JPH03185970A - Color document picture processing unit - Google Patents

Abstract

PURPOSE:To obtain a color picture processing unit with a high coding efficiency to a document in which monochroic color is eminent and color is not important by applying efficient storage to various kinds of document picture, especially a picture in which color and monochroic color exist in mixture. CONSTITUTION:A color scanner 100 reads a document color picture, extracts the picture as an RGB multi-value digital data and only a luminance data of a picture is outputted to a data input section 200 as brightness information. The input section 200 binarizes the data and stores the result tentatively in frame memories 310-340. The stored data is converted into a mode suitable for the input of a display section 500 and displayed by the recording form mode designated by an input mode designation section 150 at a data conversion section 400. Thus, the binary data in the memories 310-340 is coded and stored by a data storage section 600. Thus, the color and monochroic picture are stored efficiently and high coding efficiency picture processing is attained.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention is applicable to functions for inputting, storing, and searching (10) documents, especially those containing color information such as color photographs, as image data, displays, printers, etc. This involves a system that has the ability to output data, efficiently encodes images that are a mixture of monochrome and color images, and encodes them in a format that is compatible with devices that output only monochrome binary data, such as conventional FAX machines. The invention relates to an image processing system having functions.

[Conventional technology]

Conventionally, in devices that handle color images as digital data, images are separated into three primary colors of red, green, and blue, resulting in three types of multivalued data (hereinafter, each component will be referred to as R,G.

B, and the data of each color component is R data, G data,
B data).

For example, in color copying machines, R, G, etc.
, B multivalued image data is output. Also, when binarization processing is required, such as when outputting to a printer, multi-value RGB
The data is independently binarized and output for each color. A known example of a highly functional version of this system is, for example, Japanese Patent Application Laid-Open No. 64-51583. This known example shows a method in which a color image is separated into four colors: yellow (11), low, magenta, cyan, and black, and each color is binarized and stored. However, when image information is separated into four colors, yellow, magenta, cyan, and black, if only the black information is output, the colored portion of the image will be missing. Further, the case where intermediate colors such as color photographs are displayed is not considered.

On the other hand, in T'V or VTR, etc., the widely used method is to convert RGB color image data into luminance information and two types of color difference information, then perform orthogonal transformation, and encode and store the coefficients. It is used. This method handles only luminance information in monochrome areas that do not have color difference information.

Therefore, encoding efficiency is high for images containing many monochrome areas. Therefore, an example is shown in which this method is used for documents as well. (For example, JP-A-63-928
(2, etc.) [Problem to be solved by the invention] Since the color tone and saturation (12) of color images are greatly affected by the characteristics of the output device, it is necessary to correct these when outputting data that has been accumulated - There is a need to.

However, in order to perform the correction process, it is necessary to treat the image as multivalued data. Therefore, the method described in the known example of binarizing and storing images cannot be applied to images that require reproduction of color tone.

Furthermore, commonly used color documents often include large monochrome areas even though they include a color image as part of the image. In particular, in monochrome documents with red stamps, color information is meaningful only in the presence of color, and accurate reproduction of color is not so important.

It is inefficient to accumulate all R, G, and B data for these images as well. Further, even in a document containing a mixture of color photographs, a single area ratio often includes many monochrome character areas. Therefore, systems that target documents, even systems that target color images, are required to have high encoding efficiency for monochrome images and characters. However, (13) the method of encoding each of R, G, and B independently has high redundancy and low encoding efficiency, so it is not suitable for electronic file systems that accumulate large amounts of document image data or for low data transfer speeds. It is not suitable for faxes that require sending document images. In particular, a method that handles RGB multivalued data as is is not suitable for the above-mentioned apparatus because the amount of data is enormous. (Note that from now on, the part of the image that is expressed in monochrome will be referred to as the monochrome area.) On the other hand, the data encoding method that uses luminance/color difference conversion was originally devised for devices that handle natural images such as TVs. be. Therefore, although it is suitable for images with gradual density changes, when handling document images that include many parts such as characters with extremely large density changes, the following problems occur.

(a) Line figures such as characters have sharp changes in density, so encoding efficiency is significantly reduced.

(b) When the compression efficiency of encoding is increased, it becomes irreversible encoding. Therefore, when data transfer is performed using rask image data as an intermediary, such as by FAX (14), the data changes sequentially each time the data is converted, that is, each time it is transferred.

(c) Documents cannot be directly output to devices that handle them as binary images, such as faxes or electronic file systems.

(d) Data compatibility with conventional monochrome systems is impaired.

Therefore, there has been no means for efficiently storing color photographs, which require gradation for display, and color documents, which contain a mixture of characters that require high resolution, using a single encoding method.

A first object of the present invention is to provide a color image processing device that efficiently encodes and stores various document images, regardless of whether they are color documents or monochrome documents. Among these, it is possible to efficiently store images with a mixture of monochrome parts, which are common in color documents, and furthermore, it is possible to efficiently store images with a mixture of monochrome parts, which are common in color documents, and furthermore, it is possible to efficiently store images with a mixture of monochrome parts, such as documents with red stamps, and color documents where color reproduction is not important. An object of the present invention is to provide a color image processing device with higher encoding efficiency.

Another second object is to provide a color image processing device that is compatible with the conventional image processing device that deals with monochrome (15) mouth images.

At the same time, a third objective is to provide a color image processing method for storing color image data in a format that maintains compatibility with conventionally stored monochrome image data.

Also, even when outputting a color image with -degrees accumulated,
A fourth object is to provide a color image processing device that allows the color tone of an image to be changed again.

Furthermore, another fifth object of the present invention is to provide an image processing device that can display the contents of each image at high speed, for example, when searching the contents of a large amount of color image data stored on an optical disk or the like. There is a particular thing.

Furthermore, most of the FAX and electronic file systems that have been widely used so far are equipped with input/output devices for binary monochrome images. Therefore, if color document imaging systems are made compatible with these systems, they will be able to be used more widely (16). To this end, it is necessary to maintain compatibility in data recording formats, but conventionally, this compatibility has not been taken into account.

[Means to solve the problem]

In order to solve each of the above-mentioned problems, the present invention includes means for handling color image data as a plurality of independent binary image data.

Specifically, when recording one color image, three binary images each recording an R component, a G component, and a B component of the image are each recorded as R data.

