WO1991009488A1 - Picture filing device and method thereof - Google Patents

Abstract

A picture filing device which inputs pictures corresponding to documents as multivalued data, converts the multivalued data to binary picture data and stores the binary picture data in it, and a method of filing pictures. Picture elements in the areas having a specific color are extracted from an input picture, and luminance information representing monochromatic binary picture data and binary picture data specifying colored areas are stored on different planes respectively. On the plane for the luminance information, not only black picture elements but also other picture elements to be displayed in specific colors such as red, etc. are described inclusively. Picture elements, for example, of a red character, of a specific color to be displayed are extracted. The extracted picture elements of the specific color are recorded as '1s' on the plane of the specific color, e.g. an R-plane different from the plane for the luminance data. On the R-plane only picture elements having red color are recorded as '1s' and other picture elements are recorded as 'Os'. When outputting, picture elements whose contents on the plane for the luminance data are '1s' and whose contents on the R-plane are 'Os' are displayed in black. Picture elements whose contents on the luminance data plane are '1s' and whose contents on the R-plane are 'Os' are displayed in red.

Description

 Light w

 Image filing apparatus and method

[Technical field]

 The present invention provides a function for inputting and accumulating a document, particularly a color document including color information such as color photographs and redlining with red brush as image data, a display thread layout and a printer. Output to field

 Image filing apparatus and image information processing device

10 Regarding the processing method. In particular, mono click B image (mono ch r Qma tic images) and force La first image (co 1 or images) and the force ^s mixed picture can coding efficiency rather good, and also mono- click b binary data only the output Devices such as electronic filing devices that can encode a single image in a format that is compatible with faxes, etc.

15 Related to the image processing system. [Background technology]

Conventionally, in a device that treats a color image as digital data, the image is divided into three primary colors of red, green, and blue (primaryco 1 or _S ) and three types of multi-valued data ( Since then

20 In this specification, each component is referred to as R, G, and B, and the data of each color component is referred to as R data, G data, and B data. For example, a color copier or the like outputs the input R, G, B multi-valued image data as it is. Also, output to pre and

If binarization processing is required, such as 25 And binarize the data independently for each color. Examples of such a high-performance processing are, for example, JP-A-63-174472 and JP-A-64-51583.

 Japanese Patent Laid-Open No. 63-1747442 is a color image in which luminance information (luminance) is calculated from three components of R, G, and B, and the binarization process is switched according to the value. Japanese Patent Application Laid-Open No. Sho 644-511583, which discloses a processing device, decomposes a color image into four colors of yellow, magenta, cyan, and black, and converts each color into a binary value. It discloses an electronic file device that stores and converts it into a file. However, if the image information is decomposed into four colors of yellow, magenta, and cyan black, and if only the information of the black is output, the image of the colored portion is lost. No consideration is given to displaying intermediate colors such as color photographs.

 Since the color tone and saturation of a color image are greatly affected by the characteristics of the output device, these must be corrected when outputting accumulated data. However, in order to perform the correction process, it is necessary to treat the image as multi-valued data. Therefore, the method of binarizing a color image and storing it as in the above-described conventional technique cannot be applied to a single color image in which color reproduction is required.

In addition, a color document often has a large area of a mouth opening while including a color image as a part of the image. In particular, in black-and-white documents with red seals, color … Information is meaningful only with color, and it is not so heavy to reproduce colors accurately. Therefore, storing all R, G, and B data is inefficient.

 In addition, even in a color document in which color photographs are mixed, the area ratio often includes a large number of character areas of a monograph.

 Therefore, an image processing system that targets documents is not only a color image processing system that targets only one color image, but also has high coding efficiency for mono-text images and characters. Desired.

 However, if R, G, and B are individually coded, the redundancy is high and the coding efficiency is low. For this reason, it is not suitable for electronic file systems that store a large amount of document image data, or for faxes that require sending document images at a low data transfer rate.

 Also, the method of storing and outputting the RGB multi-valued data as it is would not be suitable for the above-described apparatus because the data amount would be enormous. In the following, the portion represented by a monochrome in the image is called a monochromatic region.

On the other hand, TVs or VTRs convert RGB color image data into luminance information and two types of color difference information (chr ₀ minance), perform orthogonal transformation, and calculate the coefficients. The technology of encoding and storing is widely used. This encoding method handles only luminance information in a monochromatic aperture region having no color difference information. Therefore, an image containing a large number of monochromatic regions And high coding efficiency.

 For this reason, Japanese Patent Application Laid-Open No. 63-92282 discloses an example s using this encoding method for a document.

However, the data code using this luminance / color difference conversion

The conversion method is intended primarily for devices that handle natural images such as TVs.

It was devised.

Therefore, it is suitable for images with gradual density changes,

Documents containing many parts with extremely large changes in density, such as characters

When dealing with images, the following problems arise.

(a) Since the density of line figures such as characters changes sharply,

Efficiency drops significantly.

 (b) When the compression efficiency by encoding is increased,

It becomes. .

Therefore, raster image data such as FAX

Data transfer, every time you convert data,

In other words, the data changes sequentially with each transfer.

 (c) Binary conversion of documents such as FAX and electronic file systems

Cannot output directly with a device that treats it as an image.

 (d) A conventional electronic filling device for black and white

Data compatibility is lost.

 As a specific color extraction technique, for example,

 6 2-2 3 0 1 6 5

From the top, has a strong specific color component defined for each device

And print it out with color toner on the paper to be printed

A two-color original reading device is known. This extraction method This method optically reads a color component determined in advance by the device and its corrected color component, and identifies whether or not the target portion is a set color based on the difference and the sum of the two. In this method, the center wavelength of the extracted hue is determined by characteristics of an optical finoletor or the like.

 Also, as an image data storage device that handles a single color image, various devices that target still images of a color are known. However, most of these types of devices are image data storage devices. Discrete cosine transform is used as the data encoding method.

 [Disclosure of the Invention]

 One object of the present invention is to efficiently encode and store various types of document images, regardless of whether they are a single document or black and white text. .

 Above all, it is possible to efficiently accumulate documents that have a mixture of a character area and a monochromatic mouth area. In particular, the monochromatic area such as a document with a red seal is mainly used, and the color For a multi-color document where reproduction is not important, the encoding is to be performed with higher efficiency.

 Another object of the present invention is to make it possible to change the color tone and saturation of an accumulated color image when outputting the image.

Furthermore, another object of the present invention is to allow the range of colors to be extracted to be freely specified from the outside according to the operator's sense, and to be able to specify the hue while checking the extraction results. It is here. Many fax and electronic file systems that have been widely used so far have an input / output device for binary black-and-white images. Therefore, the color document filing systems can also be used more widely if they are compatible with them. For that purpose, it is necessary to maintain compatibility in the data recording format.

 The image filing apparatus of the present invention is compatible with a conventional image filing apparatus that targets a monochromatic image. For this reason, color image data is stored in a format that is compatible with previously stored monochrome image data. In addition, the contents of each image can be displayed at high speed when searching for the contents of color image data stored in a large amount in a data storage device such as an optical disk.

An image filing apparatus of the present invention comprises: input means for inputting an image corresponding to a document as multi-valued data; and input data conversion means for converting the multi-valued data into binary image data. Image memory means for temporarily storing the binarized image data, and performing predetermined encoding to accumulate the image data stored in the image memory means. And an image storage unit having a code decoding processing function of decoding stored document image data, and outputting desired image data based on the binary image data of the image memory unit. Output data conversion means; and image display means for displaying a desired image based on image data from the output data conversion means. This type of image filing apparatus is disclosed in US Serial No. 07/479615 (EP Applicatio No 90102974.4) previously proposed by the present inventors. '

 This application is a continuation-in-part of U.S. serial number 07/479, 615, filed on February 13, 1990

 (Cont inua ti on-In-Part appli cation), the disclosure of which is incorporated herein by reference.

