1 is referred to as an area number.

Next, area selection signals S, . . . are generated for each area number, and area selection signals S5, . The point where image processing operations are performed separately will be explained. These area number string data are supplied to the area designation circuit 32 from the system controller 5 for each scanning line in the form of 32 bits x 297 word data. As described above with respect to the area specifying circuit 32, the given data string is expanded into a serial area number according to the scanning pixel position in the scanning line at the next timing, and is then stored in the area register group 51.
given to. By the function of the internal control means of this area register group 5, area registers a - d corresponding to area numbers are
One of them is selected, and the data in that register continues to be output as a processing selection signal S, . That is, when the data given by the system controller 5 is divided into 2-bit units and made to correspond to scanning pixel positions, the data in the area register corresponding to the divided 2-bit data is output in accordance with the scanning position.

The data in the area register group 51 is loaded in advance from the system controller 5 before the copy operation is started, and if the data held in the area register group 51 is made different from each other, different processing selection signals S sw can be generated for each area number. can get. Furthermore, if the data in the four area registers are the same, the same processing selection signals S and 3 will be obtained as a result.

The processing circuits 34 to 41 output one of the plurality of parallel image processing results in each circuit to the next stage in response to the processing selection signals S, 1, and image processing is performed for each designated region.

Finally, area O and area 1 are placed on the copied transfer paper 21.
Different images will be obtained.

To explain the details again, prior to the copy operation, images are transferred from the system controller 5 to the bus 42 for four areas (a-d) of the area register group 51 (two areas in this example), so at least two area registers a and b are stored. Write data according to the processing content. When the copy operation is started, the system controller 5 continues to send 297 words of 32-bit data to the area designation circuit 32 every l scanning lines. The sent data is alternately written to the toggle memory 53 or 54 through the bus 42, and data is read out from the other toggle memory 54 or 53 one word at a time for every 16 pixels, and each read word of 32-bit data is is divided into 2-bit units starting from the lower bit and supplied to the area register group 51 in synchronization with the pixel clock VCLK. These 2 bits select one of the registers a%d configuring the area register group 51, so for example, scanning line 1
3, the values of this 2-bit data string are all roJ between XO and Xl, all rlJ between XI and X2, all rOJ between X2 and x3, and all between X3 and Xn.
1'', everything between Xn and Xn may be set to rOJ. Note that the system controller 5 controls the image area designation circuit 3.
The data unit sent to 2 is 32-bit data in which 16 sets of these 2-bit data are arranged.

In addition, the written data in the toggle memory 53 or 54 is retained until it is rewritten, so when scanning lines with the same area number data continue, writing the 297 word data for each scanning line may be omitted. be. In other words, for a simple area such as a rectangle, the system controller 5 only needs to perform data transmission processing a few times out of 6720 scanning lines.
Furthermore, in order to obtain an area consisting of a smooth curve like a circle, it is necessary to send new area data almost every scanning line. However, with the same simple circuit configuration, it is possible to process curved areas such as circles with pixel-by-pixel smoothness.

[Detailed explanation of color correction circuit 38 1 The internal configuration of the color correction circuit 38 is shown in FIG.

First, four sets of color correction calculation ports 60a to 60d are provided. These color correction calculation circuits 60a to 60d each perform color correction calculation on each 8-bit RGB input, and
It outputs 8-bit values for each of MYK. On the output side of these color correction calculation circuits 60a to 60d, there is a multiplexer circuit 6 for selectively sending color data to the next stage circuit.
1 is connected. SEI is its selection signal input line.

Further, an internal bus 62 having the same function as the system controller 5 and an address decoder 63 having an internal register selection function are provided.

By the way, the four sets of color correction calculation circuits 60a to 60d have the same configuration, and one of them is, for example, the 14th color correction calculation circuit.
It is configured as shown in the figure. That is, it consists of a coefficient register section and a product-sum calculation section. In the coefficient register section, the subscript n of anxy represents any of the four types of parallel multiple processing a; b, c, and d, and X and y are the row number and column number of the color correction 7-trix calculation.

The color correction calculation circuit 60 executes the following product-sum calculation formula.

This equation is generally well known as a masking equation, and by appropriately setting coefficient values, it is possible to cancel out the illegal components contained in the CYK toner and obtain a beautiful full-color image. It is also easy to see from the above formula that it is possible to convert a full-color original image into monochrome, perform color conversion, and paint to fill in a specific color regardless of the original image. For example, if the coefficients from anxl to anx4 are made the same, the CMYK output will depend uniformly on RGB and will be a monochrome output.For example, if anly is set to a certain value and all from an2y to an4y are set to O, only C will be a monochrome output. It becomes a copy.

Also - as an example, anl 4. an24. If an34°an44 is set to a value other than O, and all other coefficients are set to O, it does not depend on the original RGB data and always remains constant.
MYK data is calculated and output. That is, it is painted. The color of the paint depends on the ratio of the four anxls; for example, if anl4 and an24 are 1 and the others are O, then C and M are in equal amounts, so they are painted blue.

There are 16 coefficient registers per set of arithmetic circuits, therefore,
Although there are four sets of 64 data in total, these data can be arbitrarily rewritten by the system controller 5.

As an example, let us consider a case where the character "4" shown in FIG. 12 is painted, and the remaining parts are subjected to normal full color processing. In this case, first, prior to the copy operation, the system controller 5 fills the 16 coefficient registers of the color correction calculation circuit 60a with full color processing coefficients, and the 16 coefficient registers of the color correction calculation processing circuit 60b with paint coefficients. Set it. Further, bits b15 to b related to color processing selection of the area register 51a of the area specifying circuit 32
"0" in the 4 bits of 12, "1" in the 4 bits of bits b15 to b12 related to color processing selection of the area register 51b
Set. Next, after the copy operation is started, 297 words of area switching data corresponding to the shape r4) continue to be sent for each scanning line as described above. In this way, the area specifying circuit 32 sends the first
In Figure 2, XO is rOJ, xl is rlJ, x2 is “0”, x
A signal is sent such as "l" at 3 and "O" at x4,
The inside of the ``4'' will be painted, and the rest will be given the usual full color treatment.

[Detailed explanation of color detection circuit 44] The internal configuration of the color detection circuit 44 is shown in FIG.

First, an internal bus 65 directly connected to the bus 42 is provided. Further, position registers 66a to 66cl that hold sub-scanning position data to be color detected and color registers 67a to 67d that store detected color information are connected via an internal bus 65.

With this configuration, the RGB data of four scanning lines at the sub-scanning position specified by the position register 66 is stored in the color register 67, and the stored data is arbitrarily read out by the system controller 5. Furthermore, any value can be set in the position register 67 by the system controller 5.