It has a means for storing G data and B data, and synthesizing the three binary images at the time of output to express one color image.

Further, it has means for accumulating only the luminance data of the image as a single image and expressing a monochrome image. For example, G data may be substituted for the brightness data. In addition,
Hereinafter, in this specification, data for the three primary colors of a color image will be referred to as R data, G data, and B data, respectively, and three pieces of binary image data used to express one image will be referred to as (17 ) When referred to individually, each is called a plane. Further, the luminance data is indicated as Y.

First, in order to express only a specific color such as red or blue in an image, such as a document with a red stamp, the present invention uses binary data Y indicating brightness information and binary data Y indicating only the part of the color concerned. It has means for storing data as independent images. Specifically, there is a means for accumulating a plane of luminance data for recording a monochrome binary image, and a means for accumulating binary image data showing only pixels to be displayed in red, for example. In other words, image data is stored as two binary images. Therefore, a means is provided for extracting only pixels of the relevant color from the image.

On the other hand, in order to express full-color images such as photographs,
It has means for recording data of each plane of R, G, and B, and displaying each plane in R, G, and B, respectively.

Therefore, in order to achieve the first object, the present invention has means for storing input in each of the above formats, and further has means for switching (18) each processing means depending on the target image.

The second object is to provide a means for binarizing each multivalued data as an independent image as described above, and a means for storing the binary data in the same format as a conventional monochrome binary image. This can be achieved by having In other words, the data format of each plane is the same as that of the conventional monochrome binary image system.

Further, the third object can be achieved by having means for displaying only a luminance plane in monochrome, and displaying image data accumulated in a conventional monochrome system as a luminance plane.

On the other hand, a fourth object is to provide means for reproducing multivalued image data from binary image data of each plane when outputting an image;
This can be achieved by having means for performing image processing on the reproduced multivalued image data.

The fifth object can be achieved by having means for displaying only the above-mentioned luminance plane in monochrome when displaying an image.

In order to handle the various types of images described above (19), the present invention has means for executing multiple processing modes as a data storage and binary processing method, and furthermore, It has means for switching between them accordingly, and means for recording an identifier for identifying the mode in question along with the image data to be stored.

Further, the stored data reading section has means for reading out the recorded identifier, and the image display section has means for switching the color displayed by each plane according to the identifier.

[Effect]

First, data storage and binary processing modes executed by the present invention will be described in order to handle the various images described above.

The present invention executes a predetermined data storage and binary processing method for each mode, and the image storage characters include:
Also record the identifier for this mode.

This mode identifier is referred to in image display and used to specify data necessary for display and to determine an appropriate compositing method.

The contents of each mode are as follows.

(20) Mode (1) Monochrome mode: Only luminance data Y of the image data is accumulated.

Mode (II) Multicolor mode Brightness data Y and pixel data R to be displayed in red are extracted from the color image data created by two people, and each image data is stored in separate planes. For example, there are two binary images: a binary image obtained by binarizing luminance data, and a binary image that describes only the pixels that are recorded as black in that binary image and which you want to express as red, for example. Accumulate images.

During display, the R plane image is expressed in red, and the other pixels are displayed in black and white according to the Y plane.

Mode (Ill) Full color mode: R, G, and B color image data are stored in R, G, and B planes, respectively. When displaying, each plane is R, G, B.
Compose and display.

Mode (1v) Mixed mode: Identify color areas and monochrome areas that require color information from the image, and store luminance information in the Yti-G plane for monochrome areas,
The R and B planes are blank. -(21) On the other hand, in the color area, R, G, and B binary data are accumulated in each plane. Furthermore, area identifiers Fcm indicating color areas and monochrome areas are stored as a fourth plane.

Here, mode (1) is used when inputting a monochrome document. As mentioned earlier, mode (n) is a mode that targets documents in which seals or corrections are made in red or blue on monochrome documents, which are often found in office documents. However, this mode can be applied even when there are two or more colors to be expressed. Moreover, mode (III) is a mode used when a color photograph or the like is targeted. Furthermore, mode (IV) is a mode for documents containing a mixture of monochrome documents and color photographs.

Next, the operation of each part in the above problem solving means will be explained.

First, when dealing with monochrome images, the encoding efficiency decreases only when color information is required in one image, that is, only in mode (m), color information is stored in all R2O and B. However, in the case of monochrome portions in the image (22) or documents designated as monochrome images from the beginning, this problem can be solved by storing only the brightness information.

Note that in this specification, mode (1) and (
In the case of II), the luminance data Y is the pixel to be displayed in black.
l jl, and the blank part is recorded as '0'''.

In the case of mode (1), the data stored is only luminance information, so the above-mentioned problem does not occur.

On the other hand, in cases where accurate color reproduction is not required, such as in a document with a red stamp, mode (II) processing allows the amount of data in the monochrome portion to be the same as mode (1).

Important information is that red is the only color in documents with red seals, red-edited documents, etc.
Accurate color reproduction as in color photography is generally not required. Therefore, these images are handled in the form of monochrome image data, plus image data that specifies colored parts.

Specifically, first, the luminance data Y is divided into a single plane (23
). Here, in the luminance plane, not only black pixels are described, but also pixels to be displayed in red or other colors are also described. Then, out of the pixels that would be expressed as black when displayed in monochrome, that is, the pixels with Y="1", pixels that should be displayed as red, such as red characters, are extracted, and R
Record it as 1 on the plane. Here, in the R plane, binary image data is recorded in which only the pixels written in "red" are expressed as 1" and the others are expressed as O". Then, in the area described in black and white, the R plane becomes all "0'", so the encoding efficiency is high, and the color information is processed with almost the same amount of data as in the case of only monochrome binary image data. You can also record.

At the time of output, the content of the R plane of the luminance data Y is 1.
Displays 1011 pixels in black, and the contents of the R plane are
I II pixels are displayed in red. As mentioned above, the luminance plane includes pixels that should be displayed in red, so if only the luminance plane is displayed in monochrome, the color document will be displayed in the same way as if it were input using a conventional monochrome system. (24) Can do.