 In the present invention, similarly to the above-mentioned prior application, multivalued color image data is treated as a plurality of independent binary image data.

 Specifically, when one color image is recorded, three binary images, each of which records the R component, G component, and B component of the image, are recorded as R data, G data, and B data, respectively. These are stored as data, and at the time of output, the three binary images are combined to represent one color image. In the case of a monochromatic image, only the luminance data of the image is stored as a single image, and the monochromatic image is represented using the luminance data. The luminance data can be substituted by, for example, G data.

 Hereinafter, in the present specification, the data of the three primary colors of a color image will be referred to as R data, G data, and B data, respectively, and three binary image data used to represent one image will be referred to. Are individually referred to as planes. The luminance data is written as Y.

On the other hand, documents such as red and blue in images with red seals, etc. According to the present invention, the binary data Y indicating the luminance information and the binary data indicating only the portion of the specific color are stored as independent images. I do.

 Specifically, a plane of luminance data for recording a black-and-white binary image is accumulated, and for example, binary image data recording only pixels of a specific color to be displayed in red is accumulated in another plane. That is, the image data is stored as two binary images.

 Therefore, according to one feature of the present invention, there is provided means for extracting a pixel corresponding to a specific color from an input image.

 In multi-color documents, such as red-marked and red-corrected documents, the important thing is that red is “colored,” and it can be very accurate, as in color photos. Reproduction of rich colors is not generally required. Therefore, these images are handled in a recording format such as monochromatic image data and image data specifying a colored portion. That is, the luminance data Y is recorded as a single plane.

Here, according to one feature of the present invention, not only the black pixels are described in the luminance plane, but also the pixels to be displayed in red or another specific color are described. In addition, when the pixel is displayed at the mouth, a pixel represented by black, that is, a pixel of a specific color to be displayed in "red" such as a red character among the pixels of Y = "1". The extracted pixel of a specific color is extracted to a plane other than the luminance plane, for example, the R plane. Record as "1". Here, the R plane is recorded as binary image data in which only pixels written in "red" are represented by "1"-and other pixels are represented by "0". Then, since the pixels of the R plane are all "0" in the area described in the mono pixel, the coding efficiency is high, and the case where only the mono black binary image data is used is almost the same. Almost the same amount of data can be recorded including color information.

At the time of output, of the luminance data Y, pixels whose R plane content is "0" are displayed in black, and pixels whose R plane content is "1" are displayed in red. Since the luminance plane includes the pixels that should be displayed in red, if only the luminance plane ⁵ is displayed in mono, the multicolor document can be converted to the conventional monoc It can be displayed in the same way as if it were entered on the system.

 In addition, various writing instruments or papers with a ground color are used in documents. Therefore, even when simply saying "red", the range of the hue is wide. For example, the specifications for vermilion and stamp inks specify certain tolerances for luminance, saturation, and hue for each color, such as red and blue. Fall into that category. Therefore, in order to extract an arbitrary color, it is preferable that these characteristics can be specified.

Therefore, according to one feature of the present invention, means for determining whether or not the pixel is a pixel to be extracted based on the relative relationship between the values of the three colors R, G, and B; Pixel Two types of the above-mentioned judgment result and binary luminance data

From the combination of the above binary data, means for rewriting the determination result for the pixel is used. Hue range to extract

 In order to arbitrarily specify, and, means for externally specifying the color to be extracted, means for externally specifying the extraction range, means for converting the color of the specified hue and range to the RGB three primary color system, Based on this result, means for rewriting the contents of the means for making the above-described determination in pixel units is used. To confirm the judgment result, for example,

Means for displaying the image luminance information and the extracted pixels using different expression means, and the operator confirms the result displayed on the above-mentioned display means before accumulating the image, and extracts as necessary. Change of condition

The means of performing the change and the consequences of this

Means for inputting an image and identifying a specific color is used.

In order to display the extracted pixels by different expression means, only the extracted pixels are displayed in a specific color and the other pixels are displayed in black and white, or a binary image of luminance information and the extraction are performed.

Means for outputting an image which is an exclusive OR of the image of the output pixel, and alternately displaying the OR image and the binary image of the luminance information

Means are used.

 On the other hand, in order to express full-color images such as photographs, data of each plane of R, G, and B is recorded, and each plane is displayed as R, G, and B, respectively. In order to cope with the various images described above, the present invention is executed.

Brief explanation of binarization processing of data and its accumulation Do

 In the present invention, as the binarization processing, a binarization processing of binarizing by a predetermined threshold value and a dither processing of binarization by pseudo halftone processing are executed in parallel. Then, from these binarization processing results, the desired binarization data for each mode is selected and stored according to the color / monochrome identifier or the character / photo identifier. During image storage, the identifier of this mode is recorded together with the image data. This mode identifier is also referred to when displaying an image, and is used to specify data necessary for display and to determine an appropriate composition method. The contents of each mode are as follows.

 Mode (I) Monochrome mode: Stores only the luminance data Y in the image data. However, either the simple binarization processing result or the dither processing result is selected and recorded according to the character Z photo identifier.

Mode (Π) Multi-color mode: Extracts luminance data Y and pixel data to be displayed in a specific color such as red or blue from the input color image data, and Store image data on separate planes. For example, a binary image (Y plane) obtained by binarizing luminance data, and only pixels that are to be represented in red, for example, only pixels that are recorded as black in the binary image, Two binary images with the described binary image (R plane) are stored. The processing for accumulating the luminance data Y is the same as in mode (I). If the target pixel is a specific color, a character plane Select and record either the binarization processing result or the dither processing result according to the Z photo identifier, and if it is not a specific color, record "0" regardless of the text identifier identifier.

 At the time of display, the pixels included in the R plane are expressed in red. Other pixels are displayed in black and white according to the value of each pixel in the Υ plane.

 Mode (ΠΙ) Full color mode: R, G, and B color image data are stored in the R, G, and B planes, respectively. That is, one of the simple binarization processing result and the dither processing result is selected according to the color mono identifier and the character Z photograph identifier, and the three planes of R, G, and B are selected. Each accumulates. .

 During display, the values of the three planes R, G, and B are combined and displayed.

Mode (IV) Mixed mode: Identifies a color area that requires color information from the image and a monochromatic area, and stores the luminance information Y in the G plane for the monochromatic area. The R and B planes are left blank. However, in a document in which a character area and a photo area are mixed, for the photo area, binarized data by RG and B dither processing is stored in each plane. On the other hand, in the color area, binarized data obtained by dithering R, G, and B is stored in each plane. In addition, a color Z monochromatic identifier Fcm indicating a color region and a monochromatic region is used as a fourth plane. accumulate.

 Mode (I) is used when inputting a monochromatic document.

 The mode (Π) is a mode that is often used in office documents, etc., and is used for documents that have been stamped or corrected in a specific color such as red or blue on the mouth document. It is. However, this mode can be applied even when there are two or more specific colors to be expressed.

 The mode (m) is a mode used for a color photograph or the like.

 Mode (IV) is for documents that contain both monochrome and color photographs.

 To briefly explain the above operation, the image filing apparatus according to the present invention stores all color information of R, G, and B only when color information is required in an image. However, in the case of a document that has been designated as a monaco mouth image in the image or from the beginning, only luminance information is stored. For example, in the modes (I) and (Π), the luminance data Y is recorded as “1” for a pixel displayed in black and “0” for a blank portion. That is, in the mode (I), the stored image data is only luminance information.

In the case of multi-color documents that do not require accurate color reproduction, such as a document with a red seal, the mode (Π) can be used to reduce the amount of data in the monochrome portion by processing the mode (Π). It can be the same as the processing of I). That is, the luminance plane γ has black pixels Rather than recording only, record also including pixels that should be displayed in red or other specific colors υ and Ύ = “1” in pixels with red characters such as red characters A pixel of a specific color to be extracted is extracted, and the extracted pixel of the specific color is recorded as a plane different from the luminance plane. Therefore, the pixels of the plane of the specific color are all '0' in the area described in the mono pixel, so that the coding efficiency is high and the binary image data of the mono pixel is obtained.