RGB signals sent from the scanner 1 are input to a color register group 67. In addition, each position register 66a to 66
d has a built-in counter that counts the line synchronization signal LSYNC and a comparator that compares and matches the output value of this counter with the position data set in the register. A trigger signal is issued to start data acquisition.

The triggered color register is R for 4752 pixels of one scanning line.
Store GB data. The system controller 5 can read any register at any time.
Original RGB data of the scanning lines of the book can be obtained.

[Explanation regarding region specification, automatic image region recognition, and selection of multiple image processing] Regarding the processing circuits 34 to 41, each circuit executes multiple types of processing in parallel, and only one result is used for the next processing. As mentioned above, what is sent to the circuit and which one is selected depends on the 8-bit signal input to the image processing selection means adeC. do.

The key point is that in the past, the operator could specify the image processing contents uniformly for all pixels in the designated area, or the image processing contents could be automatically switched based on the automatic image area recognition results. However, in this example,
The point is that the content of image processing can be determined by the combination of both signals.

a. When an operator performs a specific process on the entire surface of an original image For example, if conventionally, color conversion processing is performed on the entire surface, all halftone gradation images and characters within the area will be color-converted. On the other hand, in this example, color conversion processing is applied to the gradation processing part in the same way, but a copy can be created that preserves the original color even within the specified area for the text. be. In this case, you can do as follows. First, before starting copying, the color correction calculation circuit 6 shown in FIG.
The system controller 5 sets values for normal full color processing in the coefficient register group 0a, and values for color conversion processing in the coefficient register group of the color correction calculation circuit 60b. Furthermore, the data in the lookup table 64 is the input r1000.
Corresponding to XXXXJ, rQJ, ro 100XXX
Set "0" corresponding to XJ, and set rlJ corresponding to roOXXXXXXJ. Next, after the copy operation is started, C, CC, H,
The P signal is sent data in which one bit of C, CC, H, or P is N or 1 and the remaining three bits are rQJ, corresponding to the type of original image recognized from the automatic image area recognition circuit 31. come. At this time, the color correction calculation circuits 60a and 60b are performing normal processing and color conversion processing in parallel, and either processing result is dynamically switched according to an 8-bit signal input to the lookup table 64. and sent to the next stage processing circuit 39. The sent data is rooxXX when the value of the lookup table 64 is "1".
When it is XXXJ, that is, it is a part that is neither black characters nor colored characters. In this way, a copy in which only the picture portion is color-converted can be obtained.

b. When an operator performs specific processing on a specified area of the original image. For example, in the conventional case, if the area is specified on a tablet and color conversion processing is performed within the specified area, the halftone gradation image within the area is All text and text will be converted to color. In contrast, in this example, color conversion processing is applied to the gradation image part in the same way, but it is possible to copy text while maintaining the original color even within the specified area. It is something. in this case,
You can do as follows.

First, before starting copying, the system controller 5 sets values for normal full color processing in the coefficient register group of the color correction calculation circuit 60a, and sets values for color conversion processing in the coefficient register group of the color correction calculation circuit 60b. do. Furthermore, the data in the lookup table 64 is input rl 0OOXX
"○" corresponding to XXJ, ro 100XXXXJ
i: Corresponding site rQJ is set to ``○'' corresponding to rooxxooooo, 1, and ``1'' corresponding to roOXXXXXJ. In addition, 4 bits b15 to b1 output to the color correction processing circuit 38 of area register 5]a
2 value to rOJ, 4 bits b1 of area register 51b
Set the values of 5 to b12 to "1".

Next, after the copying is started, the system controller 5 tells the area specifying circuit 32 to specify r
OJ continues to send area switching data rlJ to the area to be color converted for each scanning line.

Then, the area designation signal a3゜aO of the color correction processing circuit 38
Naturally, data of "0" is sent to the area without color conversion, and data of "1" is sent to the area subject to color conversion.

In addition, independently from this, the automatic image area identification circuit 31 outputs C and CC corresponding to the type of the recognized original image.

One bit of either H or P is “1” and the remaining three bits are r
Data that is QJ is sent. At this time, the color correction calculation circuits 60a and 60b are performing normal processing and color conversion processing in parallel, and either processing result is dynamically processed according to an 8-bit signal input to the lookup table 64. The signal is switched and sent to the next stage processing circuit 39. The data to be sent is if the value of the lookup table 64 is “
1" is rooxxoooo", that is, it is within the specified area and is neither black nor colored text. In this way, a copy in which only the picture portion within the specified area is color-converted is obtained.

These are examples of stamps, and various processing can be performed, such as monochrome conversion of only black 0 characters and whiteout (conversion to white) of only black characters.

The same applies to processing circuits other than the color correction processing circuit 38. For example, the scaling/shifting circuit 34 performs magnification that differs between text and the picture area, and the image editing circuit 35 performs tilting processing only on the text and the picture part. The gradation processing circuit 37 applies a Laplacian filter to the text area and the smoothing filter process to the halftone image area.The color correction circuit 38 performs inversion processing only for colored characters, high contrast processing for black characters, and photo processing. The area can be subjected to solarization processing, and the halftone image area can be subjected to softening processing.

[Description of the console unit 6] On the panel surface of the console unit 6, there is a start button,
It consists of a dot matrix display 70 and a transparent matrix type touch switch 71 disposed above it, as shown in FIG. 16, along with push buttons (not shown) such as a 10-key button and a clear button. FIG. 16 shows an example of a display state in a specific mode of the copying machine, and the console unit 6 will be explained with reference to this figure. The seven character strings (rh)gJrtestJ, etc.) and button-like pattern in the illustrated example are the patterns displayed on the dotted line display 70. The button pattern indicates that the operator can perform touch input. If you touch it and the copying machine accepts the input operation, the "
The color changes as shown in "10g Button". The panel display system is hierarchical, and if you touch the "rback button," the screen at the next higher level will appear. Also, parts of the screen that cannot be displayed due to screen size constraints within the same hierarchy can be obtained by touching the "rmore" button. If you touch any button other than these two buttons, if there is a lower level screen, the lower level will appear. If there is no hierarchy below it and the button is the final instruction button, the color changes to "rlog button" and the copying machine starts the predetermined operation.