On the other hand, in mode (IV), which targets documents containing a mixture of full-color images such as photographs and monochrome images, color/
By accumulating the monochrome identification information Fcm, redundancy of data in the monochrome portion can be suppressed. This mode is particularly effective when a large amount of documents are input continuously using an automatic paper feeder or the like. On some pages of the document to be input,
When a color photograph is included, in mode (nI), all pages of the document are recorded in three pieces of RGB data, so the amount of data to be stored will be three times that of monochrome storage. On the other hand, in mode (mV), only the luminance data Y is accumulated on most pages, and the R and B planes and attribute identifier Fcm are blank on most pages, so efficient encoding is not possible. realizable.

Hereinafter, in this specification, the attribute identifier F(1) outputs "0" when the target area is a monochrome image, and "1" when it is determined to be a color area.

I think so.

In this case, in the color area, R, G, and B data are recorded as data on three sheets (25), and in the monochrome area, the luminance and red and blue data described above are recorded, and when outputting an image, the data is It is necessary to have a means for switching the display method.

In this method, image data is binarized and stored, so
Similar to conventional devices for monochrome binary images, M H (M
Huffman), M R (Mo
It can be encoded using a method such as dividedRead). Therefore, recording media such as optical discs can maintain compatibility with conventional monochrome systems. Therefore, as described above, the luminance data Y can be output by a conventional monochrome system.

Furthermore, these binary image encoding methods do not cause the problem that the encoding efficiency suddenly decreases in line graphic parts such as characters.

Furthermore, in these binary image data encoding methods, O
Alternatively, the amount of data in the portion where 1s are continuous can be minimized. Therefore, the encoding efficiency of information such as color/monochrome identification information whose value changes in units of regions is extremely high. Furthermore, since the R plane stored in mode (II) (26) is blank except for the color portion, high encoding efficiency can be obtained. In this example, the case is described in which only red is displayed as color information added to black and white, but the case of displaying two or more colors can also be performed by the same means by increasing the number of planes to be accumulated.

Next, the case of storing full-color images will be described.

When a full-color image is represented by binary information, it is necessary to perform pseudo-halftone processing in order to reproduce not only density but also color. However, when pseudo halftone processing is applied to line figures such as characters, the resolution of the image will be reduced. Therefore, for parts that do not require color reproduction, such as monochrome parts of mixed documents or multicolor documents, line figures are recorded more precisely by adaptive binarization processing. On the other hand, for color areas, pseudo halftone processing is applied even to line figures in order to reproduce colors.

Furthermore, when an image is displayed by superimposing three pieces of binary image data of R, G, and B, a color that does not originally exist may appear due to the difference in values between the three pieces of data (27). This will be referred to as color noise hereinafter.

In the present invention, color noise is divided into modes (In) and (IV).
Occurs in the case of To prevent this color noise, 3
The data in the planes must have equal values.

Therefore, in the present invention, in the case of mode (III), the color/monochrome determination section determines that it is a monochrome area,
Furthermore, R and G are applied to pixels determined to be character areas by the character/photo area determination unit.

This problem is solved by having a means for recording the same value in the three planes of B.

In order to realize each of the above-mentioned operations, a means is required for performing a data binarization method and selection of data to be stored based on predetermined conditions using the mode and the output of each area identification device.

On the other hand, when outputting a color image, image processing such as affine transformation, changing specific colors, or adjusting color tone, etc.
It is necessary that the image data be multivalued at the time of processing. Therefore, from images stored as binary data,
These functions can be realized by reproducing multivalued data (28).

Furthermore, when displaying the contents of a color image consisting of 2.3 planes or 4 planes of data at high speed, for example in the case of a search, for example, among the image data stored in optical discs and disks, the brightness By reading out only the plane and displaying it in monochrome, it is possible to display a monochrome image at high speed simply by reading out the data for two planes.

〔Example〕

An embodiment of the present invention will be described in detail with reference to the drawings.

First, the data recording format of a multicolor image, which is one of the features of the present invention, will be explained using FIG. This corresponds to the case of mode (II) described above.

In FIG. 1, (a) is a document in which characters are written in two colors, black and red. In the figure, the letters in 11 are written in black, and the letters in 12 are written in red. This document is in the mode (II) described above.
For example, when inputting with
The pieces of binary image data become images shown in (b) and (Q) in FIG. (b) (29) is a plane on which luminance data Y is recorded. The image is recorded as binary data, and the characters written in black and red in the original image are both written as a monochrome binary image. The mode in which only this luminance data Y is recorded corresponds to the above-mentioned mode (T). In this specification, as described above, in the case of mode (1), the pixel where a character is written is designated as 111 II, and the pixel in the blank area is designated as "OIT".

On the other hand, (Q) shows two types of image data written in the R plane. 111 only for pixels that are determined to be red among the pixels that are described as “1” in the brightness data.
” is written. The configuration of the part that extracts this red color and binarizes it will be described later.

As a result, the luminance data Y (x e y
) and R (x + y) shown in (c), the following relationship holds true.

if R(x, y)=1 then Y(x,
y)=1 Therefore, a case where the R plane is "1" and the Y plane is "0" does not occur.

Figure 2 shows each color (30
) is an example of a data format that expresses In the figure, Y indicates the luminance data, and R indicates the content of the R data.

When the expressed colors are white, black, and red, the possible values of the data are:
There are three ways shown in the figure.

On the other hand, three binary images (b
), (Q) When displaying a multicolor image, there are two types of output: color output and monochrome output. FIG. 2 also shows the colors at the time of output. When displaying the image of the engineering drawing in color, the image in (c) is displayed in red, and the parts other than red are displayed in black and white according to (b). Specific means for realizing this display method will be described later.

Furthermore, when displaying a plurality of accumulated images in succession, high-speed display is possible by reading out only the Y data and displaying it in monochrome.

By using only this Y data, data compatibility can be maintained with devices that handle monochrome binary images, such as FAX. This feature is.

The same applies to other modes.

In this example, red is the color to be reproduced when displaying an image (31), but the same applies to other colors. Further, a plurality of pieces can be displayed as specific colors. In this case, for example, if you need n representations,
Binary image data of n+1 planes is accumulated.

As an example, FIG. 3 shows an example of expressing two colors, red and blue, in addition to black and white. In the figure, (a) is a document in which characters are written in three colors: black, red, and blue. The letters in 16 in the diagram are black, 17
The letters inside are written in red, and the letters inside 18 are written in blue. If this document is entered in the mode (n) mentioned above,
The images are stored as three binary image data shown in (b), (Q), and (d) in FIG. (b) is a plane in which luminance data Y is recorded, (Q) is a plane in which pixels to be displayed in red, and (d) is binary image data in which pixels to be displayed in blue are recorded.