Data with almost the same data amount as the case with only data, including color information.

On the other hand, in the mode (IV), which targets a document in which a funeral image such as a color photograph and a mono image are mixed, the color identification information F cm is stored. This makes it possible to suppress data redundancy in the mouth area. This mode is particularly effective when a large number of documents are input continuously using an automatic paper feeder or the like. If some pages of the input document contain a color photo, in mode (（), all pages of the document will be stored in three RGB planes. The amount of data required is three times as large as that required to store a monochrome image. On the other hand, in the mode (IV), most pages accumulate only the luminance data Y, and the R and B planes and attribute identifiers show blanks on most pages, so that efficient operation is possible. Encoding can be realized. In this case, it is considered that the attribute identifier outputs “◦” when the target area is a monochromatic image and “1” when the target area is determined to be a color area. In the area as three planes R, G, and B binary data are recorded. In the monochromatic area, the above-mentioned luminance data (which can be substituted with G data) and red and blue data are recorded. The data display method is switched according to the attribute information such as cm.

 In the present invention, since the image data is binarized and stored, the image data is stored in the M H

 (Modified Huffman) and MR (ModiίiedRead). Therefore, a recording medium such as an optical disk can maintain compatibility with a conventional monochromic system. Therefore, as described above, the luminance data Y can be output by the conventional monochromatic system. In addition, these methods of encoding a binary image do not cause a problem that the encoding efficiency is sharply reduced in a line figure such as a character.

 Furthermore, in these binary image data encoding methods, the data amount of a portion where 0s or 1s are continuous can be minimized. Therefore, information whose value changes in units of areas, such as color identification information, has extremely high coding efficiency.

In addition, the plane of the specific color stored in the mode (H) is blank except for the pixel of the specific color, so that high coding efficiency can be obtained. Here, the case where red is displayed as the specific color information to be added to the black and white image has been described, but when expressing two or more specific colors, the number of stored planes can be increased. By doing so, it can be executed similarly. Next, the case of storing a full-color image is described. <When a full-color image is represented by binary information, it is necessary to perform pseudo halftone processing to reproduce not only the density but also the color. However, if the 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 character areas in full-color documents and mixed-color documents, line figures are accumulated more finely by adaptive binarization processing. . On the other hand, in the color region, pseudo halftone processing is performed to reproduce colors even for line figures.

 When an image is displayed by superimposing three binary image data of R, G, and B, a color that originally does not exist may appear due to the difference in values between the three data. . This is called color noise. Color noise occurs in modes (m) and iN). In order to prevent this color noise, data equivalent to each pixel in three planes is recorded.

According to the present invention, in the mode (m), the color Z monochromatic mouth area determining unit determines that the area is a monochromatic area, and the character Z photograph area determining unit determines that the area is a character area. For a pixel, the same value is stored in the three planes R, G, and B. On the other hand, when a color image is output, image processing such as affinity conversion, change of a specific color, or adjustment of color tone requires the image data to be multi-valued at the time of processing. is there. Therefore, multi-value data is reproduced from the image stored as binary data. [Brief description of drawings]

 Fig. 1 is a diagram for explaining an example of image data in the case of storing a multi-color document. Fig. 1A is a diagram that is written in two colors, black and red. Fig. 1B shows a luminance data (Y) plane for recording a binary image corresponding to the black color of the document, and Fig. 1C shows the same. Indicates an R data plane that records a binary image corresponding to the red color of the document.

 Fig. 2 is a diagram showing the specific colors represented by the binary image data Y and R when storing multi-color documents.

 Fig. 3 is a diagram for explaining another example of image data when a multi-color document is stored. Fig. 3A is a multi-color document containing three colors of black, red and blue. Fig. 3B shows a luminance data (Y) plane for recording a binary image corresponding to the black color of the document, and Fig. 3C shows the brightness data corresponding to the red color of the document. An R data plane for recording a binary image is shown, and FIG. 3D shows a B data plane for recording a binary image corresponding to the blue color of the document.

 FIG. 4 is a diagram showing a specific color represented by each of the binary image data Y, R, and B when the multi-color document shown in FIG. 3A is stored.

FIG. 5 is a diagram for explaining an example of image data when a full-color document is stored, and FIG. 5A is an example of a full-color document including a color photograph, etc. Figure 5B shows the Figure 5C shows the R data plane recording the binary image corresponding to the red color of the document, Figure 5C shows the G data plane recording the binary image corresponding to the green color of the document, and Figure 5D The figure shows a B data plane that records a binary image corresponding to the blue color of the document.

 FIG. 6 is a diagram showing the relationship between the binary data recorded in each of the R, G, and B planes and the input colors in the full color mode.

 FIG. 7 is a diagram for explaining another example of image data when a full-color document is stored. FIG. 7A is a diagram illustrating a color region such as a color image and a monochromatic image. Fig. 7B shows an example of a full document with mixed areas, Fig. 7B shows an R data plane for recording a binary image corresponding to the red color of the document, and Fig. 7C shows Figure 7D shows a G data plane that records a binary image corresponding to the green color of the document, Figure 7D shows a B data plane that records a binary image corresponding to the blue color of the document, FIG. 7E shows an Fcm plane that records a color Z mockup identifier Fcm.

 FIG. 8 is a block diagram for explaining an embodiment of the image filing apparatus according to the present invention.

 FIG. 9 shows an embodiment of an image filing apparatus according to the present invention, and is a diagram showing an example of a configuration of an input data conversion section for inputting a document and converting the input image into binary image data. is there.

FIG. 10 is a diagram showing the contents of data recorded in each image (frame) memory in each input mode. FIG. 11 is a diagram showing codes of respective modes indicated by the input mode specifying unit.

 FIG. 12 is a diagram showing an example of the operation of a data selector for selecting image data to be stored according to each input mode and each area determination result.

 FIG. 13 is a diagram showing an example of code data when a target area is designated by the input mode designation unit along with the designation of the input mode.

 FIG. 14 is a diagram showing the contents of data recorded in each image (frame) memory in each input mode when a document to be input is limited to a line figure such as a character.

 Fig. 15 shows the content of data recorded in each image (frame) memory in each input mode when the input document is limited to photographs.

 FIG. 16 shows an embodiment of the image filing apparatus according to the present invention, in which each image data binarized by the input data conversion unit is shown.

FIG. 3 is a diagram illustrating an example of a configuration of a data storage unit for storing data.

 FIG. 17 is a diagram showing the format of image data recorded in the data storage section. FIG. 17A shows the mode (I), and FIG. 17B shows the mode (（). ), Fig. 17C shows the case of mode (ΙΠ), and Fig. 17D shows the case of mode (IV).

FIG. 18 shows an embodiment of the image filing apparatus according to the present invention, and the output data conversion for displaying the image data. It is a figure showing an example of 1 composition of a change part.

 FIG. 19 is a diagram for explaining an example of the operation of the data selector in the output data conversion unit for selecting each image data according to each mode and display condition.

 FIG. 20 shows an embodiment of an image filing apparatus according to the present invention, and is a view for explaining an example of the configuration of a multi-value processing section in an output data conversion section.

 FIG. 21 is a block diagram for explaining another embodiment of the image filing apparatus according to the present invention.

 FIG. 22 is a diagram showing the contents of binary image data recorded in each plane when a multi-color document is stored.

 Fig. 23 is a diagram for explaining another example of image data in the case of storing multiple documents. Fig. 23A is a diagram showing a multi-file containing black characters, red markings, and blue correction characters. An example of a color document is shown, FIG. 23B shows a luminance data plane for recording a binary image corresponding to the black color of the document, and FIG. 23C shows a luminance data plane corresponding to the red color of the document. An R data plane for recording a value image is shown, and FIG. 23D shows a B data plane for recording a binary image corresponding to the blue color of the document.