The screen shown in FIG. 16 is a service mode screen used mainly by people involved in maintenance of the apparatus, and the operating state of the copying machine related to this screen is referred to as the service mode. Here, the "rlog button" is a command button for printing out the total number of copies and the number of failures on the transfer paper 2I, and the "rtest button" is a command button for printing out the detection results of circuits 43 and 44.
adj, button" is a command button to print out various adjustment points inside the device, such as the output voltage adjustment value of the electrostatic charger, and rsample button J is a command button to print out settings that can be set by the operator, such as copy density and color tone. A command button for automatically creating a changed copy on one sheet of transfer paper 21 regardless of the setting screen (density setting screen or color tone setting screen), rdata
Button j is a command button for printing out image processing parameters overlappingly during normal copying, rc-dat
The a button J is a button for normally inserting title characters etc. during copying.

[Description of various operations in service mode] First, the operator selects a service mode screen as shown in FIG. 16 from among the hierarchical display screens of the console unit 6.

Next, by touching the desired data designation button for Noriharu and pressing the start button, various data will be printed out.

(Operation with "rlog button") Figure 17 is a data flow diagram showing various types of aggregated data, their flows, and programs that handle these data, and Figure 18 is the overall process when printing out aggregated data. FIG. 19 is a flowchart showing a program in the system controller 5 for printing out total data.

First, in FIG. 17, the system controller 5 includes an operating system (O8) program 75 for controlling the entire device, and an output program (this program is named rlog) for printing out aggregated data and adjustment setting values. ) 76, a control program 77 that performs sequence control and various timing controls of the device
There is.

Therefore, the total data 78 is updated and managed by the OS program 75 every time there is a copy or a failure. Also,
As the argument 79 that the OS program 75 passes to the log program 76, a value for specifying the range of processing of the log program 76 is entered. The return value 80 that the log program 76 returns to the OS program 75 is l
The identification code and error code of the contents requested by the og program 76 to the OS program 75 are entered. 81 is l
This is control data for operation control that the og program 76 passes to the control program 77. These data 78-8
1 exists in the battery-backed RAM inside the system controller 5; a82 is pattern data for character generation used by the log program 76, and basic data for generating various graphs such as bar graphs, pie graphs, and line graphs; is stored within. For example, character data is in vector format, which requires an extremely small amount of data compared to bitmap format, and can be output with arbitrary character shapes and sizes.

FIG. 2o shows a document 83 for printing out only specific data from the total data, and FIG. 21 shows an example of printing out the total data onto the transfer paper 21.

Comparing the original 83 shown in FIG. 20 and the printout shown in FIG. A thick rectangular number string 85 that is copied and does not exist in the original 83 is newly inserted.

These number strings 85 are visual representations of the aggregated data in the form of numbers that can be read by humans. That is, the printed transfer paper 21 is a composite copy of an enlarged image of the original image (original 83 image) and an automatically generated number string.

The operation for obtaining such a printout will be described below. When the rlog button J is pressed on the service mode screen as shown in FIG. 16, the color inside the button changes, indicating to the operator that the mode has shifted to the log output mode. Thereafter, by performing the processing shown in FIG. 18, a printout as shown in FIG. 21 is obtained. Referring to this figure, the operator selects a document 83 as shown in FIG.
21 on the platen 12 and then press the start button to obtain a print as shown in FIG.

Note that these manuscripts 83 are referred to as log cards. Fig. 20 is a log card for outputting the number of copies made by this copying machine by size. An 8-bit black and white pattern code 84 is attached to the leading edge of the original 83. The parts are pre-printed so that they can be combined and printed when printing out the size-based aggregate data. The card (original 83) is placed in such a way that the direction in which the 8-bit patterns are arranged coincides with the scanning direction of the scanning lines. Similarly, a log card or the like for outputting failure totals for each part is possible, and in this case, the black and white pattern codes 84 will be different.

When the start button is pressed, the OS program 75 in the system controller 5 switches to service mode lo.
Since we know that it is a g action, first, we set the argument 79.
is set as the request code for outputting the total data, and the log program 76 is called and executed. Note that in order to print out one sheet of total data, it is necessary to call and execute the log program 76 multiple times. The log program 76 performs different processing when called multiple times. The processing content on the call count side is shown in FIG.

The log program 76 called for the first time calculates and passes control variables 81 necessary for the color detection operation to the control program 77. Further, distance data from the leading edge of the document 83 to the pattern code 84 is provided to the color detection circuit 44 . That is, the sub-scanning position data is stored in the color registers 67a to 67a of the color detection circuit 44.
Set it to one of 67cl. Finally, a request for original image scanning operation is returned to the OS program 75 as a return value of 80.

After receiving the return value 8o, the OS program 75 gives a scan operation command to the scanner 1, and after the reading operation of one original image (original 83) is completed, the OS program 75 starts the log program 76 again.
call.

Referring to FIG. 19, the log program 76 called for the second time stores all pixels of one read scanning line at the above-mentioned sub-scanning position in one of the color registers 67a to 67d of the color detection circuit 44 after scanning the original. Since RGB data has been stored, the data in the color register is read, the mark pattern code is checked, and it is determined whether the code is correct or incorrect. If the black of the original 83 is "1" and the white is "o", then the code shown in FIG. 20 can be read as "rloololol", for example. If the code is invalid, an error code is returned as a return value of 80. An invalid case is a case where the code to which aggregated data is output cannot be detected. For example, the pattern code of the log card could be read, but rloloolo
1) or the pattern code cannot be detected (
rooo.

There are several types, such as 0000J).

The OS program 75, which received the illegal code, causes the console unit 6 to display the message ``Log card is incorrect'' and prompts the operator or operator to perform the operation again. if,
If the operator does not have the correct log card, information 78 inside the copier cannot be obtained.

If the pattern code 84 is determined to be correct, the information to be output corresponding to the determined code is determined and specified. For example, if the code is rloloololJ, the total number of copies by size is displayed. Copy summary data for each size exists in 78, and a special color paint pattern is generated to make these data visible, and it is temporarily stored in the memory unit 3 (or it can also be directly printed out). be). In other words,
To create a paint pattern such as the number string 85 shown in FIG. 21 from the total data 78, color processing coefficients for painting a number string of a desired color are stored in the coefficient register in the color correction calculation circuit 60a of the color correction circuit 38. Similarly, in order to make the background part of the number string white (blank), the coefficients for white painting, i.e., rQJ, are loaded into all the coefficient registers in the color correction calculation circuit 60b.Also, Bits b15 to b13 of the area register 51 of the area specifying circuit 32 are loaded with values rQJ and rlJ so that a blank area (painted with no color) and a desired color paint area can be selected. In order to appropriately switch between two areas, it is necessary to provide area switching data in units of scanning lines after screen processing has started, but the area switching data for all these scanning lines is calculated in advance by the log program 76. Prepared as control data 82.