In this case, the data format representing each color in the document is
As shown in the figure.

Next, a mode (In) for color photographs and the like will be explained. FIG. 5 shows the data recording format in mode (III). The data is recorded as R9 (32) G, 83 pieces of binary data. In the figure (a)
20 shows an image containing a mixture of color photographs, (b) shows the content of binary image data of the R plane, (Q) the G plane, and (d) the B plane. In the figure, the letters 21 are written in black, and the letters 22 are written in a certain neutral color. on the other hand,
23 indicates the area where the color photograph is written.

Here, in this mode, the colors recorded according to the data contents of each plane are shown in FIG.

In mode (III), white and black are also treated as one color.

Regarding binarization, a full-color image such as a photograph shown as area 21 in FIG. 5 is converted into R and G.

When expressing with B binary data, each of the R, G, and B data is binarized by pseudo halftone processing. Therefore, 33, 43, and 53 in FIG. 5 are data obtained by independently converting R, G, and B data into binarized data through pseudo halftone processing.

On the other hand, in this mode, for example, black and white character areas such as 21 are not subjected to pseudo halftone processing (33) during the binarization process. This is performed by switching between two types of binarization methods using means for separating character areas and photographic areas from the document image using a known method. Furthermore, in order to correctly represent the area where black letters are written, R, G.

B Make the values equal between each plane. Specifically, pixels that are in a character area and are determined to be a monochrome area are R,
This can be achieved by writing the same value on the three planes G and B. Therefore, 31°41.51 describes the same value.

However, even in the character area, if it is necessary to reproduce the color of a colored character such as 22, pseudo halftone processing is required to express the color. Therefore, this function is realized by having means for identifying color areas and monochrome areas in addition to identifying text areas and photo areas. In this case, 32, 42.52 are written as independent values.

Note that the conditions for selecting the binarization processing power formula based on various judgment results and designated modes will be described in detail later (34).

Next, a mode (IV) that targets images in which color areas and monochrome areas coexist will be described.

This mode is particularly effective when storing documents, etc. consisting of many pages as image data. Generally, when inputting a document or the like consisting of a large number of images, an automatic input device called a sorter is used, for example.

In this case, the efficiency of input work can be improved by eliminating the need for an operator to attend. However, if no operator is present during input, it is difficult to switch the mode for each image. Then, if some pages of the document to be input include color areas, it is necessary to input all images as color images. Here, all images mode (III)
When stored in monochrome mode, a monochrome image requires three times the amount of data compared to when input in the original monochrome mode.

On the other hand, mode (IV) has an identifier indicating whether each part in the image is a color area or a monochrome area as a fourth plane. This identifier can be realized using a binary flag, so it can be recorded in the same way as each plane described above.

FIG. 7 shows an example of the recording format in mode (IV). In the figure, (,) is an example of the same color image as in FIG. 5(a). (b), (c), and (d) are binary image data of R, G, and B, respectively, and Fcm is a binary identifier for identifying a color area and a monochrome area for each part of the image. Here, suppose that the identifier Fcm(x, y) is the pixel I
“1” if M(x, y) is color, and “1” if M(x, y) is monochrome (
l OIT. Then, in the binary data Fcl11, corresponding to the color area 22.23 in the image (,),
92.93 is 111" and the other parts are "O".

Here, in the portion where Fcm=1, R, G, and B binary image data are written in the R, G, and B planes, respectively.

In this area black is written as R=G=B=0. On the other hand, for example, in the black text portion 21, the recording format is the same as monochrome input, and the pixels described as black are R, G.

Write “1” only in one plane of B. This example (36
) Then, let the G plane be the relevant plane.

In the case of image display, with this recording method, only the G plane is read out as image data in the part where Fcm = 0, and when G = 1
If so, black is output, and if G=O, white is output. On the other hand, Fcm=
For pixel 1, R, G.

Read out the three planes of B and display them in the same way as in mode ([1). Therefore, the image output unit compatible with mode (IV) needs to have means for switching the display color of the G plane according to the value of the identifier Few.

When image data is recorded using this method, FCm=O2R=O9G=O is always maintained for monochrome images, and high coding efficiency can be obtained. It is possible to record documents containing a mixture of documents. On the other hand, even in areas where color and monochrome are mixed, the identifier Fcn1 switches between O and 1 for each area, so encoding efficiency is high, and mode (■)
Image data can be stored with the same amount of data as Habo.

Next, a method for realizing image storage using each of the above-mentioned recording formats (37) will be described. First, FIG. 8 shows an example of the overall configuration of the present invention.

In the figure, 100 is a color scanner that optically reads images such as documents by known means and outputs them as multivalued digital data such as RGB, and 150 is one of the above-mentioned recording format modes (N to (IV)). 200 is an image input unit that binarizes input multi-value data of each color of RGB for data storage; 310;
320.330.340 is a frame memory 400 for temporarily storing output results; 500 is an RGB data conversion unit that converts binary image data into a format suitable for input to a display device such as a high-definition color CRT; an image display unit such as a high-definition CRT that inputs and displays multivalued data; 600;
9 is a data storage unit that encodes binary data in the frame memory and records it on a large-capacity data storage unit such as an optical disk, or reads out and decodes already stored data;
00 is a control unit that controls the entire system.

Here, one of the characteristic parts of the present invention is the data (38) input section 200. Therefore, next is the input part 20 of the image.
0 will be explained. Among the parts shown in FIG. 8, the parts related to image input will be explained using FIG. 9.

In the figure, 100 is a color scanner that optically reads images such as documents as described above, and 150 is an input mode specifying section that specifies a mode at the time of input. Also, 310,320,
330 and 340 are frame memories that temporarily store binary data of each plane.

The feature of the data input unit 200 is that it selects not only the binarization force formula but also the data to be written on each plane based on the results of a plurality of types of region determination.