 FIG. 24 shows an embodiment of the image filing apparatus according to the present invention, and particularly illustrates a specific color identification section.

FIG. 25 is a diagram for explaining an example of the judgment used in the specific color pixel extraction unit, and FIG. 25A is represented by an RGB system. Fig. 25B shows an example of the red and blue range, and Fig. 25C shows another example of the red and blue range. Fig. 26 shows the specific color pixels. FIG. 9 is a diagram for explaining the operation of the extraction unit when expressing in an RGB system.

 FIGS. 27A to 27C are diagrams for explaining the principle of determination using a VCH system as a color system.

 FIG. 28 shows an embodiment of the image filing apparatus according to the present invention, and is a diagram for explaining the details of a specific color pixel extraction unit capable of arbitrarily specifying a range of colors to be extracted. is there.

 FIG. 29 is a diagram illustrating a configuration example of a parameter input unit.

 FIG. 30 is a diagram showing an arrangement of pixels to be referred to in the guest elimination unit.

 FIG. 31 is a diagram showing a configuration of a guest elimination unit. FIG. 32 is a diagram showing a configuration example of a data storage unit.

 FIG. 33A FIG. 33D is a diagram showing a pattern when a red ghost appears.

 FIG. 34 is a diagram for explaining an example of a configuration in which a black-and-white display device performs an image blinking display function for confirming a determination result.

 FIG. 35 shows an embodiment of an image filing apparatus according to the present invention, in which an example of a configuration of an image filing apparatus capable of designating a colored portion by an external instruction is shown. FIG.

[Best Mode for Carrying Out the Invention] First, a recording format of image data when a multi-document is stored, which is one of the features of the present invention, will be described with reference to FIG. This corresponds to the above-mentioned mode (（).

 Figure 1A shows an example of a document written in two colors, black and red. In the figure, the characters in the area 11 are black, and the characters in the area 12 are red. When this multi-color document is input in the above-mentioned mode (例 え ば), for example, the two binary image data stored in the data storage unit such as an optical disc are shown in Fig. IB and Fig. 1 respectively. As shown in Fig. C. Fig. 1B shows a plane for recording luminance data Y. The luminance data plane is recorded as binary data, and the characters described in black and red in the original image are described as monochromatic binary image data. In this case, in the mode (I) as described above, the pixel on which the character is written is "1", and the other blank pixels are "0". The mode for recording only the luminance data Y is the mode (I) described above.

 On the other hand, Fig. 1C shows the binary image data described in the R plane. Among the pixels described as “1” in the luminance data Y, “1” is described only for the pixels determined to be red. The structure of the part that extracts this red and binarizes it will be described later.

As a result, the following relationship exists between the luminance data Y (x, y) shown in FIG. 1B and R (X, y) shown in FIG. 1C. It stands.

 i f R (x, y) = 1 t h en Y (x, y) = 1

 Therefore, if the pixel in the R plane is "1" and the pixel in the Y plane is "0", no error occurs.

 Figure 2 shows the combination of binary data Y and R to represent each color in a multi-color document in mode (Π). In the figure, Y indicates the content of the luminance data, and R indicates the content of the R data, respectively. If the specific colors to be represented are three types of white, black, and red, the values that the data can take are the three shown in the figure.

 On the other hand, when displaying a multi-color image from two pieces of binary image data (shown in Figs. 1B and 1C) stored in a data storage unit such as an optical disk, There are two types of output, output and mono output. Fig. 2 also shows the colors at the time of output. When the multi-color image shown in Fig. 1 is displayed in color, the image data shown in Fig. 1C is displayed in red, and parts other than red are displayed in black and white according to the image data shown in Fig. 1B. Specific means for realizing this display method will be described later.

 Even when a plurality of stored images are displayed continuously, high-speed display can be achieved by reading out only the Y data and displaying it in a monochromatic manner. By using only this Y data, data compatibility can be maintained with devices that handle monochromatic binary images, such as FAX. This feature is the same in other modes.

Note that here, the color reproduced during image display is red. One color is targeted, but the same applies to other colors. Also, a plurality of can be displayed as specific colors. In this case, for example, when η specific colors are to be expressed, η + 1 plane binary image data is stored.

 —As an example, Fig. 3 shows an example of expressing two colors, red and blue, in addition to black and white. Figure 3 shows an example of a multi-color document in which black, red, and blue characters are described. In the figure, the characters in region 16 are black, the characters in region 17 are red, and the characters in region 18 are blue. When this multi-color document is input in the above-mentioned mode (Π), the images are stored as the three binary image data shown in Figs. 3〜 to 3D. Fig. 3 shows a plane for recording luminance data. FIG. 3C is a red plane, and FIG. 3D is a blue plane, each of which is binary image data for recording a pixel of a specific color to be displayed. In this case, Fig. 4 shows the combinations of binary data Y, R, and の for expressing each color in the document, together with the colors at the time of output.

Next, the mode (m) for full-color documents such as color photographs will be described. Fig. 5 shows the recording format of image data in mode (m). The image is recorded as R, G, B three binary image data. Fig. 5A shows an example of a document in which color photographs and characters are mixed, Fig. 5B shows an R plane, Fig. 5C shows a G plane, and Fig. 5D. Indicates the contents of each binary image data of the B plane. In Figure 5A, the characters in area 21 Is black, and the characters in the area 22 are written in a certain neutral color. On the other hand, the area 23 indicates the area where the color photograph is described.

 Here, in this mode, the colors recorded according to the contents of the image data of each plane are shown in FIG. In mode m), white and black are treated as one color. Regarding the binarization processing, when a full-color image such as a color photograph shown as area 23 in FIG. 5A is represented by binary data of each of the R, G, and B components, Each of the multivalued data of R, G, and B is binarized by pseudo halftone processing. Therefore, in the regions 33, 43, and 53 in the figure, data obtained by independently binarizing the multivalued data of 1, G, and B by pseudo halftone processing are recorded.

 On the other hand, in this mode, for example, a black-and-white character area such as the area 21 is not subjected to the pseudo halftone processing at the time of the binarization processing. This is performed by separating the character area and the photo area from the image of the input document and using the results to select the results of simple binarization processing and pseudo halftone processing. In addition, in order to correctly represent the area in which black characters are described, the pixel values are made equal among the R, G, and B planes. Specifically, pixels in the character area that are determined to be a monochromatic area are recorded with the same values in the three planes R, G, and B. Therefore, each pixel in the areas 31, 41, and 51 records the same value.

However, even in the character area, for example, the area 22 When it is necessary to reproduce colors with colored characters. Pseudo halftone processing is required to represent colors. Therefore, in addition to identifying the text area and the photograph area, the color area and the monochromatic area are identified. Each pixel of the colored character areas 32, 42, 52 has an independent value recorded.

 The selection of the binarization processing based on the various judgment results and the specified mode will be described in detail later.

 Next, the mode (IV) for documents that contain both color areas and monochromatic areas is described.

This mode is particularly effective when a document such as a document composed of many pages is stored as image data. In general, when inputting a document or the like consisting of many pages, for example, when using a device for automatic input called a sorter, there is no need for attendance by an operator. Thus, the efficiency of input work can be improved. However, if there is no operator during the input, it is difficult to switch the mode for each page. Then, if a color region is included in some pages in the input document, it is necessary to input images of all pages as color images. If images of all pages are stored in the mode (IE), the image data in the monochromatic area requires three times the data amount as when input in the original monochromatic mode (I). . On the other hand, in the mode (IV), the fourth plane has an identifier indicating whether each part in the image is a color area or a monaural opening area. Since this identifier can be realized by a binary flag, each of the Can be recorded in the same way.