Finally, the log program 76 returns the ID number (identification number) of the identified log card and the request code for the image processing operation to the OS program 75 as a return value 8o.

OS that received the log card ID number and image processing request
The program 75 first controls the display of only the data portion corresponding to this ID number in the total data 78 on the console unit 6, and then calls the control program 77.
An image processing cycle for one page is performed. During this one page image processing operation, the control program 77 uses the control data 8
2, the area switching signal continues to be given to the area specifying circuit 32. Then, regardless of the image data input to the color correction circuit 38, the area of the number string pattern is
It will be painted in the color set to 8, and the base of the number will be blank. In addition, the image data S3 subjected to image processing in this way
is stored in memory unit 3. Note that the memory unit 3 operates in the storage mode, which is the second operation mode, by command operation of the OS program 75. When this process is completed, the oS program 75 calls the log program 76 for the third time.

The log program 76 called for the third time, if
If there are no more number strings, CMY of memory unit 3
The OS program 7 sets the operation request for image reading, original image scanning, and image data synthesis copying cycle to a return value of 80.
Return to 5. Image processor 2 magnification/shift circuit 3
4 is the pattern code 84 part of the log card on the transfer paper 2
1. Set a movement parameter that will be deleted when it goes outside, and a parameter that has a magnification of 2x.

If you want to overwrite another number string in the memory unit 3, the same process as when called for the second time is performed, and the least significant bit b becomes the memory override signal for the area register group 51.

``rl'', return the same return value 80 as in the case described above, and end the process. In such a case,
For example, this problem occurs when the color correction circuit 38 is capable of processing four types of colors at the same time, but you want to use more types (four or more) of different colors for the number paint on the printout of total data. . That is, for example, in the case of painting five different colors, first paint three colors and blank, then paint the remaining two colors and blank, and then create a number string pattern painted with each different color. can be synthesized by the memory unit 3.

If the return value 80 is an image processing request as in the previous call, the OS program 75 repeats the same process as in the previous call.

If the return value 80 is a request for a composite copy of the CMY image data of the memory unit 3 and the data of the original image, an operation command signal is output to the scanner l and printer 4, and the memory unit 3 is set to the third operation mode. The data read mode is activated. Further, after the start of the operation, the control program 77 is called to form a visible image on the transfer paper 21 by combining the CMY data in the memory unit 3 and the K (black) data of the original. Note that the K data is set to a parameter of 2x enlargement by the scaling/shifting circuit 34, so the image is enlarged to the original image, and the CMY in the memory unit 3 is
The images are combined as they are.

On the upper side, there is a step of reading the document, a step of recognizing the pattern code, generating a character string in CMY colors and storing it in memory, a step of re-reading the document, reading out the image and the CMY image in the memory, and composing the image and the CMY image. It requires at least three steps, including the step of printing out the information. This is possible even if the program execution speed of the system controller 5 is relatively slow, and there is a need to form an image on a single sheet of transfer paper 21 that exceeds the number of parallel processes of the image processor 2, for example, the number of paint colors. This is to satisfy the need.

If there are no such restrictions or requirements, it is possible to make the process simpler and faster. In other words,
The memory unit 3 is activated to the first mode (delayed operation of each CMY image data), and the scanner 1 and printer 4 are operated. Then, the scanner 1 scans the pattern code 84 portion at the leading edge of the document, and sends the original image data 1 to the color detection circuit 44.
Immediately after the scanning line segments are stored, they are decoded, the total data to be output is determined, and the color processing parameters and area for generating the number string pattern are determined just before the sub-scanning advances to the position where the total value is to be recorded. After completing the processing parameter setting and region switching data generation processing and reaching the number string generation position, the region switching signal may continue to be applied until the end of original image scanning.

Furthermore, if the pattern code of the log card includes a department code, etc., it is easy to output accounting management information for each department. The pattern code 84 in the illustrated example is a black and white 8-bit code to simplify the explanation, but since the color detection capability of the color detection circuit 44 is four scanning lines, the log card (original 83) is placed on the platen 12. Even if you take into account the slight deviation when placing a log card, it is easy to have several hundred bits of information, making it virtually impossible to forge a log card through trial and error. In particular, it would be even more perfect if the pattern section of the log card included color information by utilizing the RGB color detection function.

In this way, when the code portion of the log card is made complicated, it becomes possible to display or print out the information 78 in the copying machine on the console unit 6 using a different operation method. This takes advantage of the fact that, as described above, there is almost no complicated pattern code image in a typical original to be copied. Then, without switching the console screen to the service mode screen as described above, the document is first scanned to read the document information in the general copy mode, the pattern code is recognized, and if it is not a log card, the second screen is scanned. A copy image may be formed at the time of the second scan, and if it is determined to be a log card, information corresponding to the ID number contained therein may be output. Scanning for reading document information is generally called a pre-canning method, and is widely used for detecting document size and the like.

Note that although the above-described printout pattern is an example of number pattern generation, it may also be a graph or the like. In short, it is important that the shape, color, and pattern be accurate and easily readable by the operator. Also,
These various processes are possible by setting appropriate parameters in the processing circuits 34-41.

(Operation with "rtest button") When the operator touches the "rtest button" on the service mode screen shown in Figure 16, multiple sensors to be tested will be displayed, and a sensor specification screen will appear where one of them can be specified. Now you can see that it has changed to test mode.

This test mode is a mode in which the operation results of the various detection means of the copying machine are printed out so that the operator (serviceman) can easily judge the accuracy of the detection operations of the various detection means of the copying machine.

Here, the color detection circuit 44, the size detection circuit 43, and the potential sensor (not shown) of the photoreceptor 15 are used as the detection means, and FIG. 22 shows the printout results in the test mode.

First, touch the "rtest button" to display the sensor specification screen, and then touch one of the sensor specification buttons.
When the start button is pressed, the system controller 5 activates the detection operation of the designated sensor to be tested, reads and prints out the detection results of that sensor. For example, when the sensor input by touch on the sensor specification screen is a photoreceptor potential sensor, the system controller 5 issues a command to the printer 4 to perform one image forming process and read the potential through 84, and then finishes the process. After that, time series data of potential detection is received from the printer 4.

Next, run a program that expands this data into graphs, characters, and numbers, and load the image processing parameters corresponding to the white background and the image processing parameters for the parts that will become graphs and characters into the image processor 2. , area switching data corresponding to the numerical shape is calculated in advance. Next, the image processor 2 and the printer 4 are operated, and the previously prepared area switching data is sequentially outputted to the area specifying circuit 32 of the image processor 2, and the second area switching data is printed on the transfer paper 21.
2. Form a printout as shown in Figure 2(c).