As a means for determining the area of the input image, 210 is a color/color/monochrome area that identifies whether each pixel of the image belongs to a color area or a monochrome area from three RGB multi-valued image data using means to be described later. Monochrome identification section, 22
0 is a color identification unit that determines whether each pixel of an image belongs to a predetermined specific color such as red or blue, and 230 is a color identification unit that inputs multivalued digital image data by known (39) means; This is a text/photo area determination unit that determines whether each pixel of an image belongs to a line graphic area such as characters or a halftone area such as photographs. In addition, RGB multivalued image data can be
As a means for converting into a value, 240 is a binarization processing unit that corresponds to an image such as a line figure by a known means, and 250 is a dither processing unit that performs binarization processing by pseudo halftone processing. On the other hand, 260 is RGB based on the input mode and each judgment result.
This is a selector that selects and outputs data to be accumulated from two types of binarization processing results for three images.

Here, each frame memory 310 .
The contents of data recorded at 320, 330, and 340 are shown in FIG. In this specification, data developed in the frame memory 310 will be referred to as DATA-1, 320.
, 330, and 340, respectively.
They are called TA-2゜DATA-3 and DATA-4.

In FIG. 10, "-" indicates that data is not recorded.The image data accumulated in mode (I) is recorded in the frame memory 810 only in (40) Y.Mode (
II) stores data for 2 planes, and mode (m) stores data for 3 planes. Furthermore, in mode (IV), four planes are stored at 310, 320, 330, and 340.

Now, in order to briefly explain the features of the present invention, it is assumed that the color scanner 100 simultaneously outputs RGB multivalued data of each pixel on an image. The signal line 110 is R data, 1
20 sends G data, and 130 sends B data. Further, in this example, to simplify the explanation, G data is substituted for the brightness data, but it is also possible to calculate the brightness data by arithmetic processing on RG83 types of multivalued data.

The color/monochrome identification unit 210 determines the mode (m
) and (IV). A binary color/monochrome identification section Fcl11 is output.

On the other hand, the color identification unit 220 aims to extract pixels described in a specific color plane in mode (II). In this embodiment, a case where red pixels are extracted (41) will be explained as an example, but the number of colors to be extracted can be any number greater than or equal to 1, and the white body can also be arbitrarily set. The format of the identifier Frb output from the color identification unit 220 varies depending on the number of extracted colors, and when extracting the red color, the format is 1 bi
It becomes a signal of t. The following relationship exists between the number of colors Cnb extracted here and the number N of bits of the identifier Frb.

N≧loga(Cnb+1) There are already many known methods for extracting specific colors that use bias in values between RGB.

On the other hand, the text/photo area determination unit 230 aims to identify, as a binarization formula, an area where line graphic processing should be performed and an area where pseudo halftone processing should be performed. Specific means are described in, for example, Japanese Patent Laid-Open No. 63-316566. The determination result Fdm becomes binary data.

If the determination result Fdm differs for each color, the output of the selector 700 will become unstable, so text/photo area determination is performed on luminance data. In this example, as described above, the G data is used as a substitute for the luminance data.

(42) FIG. 11 shows an example of the mode designation code F mod input from the input mode designation section 100.

The selector 260 is driven by the above four types of code information. The selector 260 selects each of the three RGB signals output from the binarization processing section 240 and the dither processing section 250, respectively.
Plain binary image data.

From the data of a total of 6 planes, 4 frame memories 31
2 to output on 1 to 4 sheets among 0,320,330,340
Select value data DATA-1 to DATA-4.

An example of the operation of the selector 260 is shown in FIG.

The data outputted in the figure are Rc, Gc, and Be, which are R, G, and Rc output from the binarization processing section 240 in FIG. 9, respectively.
Binarization processing results of B, Rd, Gd.

Bd is R and G output from the dither processing section 250.

This is the result of binarization processing of B. In mode (1),
Only the luminance information is output, and only one of Gc and Gd is output to the frame memory 310. Here, the switching between Gc and Gd is executed by the output Fdm of the character/photograph (43) true area separation section 230.

However, if the image to be input consists of only one of text or photograph, it is also possible to fix it to either one from the outside using the input setting section 100, but this will be described later.

In mode (It), binary image data for two plays is stored in the frame memories 310 and 320. The selection of data to be stored is controlled by the output Fdm of the text/photo area determination section 230 and the output Frb of the color identification section 220. Here, as in mode (I), either Gc or Gd is selected and recorded in DATA-1 using Fdm.

Furthermore, this figure shows a case where 1-bit code data is used as Frb.

If the target pixel is “red”, the identifier Frb is “
1”, otherwise ROI+.DATA=2
becomes Gc or Gd when Frb is "111", and when Frb is "O", DATA-2 becomes "O".

(0) In addition, switching between Gc and Gd is performed using F as in mode (1).
Selected by dm.

In mode (III), images are also stored as binary images of R, G, and B three planes. Here, the data to be accumulated is each plane of R, G, and B.

On the other hand, in mode (IV), R, G, B and Fc
4 frame memories of 310° and 320°, respectively.
．． 330 and 34.0, respectively. Here, one of the G plane and the luminance plane is selected and stored in the frame memory 320 based on the output Fcm of the color/monochrome discrimination section 210, but as described above, in this example,
Since the G data and the luminance data are the same, there is normally no need to select them. However, in a document containing a mixture of character areas, the monochrome line figure area is
For other areas, the dither processing unit 250 selects the output. Therefore, either Gc or Gd is selected based on the output Fcm of the color/monochrome discrimination section 210.

Now, the example described so far assumes that the target document is a document containing a mixture of text and photos. However, when inputting a document that consists entirely of text or photos, For example, in addition to (1) to (TV), [text], [photo]
By specifying the area type such as , [Mixed],
The operating conditions of the selector 260 can be determined independently of the text/photo area determining section 400. In this case, the signals from the input mode specifying section 150 to the selector 260 include selection codes for four types of input modes and two identification codes according to the target document.
Bits are added.

FIG. 12 shows an example of the output code from the input mode specifying section 150 in this case. Basically, each of the four types of modes includes [Text], in which dithering is not selected, and [Photo], which is always selected, and [Mixed], which is based on the output of two character/photo area determinations. However, in the case of [character], the output of the normal binarization processing unit 240 is accumulated. However, mode (m
) and mode (IV), it is necessary to reproduce intermediate colors, so the binarization processing unit 240 (4)
6) No output is accumulated. Therefore, mode (I[
For [) and mode (mV), cases where [character] is specified are excluded.