 Fig. 7 shows an example of the recording format of image data in mode (IV). Fig. 7A shows an example of the same color document image as Fig. 5A. Fig. 7B, Fig. 7C, and Fig. 7D are the binary image data of R, G, and B, respectively, and Fcm is the color area and monoc for each part of the image. This is a binary identifier for identifying the mouth area. Here, it is assumed that the identifier F cra (X, y) is “1” when the pixel IM (x, y) is a color and “0” when the pixel IM (x, y) is a monochrome. Then, the binary data F cm becomes “1” in the areas 92 and 93 corresponding to the color areas 22 and 23 in the image of FIG. 7A, and in other parts, It becomes "0". Here, the portion where F cm = 1 describes the binary image data of R, G, and B in each of the R, G, and B planes. In this region, black is described as R = G = B = 0. On the other hand, for example, in the black character area 21, the recording format is the same as in the monochromatic mode, and the pixels described as black are the R, G, and B planes. Write "1" only in one plane. Here, the G plane is defined as the plane.

In the image display mode, in mode (IV), only the G plane is read as image data in the area of F cm = 0, black if G = 1, and white if G = 0 I do. For pixels with F cm = l, read three planes of R, G, and B and display them in the same way as in mode (Π). Therefore, in the mode (IV), the image output unit performs G-play according to the value of the identifier Fcm. Switch the display color of the buttons.

 When image data is recorded in mode (IV), Fcm = 0, R = 0, and G = 0 are always obtained for monochromatic character images, and high coding efficiency is obtained. Therefore, it is possible to accumulate documents mixed with color images with almost the same data amount as the input in the mono mode. On the other hand, in the color area as well, the identifier Fcm switches between 0 and 1 in units of area, so that the coding efficiency is high and the image size is almost the same as the mode (IE). Data can be stored.

 Next, an embodiment of an image filing apparatus that stores image data in each of the above modes will be described. First, FIG. 8 shows an overall configuration of an embodiment of an image filing apparatus according to the present invention.

In the figure, 100 is a color scanner that optically reads an image such as a document and outputs multi-valued digital data with respect to, for example, RGB three components, and 150 is a color scanner that has the modes (1) to (IV) described above. ) Is an input mode designator that designates one of the input data. Reference numeral 200 denotes an input data converter that binarizes the input multivalued data of each RGB color for data storage. 6 β0 is an image memory for temporarily storing the binary image data output from the input data conversion unit 200, for example, four frame memories 6 10, 6 2 0, It consists of 630 and 640 forces. 7000 is an output data conversion unit that converts binary image data into a format suitable for input to a display device, such as a high-definition color CR, for example.750 is an input unit that inputs RGB multivalued data. Power This is an image display unit such as a high-definition CRT that displays color images. On the other hand, 800 is a large-capacity memory such as an optical disk that encodes and stores binary data in the image memory 600 or reads and decodes already stored data. Reference numeral 900 denotes a data storage unit composed of a re-writing device, and 900 denotes a control unit for controlling the entire filing device.

 Here, one of the characteristic parts of the present image filing apparatus is the input data conversion unit 200. Of the components shown in FIG. 8, the part related to image input will be described with reference to FIG. In the figure, 6 1 0, 6 2 0, 6 3 0,

 640 is a frame memory for temporarily storing binary data of each plane of the input image. The input data conversion unit 200 is composed of an area determination unit 400, 2. a value conversion unit 300 and a selector 500, and is based on a mode designation signal and various area determination results. In addition to selecting the image data to be recorded in each frame memory together with the binarization processing.

The area determination unit 400 that determines the area of the input image is based on three RGB multi-valued image data obtained from the color scatterer 100, and each pixel of the input image is defined as a color area and a model. Color identification block 4101 that identifies which of the nodal areas belongs, whether or not each pixel of the input image belongs to a predetermined color such as red or blue Color identification unit 420 inputs multi-valued digital image data and determines whether each pixel of the input image belongs to a line figure area such as a character or a halftone area such as a true Photograph territory to do It is composed of an area judging section 4300.

 Further, a binarizing unit 300 that binarizes RGB multi-valued images independently of each other includes a binarizing processing unit 340 that binarizes images such as characters and line figures, It is composed of a dither processing section 350 for performing binarization processing by pseudo halftone processing. The binarization processing section 340 binarizes multi-valued image data with a predetermined threshold, and is suitable for binarizing images such as characters and line figures. The dither processing section 350 is adapted to binarize by periodically changing the threshold value according to the position of the pixel, and is suitable for binarizing a photograph or a color image. As the pseudo halftone processing, besides dither processing, there is an average error minimum method, and this may be used. Halftone processing by the average error minimization method is suitable for binarization of periodic images such as halftone images.

 Based on the mode designation signal F mod from the input mode designation unit 150 and the judgment results from the area judgment unit 400, the selector 500 outputs three RGB images. The data to be stored in the image memory 600 is selected and output from the two types of binarization processing results for.

Here, in each mode, the contents of the binary image data recorded in each frame memory 61 0, 62 0, 63 0, 64 0 in the image memory 600 are described. Figure 10 shows. Hereafter, the data developed in the frame memory 610 will be referred to as DATA-1, 620, 630, 640, and the data accumulated in DATA 620 will be referred to as DATA-2, respectively. DATA— 3, Called DATA-4. In FIG. 10, "1" indicates that data is not recorded. The binary image data stored in the mode (I) is only the luminance data Y and is recorded in the frame memory 610. In the mode (Π), two planes of binary image data are stored, and in the mode (IE), three planes of binary image data are stored. Further, in the mode (IV), four planes of binary image data are stored in the frame memories 61, 62, 63, and 64.

 Now, in order to briefly describe the features of the present embodiment, it is assumed that the multi-value data of RGB of each pixel on the input image is simultaneously output from the force raster 100. Signal line 110 sends R data, signal line 120 sends G data, and signal line 130 sends B data. Further, in the present embodiment, the luminance data Y is substituted by G data for simplification of the description, but the luminance data Y can be calculated by performing arithmetic processing on three types of multi-valued data of RGB. .

 Based on the input RGB multi-value data, the color Z monochromatic discriminating section 410 uses a known method to generate binary colors used in the modes (m) and (IV). Outputs the mono identifier F cm.

On the other hand, the purpose of the color identification section 420 is to extract the pixels recorded in the plane of the specific color in the mode (Π). In this embodiment, a case where a red pixel is extracted will be described as an example. However, the number of specific colors to be extracted is one or more, and the color itself is also set arbitrarily. Out of color identification section 420 The identifier F rb to be input has a different format depending on the number of colors to be extracted, and becomes a signal of lb it when one color of red is extracted. The following relationship exists between the number of specific colors Cnb extracted here and the number of bits N of the identifier F rb.

N≥ log ₂ (Cnb + 1)

Various methods for extracting a specific color, such as a method using deviation of values between R, G, and B, are already known, and these may be used.

 On the other hand, the character Z photograph area determination section 4

 The binarized data obtained by the simple binarization process suitable for characters and line figures obtained from 300 and the binarized data obtained by the pseudo-halftone process suitable for photographs and color images The purpose is to make a selection. Specific means are described in, for example, JP-A-63-316656. The judgment result Fdm is binary data. If the determination result Fdm is different for each color, the output of the selector 500 becomes unstable. Therefore, the determination of the character / photo area is performed on the luminance data. In the present embodiment, the luminance data is substituted for the G data as described above.

FIG. 11 shows an example of the mode designation code Fmod input from the input mode designation section 150. The selector 500 is driven by an identification code for selecting four types of input modes. The selector 500 is composed of three planes of RGB binary image data output from the binarization processing section 340 and the dither processing section 350, respectively, and a total of 6 planes of data. The four frame memories in the plane image memory 600 Of 6 20, 6 3 0, 6 4 0:! Select the binary data to be output to 4 to DATA-4 to DATA-4.

 FIG. 12 shows an operation example of the selector 500 in this embodiment. In the figure, the data output from selector 500 is represented by Rc, Gc, and Bc, and the data of R, G, and B obtained by binarization processing section 340 in FIG. The results of the binarization processing, denoted by R d, G d, and B d, are the binarization processing results of R, G, and B obtained by the dither processing section 350.