For example, if the touch-designated sensor on the sensor designation screen is a document sensor, the scanner I repeats scanning three times in total when the start button is pressed. During the second and third document scanning, the printer 4
When the second scan is completed, a printout of the operation result of the color detection circuit 44 as shown in FIG. 22(a) is obtained, and when the third scan is completed, a printout is obtained as shown in FIG. 22(b). A printout of the operation results of the size detection circuit 43 can be obtained. Note that the document to be detected is set on the platen 12 before pressing the start button.

Referring to FIG. 22(a), the color detection result information is rct
Heading 86 painted with the test code shown in estlJ, RGB shown in each of the character strings 87a to 90a
Paint string consisting of heading and value, 87b-90
It is output as a rectangular paint frame centered on the detection position indicated by b. Note that the same image as the original on the platen 12 is copied except for these parts to be painted. That is, an image in which the original image and the detection information are combined is obtained. The center of the paint frame 87b coincides with the first detection position, and its color is painted with the detected color. The paint character string 87a is the RGB component value of the detection result data. Since the center of the paint frame 88b is a copy of the original, the inside of the paint frame 88b and the paint frame 87b itself are visually compared, and if they match, it is known that the detection circuit has operated correctly. color detection circuit 4
It is possible to discover a failure in the color correction circuit 38 that paints the 4 or frame.

In such a case, it is possible to further examine the painted number string data for each RGB color of the painted character string 87a to determine whether the color detection circuit 44 has malfunctioned or the color correction circuit 8 has failed. There are three other detection locations, and information is output as in the previous section.

Referring to FIG. 22(b), the information on the size detection results includes the heading 91 where the test item is painted, the platen 12
Paint frame 92 showing the planar shape of , portions 93 a and 93 b showing the size (number of pixels) of the detection results in each direction of main scanning and sub-scanning by painting them in number pattern rows, and the detection data are the same as the paint frame 92 The portion 94 painted on a proportional scale is printed out. Therefore, by comparing these painted patterns and number strings with the actual document on the platen 12, the operator can judge whether the detection accuracy is good or bad.

Note that 95 is an image portion obtained by copying the original 13 on the same proportional scale as the platen paint frame 92.

In order to match one corner of the platen paint frame 92 with one corner of the copy image, image movement processing is performed using the magnification/shift circuit 34. Therefore, although the image painted portion 95 is shown in black for convenience in the illustrated example, it is actually an image similar to the original image.

In this way, there is no need to compare the actual original 13 on the platen 12 and the printout result, and it is convenient to conduct tests all at once even when testing different original images multiple times.

Next, the operation will be explained.

First, when the test button is pressed, the system controller 5 gives the sub-scanning position parameters of a predetermined color detection position to the color detection circuit 44, and clears the size register (not shown) inside the size detection circuit 43. .

When the start button is pressed, the system controller 5
outputs a scan start command to the scanner 1, and the scanner 1 scans the original 13. After completing the scan, the color detection circuit 44
The color registers 67a to 67d store RGB data for each one scanning line corresponding to the sub-scanning positions set in advance in the position registers 66a to 66d, and the size register of the size detection circuit 43 stores RGB data in the main scanning direction. Original size value (
The number of pixels) and the size value in the sub-scanning direction are set as the detection result.

Since the values of these color registers 67 and size registers can be arbitrarily read by the system controller 5, they are read and temporarily stored in the RAM, which is a work memory within the system controller 5. After that, processing for printing out the color detection result information onto the transfer paper 21 is performed.

The information is printed out by the area specifying circuit 32 outputting a shape area signal such as a number string, character string, frame, etc., and the color correction circuit 38 receiving this signal painting the area in a predetermined color. This process is the same as the operation when operating the "1og button" described above, so a detailed explanation will be omitted.

Briefly, the system controller 5 calculates a character string, a number pattern, and a frame shape from the color detection data in the work memory, and stores them in the work memory as area switching data. Further, the color correction circuit 38 is set to parameters for normal processing and paint processing, and the normal color processing selection and paint processing selection numbers are set in the area register of the area designation circuit 32.

After completing such processing, the system controller 5
again issues the first operation start command to the printer 4, causing it to perform a series of copy cycles. While the image processor 2 is processing the image, the system controller 5 continues to supply area switching data prepared in advance in the work memory to the area specifying circuit S2 without updating it for each scanning line. When this cycle is completed, a printout as shown in Figure 22(a) is obtained.

Thereafter, the shape of the number string, character string, and frame to be generated is calculated from the size detection data in the work memory, and the parameters are loaded into the color correction circuit 38 and the area designation circuit 32 as in the previous cycle. Furthermore, parameters for a predetermined magnification and movement amount are set in the magnification/shift circuit 34.

When these preparatory processes are completed, the system controller 5 issues a third scan command to the scanner 1 and a second print scan command to the printer 4 to execute the copy cycle, as shown in FIG. 22(b). ) will produce a printout like the one shown in ).

(Operation with radj, button J) The operator touches radj, button J in FIG.
is placed on the platen 12 and the start button is pressed to obtain a printout as shown in FIG. 22(C).

This is a visual record of the adjustment values of the adjustment points of the copying machine. Examples include a scale representing a displacement from a standard delay amount of M, Y, and C image data in the first operation mode of the memory unit 3, that is, a scale representing an adjustment setting value of the delay amount, and a scale representing an adjustment value of register timing. The document in this case is called an adj card, and at its leading edge is provided a black and white pattern with a code corresponding to 8 bits, with black as "1" and white as rQJ. There are a plurality of types of such adj cards, and each type is set to have a different black and white code. Further, in addition to the code portion, characters and memory are printed on each adj card. These printed portions become enlarged copies when printed on the transfer paper 21. In short, adj is a composite copy of the enlarged image of the card and the paint pattern generated internally in association with the adjustment value.

The operating procedure and processing procedure are almost the same as those for the "rlog button" operation. That is, the log program 76 in FIG. 17 also has the function of outputting adjustment values.
The data 78 has a data structure in which these data to be output are added. The OS program 75 uses these data 7
8, and when calling the log program 76, the argument 79 must be a value indicating that it is in the range of adjustment value output, unlike when outputting aggregated data. Other than this, the operation is basically the same as the operation with the rlog button J, and the explanation will be omitted.

<Operation with "rdata button"> When the "rdata button" shown in FIG. 16 is touched, the color inside this button changes, and when it is touched again, it returns to the original color. When the color changes, it indicates that the data mode is activated. In the other service modes mentioned above, even though the original may be copied, the main purpose is not to copy the document itself, but to print out internal information using the paint function. On the other hand, this d
The purpose of the ata mode is to adjust the magnification, density, color tone, etc., and then add these adjustment values to the copied image, as in normal copying work.