Here, an example of the operation of the selector 260 when [Text] is specified is shown in FIG. 14, and an example when [Photo] is specified is shown in FIG. 15, respectively.

Next, an example of a method for storing data in each of the above-mentioned recording formats will be described.

DATA-1, DATA-2, DATA-3, stored in frame memories 310, 320, 330, 340
The binary data of DATA-4 is encoded by known means and then stored in a mass data storage device such as an optical disk. An embodiment of the data storage section will be described using FIG. 16.

In the figure, 310, 320, 330, 340 are as mentioned above,
A frame memory 600 that stores binary image data for each plane corresponds to the data storage section in FIG. The data storage section 600 can be realized with the configuration described below. 6 in the diagram
10 sequentially reads data in each connected memory,
This is a selector that outputs to the code/decoding processing unit (47) 620. Code/decoding processing unit 62
0 encodes binary data using known means and stores it in a data storage unit 650 such as an optical disk.

When the control unit 900 gives an instruction to accumulate data, the system first transfers and accumulates header information on the optical disc 890. The header information includes, for example, information used when searching, such as the name of the document, and the size of the document. However, in the present invention, the mode at the time of input is recorded as a code or the like as an item of header information.

After storing the header information, the binary image data of 1 to 4 planes is sequentially encoded by the encoding processing unit 620 depending on the mode, and is stored in the data storage unit 650. The order of data to be stored is selected by selector 610 based on instructions from control unit 900.

Here, the recording order of data stored in the data storage section 650 will be explained using FIG. 17.

Figures 17(a) to (d) are modes (I) to (IV).
(48) This is an example of a recording format for input data.

In the figure, 910 is an area for recording header information, and 920 is a DA
TA-1, 930, 940, 950 are each DATA
-2, DATA-3, and DATA-4 are recorded in this area.

In mode (1), the image data is only the luminance plane, and only DATA-1 in the frame memory 310 is stored. On the other hand, in modes (II) to (m), a luminance plane is accumulated at the beginning as image data,
Then record a plane for each color. In mode (TI), when expressing a plurality of colors, each plane is recorded in order after this luminance plane. As a result, when searching and displaying two images, just read the data for the first image.
It is also possible to output a monochrome binary image. On the other hand, in mode (IV), G data is recorded at the beginning of the image data. This is because G data is located in the middle of the spectrum among the three types of data, R, G, and B, and therefore the phenomenon in which a specific color is not reproduced is least likely to occur. Therefore, in this case as well, when displaying (49), it is possible to output a monochrome binary image by simply reading out the data of the first page.

On the other hand, code information representing the mode at the time of inputting each image is recorded in a recording area 911 in the header information.

Next, a process of reading image data recorded in the data storage section 500 and displaying the image will be described.

First, the image data stored in the data storage section 650 is read out, and the frame memories 310, 320, 330.
The input section 40 will be explained using FIG. 16. First, the header information of each data recorded in the data storage section is read out, and the target data is searched.

Once the target data is confirmed, the mode identifier F mod recorded in the header information is used to
The number of planes to be read out is determined, the applicable data in the data storage section 650 is read out for the required number of planes, each is decoded into binary data by the code decoding processing section 620, and then the frame memories 310, 320, 330 ,340 to DAT
Accumulate as DATA-4 from A-1 (50).

Next, frame memories 310, 320, 330° 340
The data converter 400 converts DATA-1 to DATA-4 expanded into data into data in a format suitable for input to an image display unit 500 such as a high-definition color CRT, and displays the data on the image display unit 500.

The operation of the data conversion unit 400 is to first generate 3-plane image data of RGB from input binary image data of 1 to 4 planes. Hereinafter, in this specification, R, G, and B data reproduced from multilevel information will be referred to as DATA-R, DATA-G, and DATA-B.
It is called. Here, DATA-R, DATA-G, DAT
The relationship between A-B and the colors displayed on the image display section 500 is the sixth
It is similar to the figure. DATA-R, DATA-G.

If both DATA-B are "1", "white" is displayed, and in the case of di Ojl, "black" is displayed.

Next, an example of the configuration of the data conversion section will be explained using FIG. 18. The data selector 410 in the figure is controlled by the control unit 900 according to the mode identifier F mod in the read (51) loaded header information, and the data selector 410 reads (51)
Combine necessary data from A-2, DATA-3, and DATA-4, and create DATA-R, DATA-G, and DATA-B.
Output. The details of this operation will be described later. The RGB three-plane binary data output from the data selector 410 is converted into data in a format suitable for the image display section 500 in a multi-value processing section 4.20. In this process, for example, a bit shift or the like is used to convert the value that controls the blinking of each pixel of the image display section using binary data.

The operation of the data converter differs depending on the mode of the target image, the type of display, and further the content of the data.

FIG. 19 shows an example of the operation of the data selector 410.

Here, for example, when displaying an image in mode (If) in color, the logical formula of DATA-1 and DATA-2 is as follows:
DATA-R, DATA-G.

DATA-B is determined. When only the brightness information DATA-1 is "1", DATA-R, DATA(52)-G, and DATA-B all become "O" and "black" is displayed, and DATA-1 and DATA-2 are If “1”,
By setting only DATA-R to "1", "red" is displayed.

On the other hand, the timing for displaying images can be determined as follows. For example, when displaying a monochrome image, luminance data Y is input to the frame memory 310, and at the same time, the luminance data is output from the data selector as three plane data of RGB. Then, the multivalue processing unit 420
A monochrome image is displayed on the image display section 500 by sending image data having equal values for each of the R, G, and B planes to the image display section 500 .

On the other hand, when displaying a full-color image in mode (m), the values of DATA-R, DATA-G, and DATA-B are
DATA-1゜DATA-2, DATA each independently
-3.

Also, the timing of image display is DATA-1゜DAT.
At the same time as A-2 and DATA-3 are written into the frame memory, G, R, and B pre (53) lines are displayed in sequence.

When displaying mode (II) image data, the final color cannot be determined until three planes of data are stored in the frame memories 310 to 330.

In this case, it is also effective to display the luminance data as a monochrome image in the same way as the initial mode (I) and to rewrite it sequentially.