 In the mode (I), only the luminance data is output, and only one of Gc or Gd is output to the frame memory 610. Here, switching between G c and G d is executed according to the output F dm of the character / photograph area determination unit 430. However, if the image to be input is composed of only characters or photos, it is possible to specify one of them externally using the input mode specification unit 150. This will be described later.

 In the mode (Π), in the present embodiment, binary image data for two planes is stored in the frame memories 610 and 620. The selection of the data to be stored is controlled by the output Fdm of the character / Z photograph area determination unit 430 and the output Frb of the color identification unit 420. Where D ATA — 1 is in mode

Similarly to (I), one of the simple binarization processing result Gc and the dither processing result Gd is selected and recorded according to the identifier Fdm. In addition, the brightness plane of DATA-1 should be displayed in a specific color instead of recording only black pixels. Record including pixels. This embodiment shows a case where 1-bit code data is used as the identifier F ib. The identifier F rb is “1” when the target pixel is “red” of a specific color, and is “0” otherwise.

DATA—2 selects one of the simple binarization processing result G c and the dither processing result G d according to the identifier F dm when the identifier F rb is “1”, and the identifier F rb is “0” In this case, DATA-2 is "0" regardless of the identifier Fdm. In the mode (Π), the switching between G c and Gd is selected by the identifier F dm similarly to the mode (I).

 In mode (（), images are stored as R, G, and B 3 plane binary images. Here, the data to be stored are the R, G, and B planes. When the Carono Mono identifier F cm is “0”, one of the simple binarization processing result G c and the dither processing result G d, R d, or B d is performed according to the identifier F dm. Select one and accumulate 3 planes in DATA-1 to 3 respectively. When the identifier Fcm is "1", the dither processing results Gd and RdBd are selected regardless of the identifier Fdm, and three planes are stored in DATA-1 to DAT-3, respectively.

On the other hand, in the mode (IV), the image data of R, G, and B and the identifier Fcm are stored in four frame memories 61, 62, 63, and 60, respectively. Each accumulates. Here, one of the G plane and the luminance plane is added to the frame memory 610, and the output of the color / monochrome discriminating section 410 is provided. The data is selected and accumulated by the force Fcm, but as described above, in the present embodiment, the G data and the luminance 'data are the same, so there is usually no need to select. However, in a document in which a text area and a photo area are mixed, the output of the binarization processing unit 340 is selected for the line-shaped area of the mouth opening, and for the other areas, Select the output of the dither processing section 350. Therefore, in the mode (IV), when the output Fcm of the color Z monochromatic mouth identification unit 410 is “0”, the simple binarization processing result Gc and the dithering are performed according to the identifier Fdm. One of the processing results G d is selected and one plane is stored in DATA-1. Then, according to the identifier Fdm, “0” and one of the dither processing results R ci and B d are selected and stored in DATA-2 and DATA-3, respectively. On the other hand, when the identifier F era is “1”, the dither processing results G d, R d, and B d are selected regardless of the identifier F dm, and three planes are stored in DATA-1 to DATA-3. In addition,

D ATA — 4 stores the identifier F cm.

 The examples described so far assume that the target document is a document in which characters and photographs are mixed. However, when inputting a document in which the entire document is composed of characters or photos, for example, the input mode specifying section 150 can be used to specify other modes (I) to (! V). , [Text], [Photo],

By designating the type of area such as [mixed], the operating condition of the selector 500 can be determined independently of the character / photograph area determination unit 4300. In this case, the input mode finger The signal from the fixed section 150 to the selector 500 is input to the four input mode selection codes and the identification code according to the target book.

2 b i t is added.

 FIG. 13 shows an example of the output code from the input mode specifying unit 150 in this case. Basically, for each of the four modes, there are two types: [Text], which does not select dither processing, [Photo], which must be selected, and [Mix], which is based on the output of the text Z photo area judgment. I do. However, in the case of [character], the output of the binarization processing section 340 is normally accumulated. However, in the mode (m) and the mode (IV), the intermediate colors need to be reproduced. Therefore, the output of the binarization processing section 340 is not stored in the color area. Absent. Therefore, in mode (II) and mode (IV), the case where. [Character] is specified is excluded.

 Here, an example of the operation of the selector 500 when [Character] is specified is shown in Fig. 14, and an example of the operation when [Photo] is specified is shown in Fig. 15 respectively. These operations are the same as the operations of the selector 500 shown in FIG. 12 divided into the case of the character area and the case of the photograph area. Is omitted.

 Next, storage of the binary image data selected by the selector 500 will be described. DAT A—1, DAT A—2, D AT A—3, stored in the frame memories 61 0, 62 0, 63 0, and 64 0 of the image memory 600 mm, respectively.

DATA— Binary data of 4 is encoded by known means After that, it is stored in a large-capacity data storage device such as an optical disk. An example of the configuration of the data storage unit 800 will be described with reference to FIG. In the figure, reference numerals 61, 62, 63, and 64 denote frame memories for storing binary image data for each plane, as described above. 0 0 corresponds to the data storage unit shown in FIG. In the figure, reference numeral 8100 sequentially reads out the data held in each frame memory and outputs the data to the code decoding processing unit 820, and the code decoding processing unit 820 decodes the data. This is a selector that outputs the obtained data to each frame memory. The encoding / decoding processing section 820 encodes the binary data by a known means, outputs the encoded data to a data storage section 850 such as an optical disk, and stores the data in the data storage section 850. The decrypted data is decrypted and output to the selector 810.

When an instruction for data storage is given from the control unit 900, the filing apparatus first transfers the header information to the optical disk 850 and stores it. The header information includes, for example, information used at the time of retrieval, such as the name of the document, and the size of the document. However, in this embodiment, the mode at the time of input is recorded by a code as one item of the header information. After accumulating the header information, according to the mode, the binary image data of 1 to 4 planes is sequentially encoded by the encoding / decoding processing section 820 and accumulated in the data accumulation section 850 I do. The order of the data to be stored is selected by the selector 8100 according to an instruction from the control unit 900. Here, the recording order of the data stored in the data storage device 850 will be described with reference to FIG. FIGS. 17A to 17D are examples of recording formats of image data input in modes (I) to (! V), respectively. In the figure, 910 is an area for recording header information, 920 is DATA-11, and 930, 940, 950 are DATA-2, DATA-3, and DATA-4, respectively. It is a local area.

 In the mode (I), the image data to be stored is only the brightness plane, and is stored in the frame memory 61 0.

DATA— Stores 1. On the other hand, in modes (Π) to (m), the luminance plane is stored next to the header information as image data, and the plane of each color is recorded in: ^. When multiple specific colors are expressed in mode (Π), the plane of each color is recorded after this luminance plane. As a result, when searching and displaying images, a monochromatic binary image can be output simply by reading the first data immediately after the header information. On the other hand, in mode (IV), G data is recorded immediately after the header information. This is because the G data has the spectral spectrum located at the middle of the three types of data of _R , G, and B, so that the phenomenon in which a specific color is not reproduced is least likely to occur. Therefore, even in this case, when displaying, it is possible to output a monochromatic binary image simply by reading the first data immediately after the header information. On the other hand, in the recording area 911 in the header information 910, Record the code information indicating the mode at the time of image input. Next, a process of reading out the image data stored in the data storage unit 800 and displaying the image will be described.

 First, the image data stored in the data storage device 850 is read, and the frame memory 610, 620, 630, 640 of the image memory 600 is read. A description will be given with reference to FIG. First, the header information of each data recorded in the data storage device 850 is first read, and the target data is searched. If the target data is confirmed, then the number of planes to be read is determined by the mode identifier F m Q d recorded in the recording area 911 in the header information. Then, the data stored in the data storage device 850 is read out for a required number of times, and each is decoded into binary data by the code decoding processing section 820, and then the frame memo is read. Re-store the data from DATA-1 to DATA-4 in 61 0, 62 0, 63 0, 64 0.