Therefore, even if you touch the "back button" with the color of the "rdata button" changed on the console screen in Figure 16 and move to a higher level screen, the data will not be displayed.
The mode attribute is set to remain preserved. When a normal copy operation is performed with the data mode attribute attached, a color tone adjustment value of 100 is printed in the upper right corner as shown in FIG. 22(d). This adjustment value 10
0 uses the paint function of the color correction circuit 38 like the other modes, and each print part 1 except for the adjustment value 100
01 etc. are completely normal original copy images. In this example,
The reason why only the color tone adjustment value is output and other adjustment values such as density and magnification are not output is because they are still set to standard values. Although it is possible to output all of the information, it has the disadvantage of being complicated.

(Operation of the rC-data button) The re-data button shown in Figure 16 is divided into two parts, left and right. A screen similar to a typewriter keyboard will appear, allowing you to enter the string to be added.

When the right side of this button is touched, a screen as shown in FIG. 23(a) appears, and character string modification input that determines the printing mode of the input character string becomes possible.

Additional character strings entered using these screens are
The pattern will be "rTITLE" 101 shown in Figure (d).

First, when the operator touches the left side of the "re-data button" on the screen shown in Figure 16, a character string input screen appears.
In addition to the keyboard screen, a message saying ``Please specify the print position on your tablet'' is displayed. Then, the operator uses the tablet to enter the upper left point of the character string that is about to be entered, and then selects "rTITLE".
Enter the string .

Then, by touching the "r back button" on the same screen, the screen returns to the screen shown in FIG. 16. This time, if you touch the right side of the ``rc-data button'', a character string modification screen as shown in FIG. 23(a) will appear. Here, the copy paper range indicated by 102, the previously input character string 103, and the specified position 104 are displayed and can be confirmed. The font, size, etc. of the characters are determined by the modification information that the system controller 5 has as default. If this is fine, press the start button to obtain a print in which this character string is combined with the copied image.

If you wish to change it, touch the modification button 105 xlxx4 corresponding to the modification item shown in FIG. 23 (m) and perform the modification operation. In FIG. 23(C), 106 is a modification item heading when changing the character string size, 107 is a size scale and current value, and 108 and 109 are reduction/enlargement buttons.

Various types of image processing copy operations and operations by image area> Figure 24 shows the display screen of the console unit 6 of the Lebel in the image processing mode, and shows the processing type designation screen for specifying and inputting the type of processing content. becomes. In the figure, 111 is outlined, 112 is shaded text, 113 is bold, 11
4 is a button for instructing italicization, 115 is a button for instructing conversion of shaded text, and 116 is a button for instructing double-width text. That is, it is an input means that automatically performs these image processing processes only on the character portions recognized by the automatic image area recognition circuit 31. Also, 117 is solarization, 118 is bossalization, 119
is a button for high key, 120 is a button for high key coarse graining, 121 is a mirror, 122 is a button for pseudo colorization, and automatic image area v! ? It serves as an instruction means for performing image processing only on the continuous tone image portion and the halftone image portion, that is, the light and shade portions recognized by the recognition circuit 31. 123 is color conversion, 124 is negative inversion, 125 is image area scaling, 126 is deletion (blanking), 12
Reference numeral 7 indicates a monochrome button, and 128 indicates a sharpness button, which is an instruction means for performing image processing on a text portion or a grayscale image portion as selected by the operator.

The various processing processes described above are classified into those whose processing specifications are predetermined (for example, negative inversion) and those whose detailed processing specifications can be designated by the operator (for example, sharpness). Detailed specifications can be specified and a processing image area can be determined using buttons such as the high-key coarse graining button 12.
When you touch 0, the screen immediately moves to the detailed processing specification screen, and you can select buttons for image area, such as color conversion button 123.
When touched, the screen is moved to the detailed machining specification screen via the image area selection screen as shown in FIG. 25(a).

First, FIG. 25(a) is a console screen at a lower level of image processing in the processing mode, which appears when the processing type of buttons 123 to 128 is specified in the second level. This screen allows you to specify where to apply the specific image processing to C, CG, P, black text, color text, continuous tone area, and halftone image area, respectively.
This is a screen for specifying with the H button 130.

Further, FIG. 6B shows an example of a console screen at a different lower level in the image processing mode, which is a detailed processing specification screen for inputting detailed specifications of the processing specified by the processing type in the first label. In other words, there are as many detailed specification screens as there are machining types for which detailed specifications can be specified. In the figure, 131
is an indication of where to perform processing in black text, color text, continuous tone area, and halftone image area. In the illustrated example, rP and HJ are displayed, so it is applied only to continuous tone area or halftone image area. Processing is performed.

FIG. 2C shows an area screen when an area is input using the digitizer unit 7.

a. Designation of image processing contents based on automatic image area recognition results As described above, image processing contents can be made different for each image area based on automatic image area recognition results. Here, we will explain how to specify such different image processing contents.

First, the type of processing is specified on the image processing type screen shown in FIG. If the image area has been determined and detailed specifications do not need to be specified (defaults are acceptable or cannot be specified), a processed copy can be obtained by pressing the start button here.

In the type of processing that specifies an image area, select the desired one using the C, CG, P, and H buttons 130 that express the recognized image area of the automatically recognized image area on the console screen in FIG. 25(a). touch. These buttons 130 represent black text, color text, photographs, and halftone image areas, respectively. Among these, buttons whose colors are inverted indicate that processing other than standard image processing will be performed.

On this screen, if one of the four buttons 130 is touched, the screen shifts to a screen for specifying detailed processing specifications for the image area indicated by that button.

Figure 25(b) shows the console screen in Figure 23.
This is a detailed specification screen that appears when the high-key coarse graining button 124 is pressed. rP and HJ of the display area 131 indicate that processing is performed only on the continuous tone image (photo) and the halftone image area.
This can be confirmed by the display.

On this screen, buttons 132 from Xl to X8 are buttons for specifying subdivisions of processing necessary for high-key coarse graining, which are subdivided into processing ranges such as gradation processing, spatial filter processing, and halftone processing. Touching one of these buttons 132 causes
Same mark as that button 133, scale & pointer 134, pointer 1
Buttons 135 and 136 for moving 34 left and right appear. In the illustrated example, the gradation processing button is touched and the scale & pointer 134 displays a maximum image density scale and a pointer indicating the default density of high-key coarse graining. In other words, the pointer swings considerably to the left, so it can be seen that the density is considerably lower than the center in standard processing. If you want to shift left and right instead of the default, press the 10, - button 1
35 or ] 36. The color of the process specification button for which non-standard processing is specified is reversed. Also, rbac
When the k button J is touched and the screen shown in FIG. 25(a) is returned, if there is even one image processing item in a non-standard state, the button color in that area will be displayed in reverse.