On the other hand, by controlling the operation of the data selector 410 based on instructions from the control unit 900, it is possible to display a color image in monochrome, for example. Specifically, for example, in the case of monochrome display, only DATA-1 is read out, and R and G are read out.

This can be realized by displaying this DATA-1 in both B and B.

For example, among the two plane image data of a multicolor document input by mode (IF), DATA-2
By displaying only the red plane, it is also possible to select and display only the red plane.

Now, a full-color image targeted by mode (m) or (IV) originally needs to express light and shade using intermediate tones (54). Particularly in color photographs, subtle differences in color tone due to the characteristics of the output device can be perceived as large differences by human vision.

Therefore, in order to display a full-color image with high quality, it is necessary to correct or change the brightness and chromaticity of the halftone data. In the present invention, images are stored as binary data, so in order to achieve this purpose, means for converting binary image data into multi-value data having shading is required.

FIG. 20 shows an example of the configuration of an image display section for realizing this purpose. An example of this is DATA-R.

Although a case is shown in which three types of binary data, DATA-G and DATA-B, are simultaneously input and output to the image display section 500, it is also possible to convert one plane at a time by using a memory that temporarily stores the data. can.

In the figure, 431, 432, 433 are halftone conversion units that reproduce multi-value data from binary image data; 441, 442
, 443 is a color tone conversion unit that converts the obtained halftone data, 451, 452, 453 (55) is a shift register that multiplies the binary data by a specific value, and 470 is a color tone conversion unit that converts the obtained halftone data. area determination units 461, 462, 4 for determining whether the
63 is a selector that selects one of two types of halftone data based on the output of the area determination section 470; 500 is an image display section that receives RGB multivalued data and displays a full-color image.

Here, for the sake of simplicity, the operation for one play will be explained, but the same applies to other planes.

Here, the halftone converter 431 reproduces multi-value halftone data from the pseudo halftone image. Many methods of reproduction are already known, for example, Japanese Patent Application No. 63-240973.
A method of extracting the distribution density of black pixels in a nearby local area can be applied. The image converted into halftones is subjected to necessary conversion for each RGB plane by a tone conversion section. The color tone conversion unit 441 is, for example, RAM (
Read Only Memory). In addition to setting the contents of RAM in advance,
For example, it is possible to implement a method in which the information is transferred from the control unit or selected from a plurality of preset types (56) according to the connected device.

On the other hand, in areas of line figures such as characters, the positional relationship of black pixels is important, and expressing shading actually degrades image quality by lowering resolution. Therefore, for example, if the input range of the display unit 500 connected to the system is 8 bits, deterioration of image quality can be prevented by simply shifting the input binary data by 8 bits.

The selector 461 selects and outputs one of the outputs of the color tone converter 441 and the shift register 421. The selection is performed by an area determining unit 47 that determines whether the area to be output belongs to, a line figure area or a pseudo halftone area.
It is switched by outputting 0. Many methods are already known as means for determining the area of the input binary image, and these are used in this embodiment.

The image selected by the selector 461 is transferred to an image display unit 500 such as a CRT and displayed.

As a result, halftone images such as color photographs can be stored as 2 (57) value data and then displayed as halftone images. It is also possible to perform conversions such as density and chromaticity.

In this embodiment, an example of a method of converting binary data of three planes at the same time has been described, but by providing a means for temporarily storing data, it is also possible to convert each plane in turn.

Although the case where an image is displayed on a CRT has been described as an example, the same method can be used for other output devices such as a printer.

〔Effect of the invention〕

According to the present invention, regardless of whether the document is a color document or a monochrome document,
It is possible to efficiently store various document images, especially images with a mixture of monochrome parts, which are often found in color documents, and it can also be used for color documents that mainly have monochrome areas, such as documents with red stamps, where color reproduction is not important. As a result, a color image processing device with higher encoding efficiency can be realized.

Further, it is possible to realize a color image processing device that stores color (58) images in a data format that is compatible with conventional image processing devices that handle monochrome images. At the same time, it is necessary to store color image data in a format that maintains compatibility with conventionally stored monochrome image data.
A color image processing method can be realized.

Also, even when outputting a color image with -degrees accumulated,
Color image processing devices that can change the tone and saturation of images are now available. Furthermore, for example, when searching the contents of large amounts of color image data stored on optical discs, etc.,
An image processing device that can display the contents of each image at high speed can be realized.

[Brief explanation of drawings]

Figure 1 is a diagram explaining the image data format when storing multicolor documents, Figure 2 is a diagram showing the colors expressed by each data when storing multicolor documents, and Figure 3 is a diagram showing the color represented by each data when storing multicolor documents. FIG. 4 is a diagram illustrating the image data format when storing a multi-color document containing red and blue, and FIG. 5 is a diagram showing the colors expressed by each data when storing a multi-color document containing red and blue. The figure shows the image (59) data format when storing a full-color document, Figure 6 shows the colors expressed by each data in full-color mode, and Figure 7 shows the color area and monochrome area. FIG. 8 is a diagram explaining the overall outline of the system for realizing the present invention, and FIG. 9 is a diagram explaining the data format when inputting a document containing a mixture of FIG. 10 is a diagram showing an example of the configuration of the part that converts into binary image data. FIG. 10 is a diagram showing the types of data written in the frame memory in each mode. FIG. 11 is a diagram showing the type of data written in the frame memory in each mode. FIG. 12 is a diagram showing an example of code data indicating each mode that is instructed. FIG. , A diagram showing an example of code data when inputting an instruction to limit the input mode and the target area using the input mode specifying section. FIG. Figure 15 shows the content of the data stored in each frame memory when the input document is limited to text or line figures (60). FIG. 16 is a diagram explaining the process of recording each image data in a large-capacity data storage unit such as an optical disk, and FIG. 18 is a diagram illustrating the process of displaying image data, FIG. 19 is a diagram illustrating the operation of the data selector in each state, and FIG. ) (C) (C) (d) No. 1 <C> (cl) 7 ■ Rain 14 (2) No. 5

Claims (1)