 Next, the frame memory 61 0, 62 0, 63 0,

6 4 0 deployed in DATA - 1 from DATA - at 4 output data converter ₇ 0 0, the format of the data that is appropriate to the input of the image display unit 7 5 0 etc.髙精thinning La one CRT It is converted and displayed on the image display section 750. Output data converter

The operation at 700 is to generate three planes of display image data of RGB from the inputted one to four planes of binary image data. Hereafter, here, the display data of each of R, G, and B reproduced from the image data will be described. Are called DATA-R, DATA-G, and DATA-B. Here, the relationship between DATA-R, DATA-G, and DATA-B and the colors displayed on the image display section 750 is the same as in FIG. If all of DATA-R, DATA-G, and DATA-B are "1", "white" is displayed, and if "0", "black" is displayed.

 Next, an example of the configuration of the output data converter will be described with reference to FIG. In the figure, a data selector 710 is controlled by a control unit 900 in accordance with a mode identifier Fm0d in the read header information, and DATA-1, DATA-2,

The required data of DTA- 3 and DTA- 4 are synthesized, and DTA-R, DTA-G, and DTA-B are output. The details of this operation will be described later. The RGB three-plane binary data output from the data selector 710 is converted to data in a format corresponding to the image display unit 750 by the multi-value processing unit 720. Is done. In this process, the binary data is converted into a signal for controlling the blinking of each pixel of the image display unit by, for example, a bit shift. The operation of the output data converter 700 depends on the mode of the target image, the type of display, and the content of the data.

FIG. 19 shows an example of the operation of the data selector 710. Here, for example, when an image in mode (Π) is displayed in color, DATA-R, DATA-G, and DATA-B are determined by the logical operation of DATA-1 and DATA-2. If only luminance information DATA—1 is "1", DATA—R, DATA—G, and DATA—B all become “0” and “black” is displayed.

 D ATA—2 force s If “1”, set “1” to D ATA—R only to display “red”.

 —On the other hand, the timing for displaying an image can be as follows. For example, when displaying a monochrome image, the luminance data Y is input to the frame memory 610, and at the same time, the data selector Ί10 to the RGB three-plane data are input. As a result, luminance data is output. Then, image data having values equivalent to the R, G, and B planes are sent from the multi-value processing section 420 to the image display section 750, so that The image is displayed on the image display section 7500.

 On the other hand, when displaying a full-color image in mode (Π), the values of DATA-R, DATA-G, and DATA-B are independently set by DATA-1, DATA-2, and DATA-13, respectively. I can decide. As for the timing of the image display, the G, R, and B planes are sequentially displayed at the same time as DATA-1, DATA-2, and DATA-3 are written to the frame memory.

 When displaying image data in mode (ΠΙ), the final color cannot be determined until three planes of data are stored in the frame memories 310-330. In this case, it is also effective to display the luminance data in a monochromatic image as in the first mode (I) and to sequentially rewrite the luminance data.

On the other hand, according to the instruction from the control unit 900, data selection is performed. By controlling the operation of the data 710, for example, a color image can be displayed in a monochromatic image. Specifically, for example, in the case of a monaco display, this can be realized by reading out only DATA-1 and displaying this DATA-1 in all of R, G, and B.

 Also, for example, by displaying only DATA-2 of the image data of two planes of a multi-color document input in the mode (Π), only the red plane is selected and displayed. You can also do it.

 By the way, the full-color image targeted by the mode (ΙΠ) or (IV) originally needs to express halftone in halftone. In particular, in color photographs and the like, subtle differences in color tone due to the characteristics of the output device are perceived as significant differences in human vision. Therefore, in order to display a full-color image with high image quality, it is necessary to correct or change the luminance and chromaticity of the halftone data. In this embodiment, since the image is stored as binary data, for this purpose, it is necessary to convert the binary image data into multi-valued data.

FIG. 20 shows an example of the configuration of the multi-value processing section 720. In this example, three types of binary data, DATA-R, DATA-G, and DATA-B, are simultaneously input and output to the image display section 750. It is possible to convert one plane at a time by using memory. In the figure, reference numerals 731, 7332, and 7333 denote halftone converters that reproduce multi-valued data from binary image data. 741, 742, and 743 are color tone conversion units that convert the obtained halftone data, and 751, 752, and 753 are shift registers that multiply binary data by a specific amount. Reference numeral 70 denotes an area judging section for judging whether the input data is binary data such as a line figure or pseudo halftone data, and 761, 762, and 763 are area judging sections. This selector selects one of the two types of halftone data based on the output of.

 Here, for simplicity, the operation for one plane is described, but the same applies to other planes. Here, the halftone conversion section 731, reproduces multi-value halftone data from the pseudo halftone image. Various types of reproduction methods are already known, for example, disclosed in US Serial No. 07 / 272,447 (Japanese Patent Application Laid-Open No. 2-93037), A method of extracting the distribution density of black pixels can be applied. The image converted into the halftone is subjected to necessary conversion by the color tone conversion unit for each of the RGB planes. The color tone conversion unit 741 can be represented by, for example, RAM (Read On Iy Memory). In addition to setting the contents of the RAM in advance, for example, a method of transferring the data from the control unit 900 or selecting it according to a plurality of types of internal devices set in advance It can be realized by such as.

On the other hand, in the area of a line figure such as a character, the positional relationship of the black pixels is important, and expressing the shading, rather, lowers the resolution and deteriorates the image quality. . Therefore, for example, input of the display unit 75 0 connected to the system When the input range is 8 bits, the image quality is prevented from deteriorating by simply shifting the input binary data by 8 bits.

 The selector 761 selects and outputs one of the output of the color tone conversion section 741 and the output of the shift register 751. This selection is switched by the output of the area determining unit 770 which determines whether the area to be output belongs to the linear graphic area or the pseudo halftone area. As a means for determining the region of the input binary image, many methods are already known, and these methods are also used in this embodiment.

 The image selected by the selector 761 is transferred to an image display unit 750 such as a CRT and displayed.

As a result, halftone images, such as color photos, can be stored as binary data and displayed as halftone images. It can also perform conversions such as density and chromaticity.

 In this embodiment, an example in which three planes of binary data are simultaneously converted has been described. However, the provision of a means for temporarily storing data allows each plane to be converted. They can also be converted in the same order. Also, the case where the image is displayed on the CRT has been described as an example, but the image can be similarly output to another output device, for example, a printer.

However, many color documents used in general office work are described in black and one or two other colors. For example, a document having a vermilion imprint or a document with a redline correction. These are different colors from black It is important that the information described in the above is important, and it is not necessary to accurately reproduce colors. As described above, in most cases, color planes are blank in most parts of the document, so when encoding and storing binary images, high encoding efficiency can be obtained. it can. The color information is input by a combination of three types of multi-valued RGB data. Therefore, by specifying three types of RGB values indicating the range of a specific color to be extracted in advance, a specific color portion can be extracted. However, pixels with sharp changes in color, such as the outline of a line, do not accurately represent the original color information. This phenomenon is called ghost. It is difficult to judge the exact color of the pixel where the goster occurred. In this case, it is necessary to determine the correct color by referring to the color distribution state for the neighboring pixels.

 In order to arbitrarily specify the hue range to be extracted, it is important to specify the color using a color system that is easy for humans to specify, such as VCH or L, a *, b *. is there. In other words, it is necessary to use a color system centered on color difference that is close to human senses. To do this, it is possible to specify an arbitrary color with an arbitrary width by calculating RGB values representing the hue range specified in the VCH system and rewriting the judgment part. Wear.

On the other hand, confirmation of the judgment result can be realized by displaying an image on a CRT, for example, before accumulating the data, and specifying the re-input by the operator. Here, the image display For example, a color image based on the determination result may be displayed. However, even in the case of a black-and-white CRT, it can be expressed by blinking the extracted part.