In this way, it becomes possible to specify the image processing content for each image area and for each image processing means for each image area, and to visually confirm it.

After this, when the start button is pressed, the system controller 5 loads predetermined image processing parameters into the image processor 2, and copies of different image processing are obtained for each area.

b. Specification of image processing content based on combination of automatic image area recognition results and specified area When the operator specifies an area using the tablet of the digitizer unit 7, the area shape is displayed on the console screen. FIG. 25(c) shows an example in which two areas, a circular area 137 and a rectangular area 138, are input. These areas 1
37, 138, the system controller 5 automatically
It is given the names (area numbers) l and a2 and is displayed approximately in the center of the area shape display. Areas other than these two areas are given the area name aO and displayed as shown at 139.

If you touch the inside of these areas aO-a22, the screen changes to the one shown in FIG. 24, and in a specific designated area, you can further designate a specific type of image processing.

FIG. 24 shows al which is the circular area 137 in FIG. 25(C).
If the specified area does not exist, ralJ is displayed in the area 139 where there was no display, and it becomes possible to confirm that the area is to be subjected to specific processing and its area number. On this screen, c, cc, p.

If any of the H buttons 130 is touched, the screen for specifying each image processing means as shown in FIG. 25(b) appears as in the case where no area is specified, and the same operations as before can be performed.

In this way, it is possible to specify different image processing for each designated area, for black text, color text, photographs, and halftone dot images within a specific designated area, and to obtain a copy image as a result.

In addition, if the above a, and are expressed as an image, it will be as follows. First, FIG. 26 shows an example of a document 13 having a grayscale image area 141, a red ellipse 142 in the grayscale image area 141, and a red character string 143. Using manuscript 13 like this,
If only the grayscale image part is subjected to red/blue color conversion processing in mode a1 above, a copy image as shown in FIG. The red ellipse 142 is copied as a blue ellipse image 145a. Furthermore, when using the document 13 in FIG. 26 and performing red/blue color conversion processing on only the grayscale image portion within the area 146 in the mode described above, as shown in FIG. 27(b), even if the area 1
46, the character string 143 part remains the original red image 144, and the specified area 1 in the two red ellipses 142
46 becomes a blue ellipse image 145a, while those outside the region 146 remain red as an ellipse image 145b.

Effects of the Invention Since the present invention is configured as described above, according to the invention according to claim 1, the image area processing parameter input means determines the first processing content or the second processing content of the image processing means. Parameters are input, and the control means selectively changes or energizes the first processing content or the second processing content of the image processing means in response to the parameters input from the parameter input means. The content of image processing can be set arbitrarily, and the degree of freedom in the content of image processing can be increased so that faithful copies of many types of original images can be reproduced.

In this case, according to the second aspect of the invention, the processing parameter input means inputs the parameters of the magnification and image movement amount for each image area, and the control means energizes the image processing means based on these parameters. The scaling/shifting means in the image processing means performs a plurality of different kinds of scaling processing and image movement processing, and the image processing selection means generates a selection signal within the specified area, so that characters and grayscale images within the specified area are It is possible to make copies with different copying magnifications or copying positions, increasing the degree of freedom.

Similarly, according to the third aspect of the invention, the processing parameter input means inputs the mosaic pitch, mirror selection, slanting angle, shading parameters, and contouring parameters for each image area, and the control means inputs the parameters. The image processing means is energized based on
Multiple types of processing are performed simultaneously, such as slanting images at different angles, shading with different lengths, colors, densities, and shapes, and creating contour images with different line widths and colors. Since it emits a selection signal, it is possible to create copies with different modes of mosaic imaging, mirror imaging, slanting, shadow imaging, and contour imaging between the characters and the grayscale image part in the specified area.
This increases the degree of freedom.

Further, according to the invention described in claim 4, the processing parameter input means inputs filtering parameters for each image area,
The control means activates the image processing means based on this parameter, and the spatial filtering means in the image processing means performs spatial filtering with different filtering coefficients, such as smoothing processing with different degrees of smoothness, sharpening processing with different sharpness degrees, etc. Since the processing is performed simultaneously and the image processing selection means issues a selection signal within the specified area, it is possible to form copies with different filter effects such as smoothness and sharpness between the characters within the specified area and the grayscale image area, increasing the degree of freedom. It will increase.

According to the fifth aspect of the invention, the processing parameter input means can select blanking for each image area, paint color and density, contrast, density, gradation inversion selection, gradation partial inversion selection, and gradation omission selection. inputting a parameter for the number of gradations, the control means energizes the image processing means based on these parameters;
The gradation processing means in the image processing means performs blanking with a plurality of different types,
Painting an arbitrary color, contrast conversion, density conversion, gradation inversion, gradation partial inversion, and gradation omission are performed, and the image processing selection means issues a selection signal within the specified area, so that the characters and grayscale images within the specified area are You can create different copies with respect to blanking, painting, contrast conversion, gradation conversion, gradation inversion, gradation partial inversion, and gradation omission, increasing the degree of freedom.

According to the invention set forth in claim 6, the processing parameter input means inputs the type of color correction for each image area, the type and degree of undercolor removal, density, blanking selection, paint color and density, and original color of color conversion. Parameters for the color after conversion, monochrome color and density, and undercolor color and density are input, and the control means energizes the image processing means based on these parameters, and the color processing means in the image processing means A plurality of different color processes are performed, and the image processing selection means issues a selection signal within the specified area, so that one color of the image is processed with specific color correction and undercolor removal processing that differs between the characters and the grayscale image part within the specified area. Copies with different conversion images, monochrome images, undercolor images, blanking, and arbitrary color painting can be created, increasing the degree of freedom.

According to the invention set forth in claim 7, the processing parameter input means inputs the halftone processing parameter for each image area, and the control means energizes the image processing means based on this parameter, thereby controlling the halftone processing in the image processing means. The processing means processes a plurality of different halftone dot shapes, halftone dot sizes, and halftone dot directions, and the image processing selection means issues selection signals within the specified area, so that characters and grayscale image parts within the specified area are processed. This allows copies with different halftone dot shapes, sizes, and screen orientations to be created, increasing the degree of freedom.