[Claims] 1. In an image processing device that handles document images including color images as digital data, means for accumulating luminance information of input multi-value color image data, and means for accumulating luminance information of input multi-value color image data in advance. A color document image processing device comprising: means for extracting pixels expressing a specified range of colors; and means for accumulating image information of the extracted pixels. 2. In the color document image processing apparatus and system according to the first claim, the luminance information storage means includes means for binarizing the luminance information of the multivalued color image data, and a means for binarizing the luminance information of the multivalued color image data;
A color document image processing device comprising: means for accumulating value image data; and means for accumulating only pixels of a specific color extracted from the binary image data as separate binary image data. 3. In the color document image processing apparatus according to claim 1, the means for extracting pixels of the specific color includes means for separately extracting two or more types of colors, and data indicating the extracted pixels for each color. means for storing the luminance information of the multi-valued color image data as binary image data; A color document image processing apparatus comprising means for displaying the color document image using image data. 4. In an image processing device that handles document images including color images as digital data, means for accumulating input color image data as a plurality of sheets of binary image data, and storing the color image as at least one of the binary image data. 1. A color document image processing device, comprising means for storing image data obtained by converting luminance information into a binary value. 5. In the color document image processing apparatus according to claim 4, the means for accumulating the plurality of sheets of binary image data stores the image data in an order in which data representing brightness information can be read out first when reading the data. A color document image processing device comprising means for accumulating. 6. The color document image processing apparatus according to claims 1 to 5, characterized in that the result of binarization processing of data of a specific color is used instead of the result of binarization processing of the luminance data. Document image processing device. 7. An image processing device that handles document images including color images as digital data, having means for reading out a plurality of sheets of binary image data, and means for displaying one of the binary image data as a black and white image. A color document image processing device characterized by: 8. In an image processing device that handles document images including color images as digital data, there is a means for reading out multiple sheets of binary image data, and pixels written in two or more of the binary image data are specified in advance. A color document image processing apparatus comprising means for displaying in a specific color and means for displaying pixels written only on one sheet in black and white. 9. In an image processing device that handles document images including color images as digital data, a means for converting a plurality of input multivalued color image data into a plurality of binary data, and a means for converting the multivalued color image data into a plurality of binary data; A color/monochrome identification means for determining whether each pixel of the image is an area written in black and white or an area containing color information; For a monochrome line figure area, the accumulated color image 1. A color document image processing apparatus, comprising means for making the binarization result of the color the same value for each color. 10. The color document image processing apparatus according to claim 9, wherein the binarization processing means includes a plurality of means for executing a plurality of binarization methods including pseudo halftone processing, and an output of the color/monochrome discrimination means. A color document image processing apparatus comprising means for selecting one of the plurality of binarization processing results based on the output of the area determination means. 11. In the color document image processing apparatus according to claim 10, the means for selecting the binarization processing result includes binarization processing using pseudo halftone processing for pixels determined to be in a color area or a photographic area. A color document image processing apparatus comprising means for selecting and storing results. 12. The color document image processing apparatus according to claim 9, comprising means for converting a plurality of input multivalued color image data into a plurality of binary data, and each of the multivalued color images. A color/monochrome identification means for determining whether a pixel is an area described in black and white or an area containing color information, and an identification result output from the color/monochrome identification means is stored as binary image data. A color document image processing device comprising means. 13. In the color document image processing apparatus according to claim 12, the means for accumulating the plurality of binary image data comprises means for independently accumulating the results of binarization processing for all colors;
means for accumulating only one of the plurality of binarization processing results; means for switching between the two types of accumulating means; and means for operating the switching means according to the discrimination result output from the color/monochrome discrimination means. A color document image processing device comprising: 14. In an image processing apparatus that handles document images including color images as digital data, means for independently binary-valued input multivalued color image data and outputting a plurality of binary image data; means for accumulating part of the value image data; color/monochrome identification means for determining whether each pixel of the multivalued color image is an area described in black and white or an area containing color information; text/photo area determination means for determining whether each pixel of the multivalued color image belongs to a line figure area or a gray area; A color document image processing device comprising a color identification means for determining whether the color of a pixel is equal to a predetermined color. 15. In an image processing device that handles document images including color images as digital data, means for binarizing and storing luminance information of input multivalued color image data;
It is characterized by having means for independently binarizing and accumulating image data for each color, and means for accumulating a binary image of luminance information and a binary image of luminance information only of a portion determined to be a specific color. Color document image processing device. 16. The color document image processing apparatus according to claim 15, comprising means for identifying, for each pixel of the multivalued color image data, whether the pixel belongs to a monochrome region or a color region; A color document image processing apparatus comprising means for storing a result as binary image data as part of the image data for each color. 17. A color document image processing apparatus according to claim 15 or 16, further comprising means for accumulating the various types of image data and means for switching between the accumulating means. 18. When storing a document image including a color image as digital data, input multi-valued color image data is converted into multiple binary image data including binary image data obtained by binarizing the luminance information of the image. A color document image processing device characterized in that the color document image processing device performs a color document image processing device. 19. Binary image data obtained by binarizing the luminance information of input multivalued color image data when storing a document image including a color image as digital data;
A color document image processing device that stores image data in the form of binary image data obtained by binarizing luminance information of only pixels displaying a specific color. 20. In an image processing device that handles document images including color images as digital data, means for independently binarizing luminance information of input multivalued color image data for each color, and for each pixel of the image. means for identifying whether or not the pixel has color information; means for identifying whether the pixel is a line figure area or a gray area; and means for determining whether the pixel is equal to a preset color. A color document image processing apparatus characterized in that a part of the binarization processing result is selected and output based on the results of the determination and identification. 21. In an image processing device that handles document images including color images as digital data, a means for identifying whether or not each pixel of the image has color information is provided in the portion that binarizes the image data. , means for identifying whether the pixel is a line figure area or a gray area, and means for determining whether the pixel is equal to a preset color,
A color document image processing apparatus comprising means for switching a binarization processing method according to the results of the determination and identification. 22. In an image processing device that handles document images including color images as digital data, means for independently binarizing and storing luminance information of input multivalued color image data for each color, and The present invention is characterized by comprising means for generating multi-valued image data for each color from the generated image, and means for synthesizing the generated multi-valued image data for each color and displaying multi-valued color image data. Color document image processing device. 23. The color document image processing apparatus according to claim 22, further comprising: means for determining whether a pixel belongs to a line figure area or a gray area from the accumulated binary image; 1. A color document image processing apparatus, comprising: a plurality of means for generating multivalued image data for each image; and means for selecting one of the outputs of the plurality of multivalued data generation means based on the determination result.