Furthermore, according to the invention as set forth in claim 8, the processing parameter input means inputs image processing parameters for each image area, the control means energizes the image processing means based on these parameters, and the image reading means reads Based on the image data and area signal, the image processing selection means selects a black character area within the specified area,
Since different selection signals for the first to fourth image processing are generated for each of the color text area, continuous tone image area, and halftone image area, the black text area, color text area, and continuous tone image within the specified area are generated. It is possible to obtain a copy image in which image processing is performed on the image area side according to the parameters specified for each of the halftone image portion and the halftone image portion, thereby increasing the degree of freedom.

[Brief explanation of the drawing]

The drawings show one embodiment of the present invention, and FIG. 1 is a block diagram showing a first example of the configuration near the image processor;
FIG. 2 is a system block diagram showing the overall configuration, FIG. 3 is a schematic front view showing the overall configuration of the copying machine, FIG. 4 is a block diagram showing a second example of the configuration near the image processor, and FIG.
The figure is a block diagram showing a third example of the configuration near the image processor, and FIG. 6 is a block diagram showing a fourth example of the configuration near the image processor.
FIG. 7 is a block diagram showing a fifth example of the configuration of the image processor; FIG. 8 is a block diagram showing the common concept of each processing circuit; FIG. 9 is a block diagram of automatic image area recognition means. , Figure 10 is a block diagram of the image area designation circuit, Figure 11 is a block diagram showing the area register configuration, Figure 12 is an explanatory diagram for explaining the operation of the image area designation circuit, and Figure 13 is color correction. Circuit block diagram, 14th
Figure 15 is a block diagram of the color correction calculation circuit, Figure 15 is a block diagram of the color detection circuit, Figure 16 is a plan view showing the service mode screen of the console unit, Figure 17 is a data flow diagram showing processing, and Figure 18 is a block diagram of the color correction calculation circuit. 19 are flowcharts, FIG. 20 is a plan view showing an example of a log card, FIG. 21 is a plan view showing an example of printout in the log mode, FIG. 22 is a plan view showing various printout examples, and FIG. 2
3 to 25 are plan views showing examples of various console screens, FIG. 26 is a plan view showing an example of a manuscript, and FIG. 27 is a plan view showing an example of its printout. 1... Image reading means, 2... Image processing means, 4.
... Image forming means, 5... Control means, 6... Image area specifying means, processing parameter input means, 21... Recording medium, 31... Automatic image area recognition means, 32... Image area Specifying means, 34... Magnification/shifting means, 35...
Image editing means, 36.40... Spatial filtering means, 37° 39... Gradation processing means, 38... Color processing means, 41... Halftone processing means 3z, %20 loss

Claims (1)

[Claims] 1. Image reading means for reading an original image by dividing it into pixels;
In a copying machine equipped with an image forming means for forming image data as a visible image on a recording medium, separating and determining at least two regions of an original image, a character region and a grayscale image region, from the image data read by the image reading means. automatic image area recognition means that performs first image processing on the character area recognized by the automatic image area recognition means on the original image data read by the image reading means, and second image processing on the grayscale image area.
an image processing means for selectively performing at least two types of image processing; and a process for inputting parameters that determine the first image processing content or the second image processing content of the image processing means. A parameter input means and a control means for selectively changing and energizing the first image processing content or the second image processing content of the image processing means in response to the parameters input from the processing parameter input means. An image processing device for each image area of a copying machine, characterized in that: 2. The image processing means is provided with a magnification/shifting means that performs at least one of image processing such as image magnification of a desired magnification ratio and image shift of a desired amount of movement for each automatically separated image area. 2. The image processing apparatus according to image area for a copying machine according to claim 1, wherein at least one parameter of a magnification ratio and an image movement amount is input and set from a processing parameter input means. 3. In the image processing means, mosaicization of a desired pitch for each automatically separated image area, mirror image/non-mirror image formation, image slanting at a desired angle, and shading of desired length, color, density, and form. An image editing means is provided which performs at least one of image processing, such as image processing and contour image formation with a desired line width and color, and at least one of the mosaic pitch, mirror selection, oblique angle, shadow casting parameter, and contouring parameter for each image area. 2. The image processing apparatus for each image area of a copying machine according to claim 1, wherein two parameters are input and set from a processing parameter input means. 4. In the image processing means, spatial filtering processing is performed for each color or by arbitrary filtering coefficients to perform image processing such as smoothing processing to a desired degree of smoothness and sharpening processing to a desired degree of sharpness for each automatically separated image area. 2. The image processing apparatus for each image area of a copying machine according to claim 1, further comprising a spatial filtering means for applying common filtering to all colors, and a filtering parameter for each image area is input and set from a processing parameter input means. 5. During the image processing means, perform at least one image processing of blanking, painting with arbitrary color, contrast conversion, density conversion, gradation inversion, gradation partial inversion, and gradation omission for each automatically separated image area for each color. A gradation processing means that can be applied separately or commonly to all colors is provided, including blanking selection for each image area, paint color and its density, contrast, density, gradation inversion selection, gradation partial inversion selection, gradation omission selection and its gradation. 2. The image processing apparatus for each image area of a copying machine according to claim 1, wherein at least one parameter of the tonality is input and set from a processing parameter input means. 6. In the image processing means, color correction processing using desired correction coefficients for each automatically separated image area, arbitrary amount of undercolor removal, K plate reprinting processing, density conversion, blanking, arbitrary color painting, arbitrary color A color processing means is provided that performs at least one of the following processes: color conversion, monochromatic imaging of arbitrary colors, and undercoloring of arbitrary colors, and the type of color correction for each image area, the type and degree of undercolor removal, density, and white space. At least one parameter of color selection, paint color tone and density, original color and converted color for color conversion, color tone and density for monochrome conversion, and color tone and density for undercolor can be input and set from the processing parameter input means. 2. The image area-based image processing apparatus for a copying machine according to claim 1. 7. In the image processing means, halftone processing such as halftone shape processing, halftone dot size, halftone direction, etc. is performed for each color or all colors to express the desired pseudo halftone for each automatically separated image area. 2. The image processing apparatus for each image area of a copying machine according to claim 1, further comprising a common halftone processing means, and a halftone processing parameter for each image area is input and set from a processing parameter input means. 8. The automatic image area recognition means outputs different signals for black characters and color characters in the character area of the original image, and outputs different signals for continuous tone images and halftone images in the gray area of the original image, and performs image processing. The means performs different image processing on black characters, color characters, continuous tone images, and halftone images, respectively, and at least one of the respective processing parameters is input and set from the processing parameter input means. 2. An image processing apparatus according to image area for a copying machine according to claim 1.