 Therefore, in the following embodiment, when storing a multi-color document in the above-described mode (で き), the range of a specific color to be extracted can be freely designated according to the operator's sense, and the extraction result can be obtained. The following describes an image filing device that can specify the hue while checking.

FIG. 21 shows the basic configuration of the image filing apparatus. The basic configuration is the same as that of the image filing apparatus shown in FIG. 8 described above. That is, in the figure, 100 is an image input device such as a color scanner that reads a color document and outputs RGB multi-valued image data, and 200 is a multi-valued image data. An input data conversion unit that binarizes the binary image data, 600 is a surface image memory that temporarily stores the binary image data, and converts two to four independent binary image data (planes). Frame memories 6 10, 6 2 0, 6 3 0 to be stored respectively (three memories are shown in the figure, but they change according to the number of specific colors to be extracted) Power. 700 is an output data converter that outputs color image data that is more multi-valued than RGB binary image data, and 500 is a CRT (Cathode Ray Tube) that displays color images. The image display unit 800 is a data storage unit that encodes binary image data and stores the data in a large-capacity data storage device 850 such as an optical disk. This is the same as the embodiment described above. The feature of the present embodiment lies in the configuration of the input data conversion unit 200. In particular, for each image in the image, whether the color of the pixel is a hue in a predetermined range is set. The specific color discriminating section 450 determines whether or not to perform the judgment. In the input data conversion unit 200, reference numeral 300 denotes a binarization unit that inputs multi-valued RGB data and outputs binary image data, and reference numerals 501 and 515 denote binary data. The selector 550 is a register which always outputs a specific value.

 The R, G, B multi-valued image data output from the image input unit 100 is input to the binarization unit 300. The binarization processing unit calculates luminance data from the RGB data and stores the binary image data of the luminance data in the memory 6100 in the image memory 600. The brightness data can be approximated by G data without calculating the brightness data.

 On the other hand, the multi-valued data of R, G, and B are also input to the specific color identification unit 450. The specific color identification section 450 identifies whether each pixel expresses a preset specific color based on the input R, G, and B data. The method of identification will be described in detail later.

In this embodiment, the purpose is to identify a plurality of colors at the same time. Here, as an example, a case where two colors, red and evening, are distinguished from other colors will be described. In this case, there are three types of judgment results: "red", "靑", and "other". FIG. 2 shows an example of the operation of the output of the specific color identification section 450 and the operation of the selectors 510 and 515.

 Here, FB / W, FLGR, and FLGB represent the output of the specific color identification unit 450. After FB / W is binarized with the same value as the luminance data of the target pixel, the pixel to be expressed in white is “0” and the pixel to be expressed in black is “1”. FLGR is a red identifier indicating whether the target pixel is red, and FLGB is a blue identifier indicating whether the target pixel is blue.

 Here, FLGR and FLGB are determined by the following conditions.

 FLGR = 1: Target pixel = "red"

 And FB / W = 1

 FLGB = 1: Target pixel = "blue"

 And FB / W = 1

 In other words, in order for FLGR to be 1, it is necessary that the color identification result is "red" and the binarization result of the luminance data is black.

The selectors 51 and 51 in FIG. 21 correspond to the memories 62 and 63 of the image memory 60 in accordance with the output of the specific color identification unit 450. Select the data to be stored in 0. On the other hand, the register 550 always outputs "0". Here, FIG. 23 shows an example of image data stored in memories 610, 620, and 630 in the image memory 500. Fig. 23A is an image of the office document at the Monok-guchi with red stamping and blue corrections. In the figure, 1 1 1 is a red seal, 1 2 1 and 1 2 2 are Indicates the area of the character added in blue.

 When this document image is input by this device, the binary image data stored in the three memories 61 0, 62 0, and 63 0 are shown in Fig. 23B and Fig. 23B, respectively. Figures 23C and 23D show it. FIG. 23B is an image stored in the memory 610 and is a result of binarizing the G data. If only this image data is used, a monochromatic binary image of the input document can be output. FIG. 23C and FIG. 23D show the image data stored in the memories 620 and 630, respectively. Only the portion marked in red in the document is stored in the memory 620, and only the portion marked in blue is stored in the memory 630.

 The image data is stored by encoding each of the three binary image data. The storage of the image data is the same as the storage of the multi-color document in the mode (II) described above, and the details are omitted.

 Next, the configuration of the specific color identification section 450, which is a characteristic part of the image filing apparatus according to the present embodiment, will be described in detail.

FIG. 24 shows a configuration example of the specific color identification unit 450 and an example of a configuration around it. In the figure, 210 indicates whether or not the pixel belongs to a predetermined hue range based on the relative relationship between the values of the R, G, and B types of multi-valued image data for each pixel. The specific color pixel extraction unit to be determined is determined by the pixel unit. The effect of the color list described later is removed from the determination result. This is the ghost removal unit.

Each of the signal lines 101, 102, and 103 in the figure transmits multi-valued data of R, G, and B sent from the image input unit. Reference numerals 221, 222, and 223 denote FB / W and FLGR, which are determination results. , FLGB, and 501 transmits image data obtained by binarizing luminance data. The judgment results FLGR and FLGBi output from the signal lines 211 and 212 indicate whether or not the hue of the pixel is within a predetermined range. However, here, for example, among the pixels determined to be “red”, even if the hue itself is in the category of ¹¹ red, it is binarized due to the high luminance. Therefore, the gates 241 and 242 are used to remove the pixels PLGR ₂ and FLGB ₂ that indicate only those pixels whose binary image data is “1”. 4 and outputs it to the 4 0. configuration Gore one be sampled removing unit 4 4 0, later for details described but, FLGR _2, FLGb _2, and inputs binary image data, "red" Oyo Pi "blue" Output the identifiers FLGR 222 and FLGB 223.

Next, an example of the determination used in the specific color pixel extraction unit 210 will be described with reference to FIG. FIGS. 25A to 25C each show a color space expressed by multi-valued image data. The colors represented by the R, G, and B types of multi-valued image data can be represented as the coordinates in the cube shown in Fig. 25A. If C l, C 2, and C 3 in the figure are the values of R, G, and B, respectively, P 1 represents black and P 8 represents white. Here, when R, G, and B are each 8-bit data, the range of CI, C2, and C3 is from 0 to 255, and the maximum value Lmax is 255. P 2, P 3, and P 4 indicate the three primary colors red, green, and blue, and P 5, P 6, and P 7 indicate the complementary colors, yellow (yellow) and magenta (sky blue). , And cyan (purple) The extraction of a specific color is performed by expanding the input image for each pixel to the coordinates in this space, and belonging to a part of a predetermined area. This can be achieved by extracting pixels.

 Two examples of the "red" and "blue" ranges are shown in Figure 25B and Figure 25C. In the case of Fig. 25B, FLGRi and FLGBri force S are output according to the following equations.

 FLGR ι = 1: R ≥ G X L 1 Lmax

 and R ≥ B X L 1 / Lmax

 0: R <G X L 1 / Lmax

 o r R <B X L 1 / L max

 FLGB! = 1: B ^ G X L l / Lmax

 and B ^ R X L l / Lmax

 0: B <G X L L max

 o r B <R X L 1 / Lmax

On the other hand, in Fig. 25C, FLGR 1 and FLGB ₂ are output according to the following equations.

FLGR 1 = 1: R 1 ² ≤ (Lmax-R) ² + G ² + B ²

0: RI ² > (Lmax-R) ² + G ² + B ²

FLGB! = 1: R 2 ² ≤ R ² + G ² + (Lmax-B) ²

0: R 2 ² > R ² + G ² + (Lmax-B) ²

The specific color pixel extraction unit 210 that realizes the above input / output relationship can be realized by a numerical calculator or a memory. memory In this case, a high processing speed can be obtained, but an enormous amount of memory is required depending on the gradation of the input image data. In this case, the amount of memory can be reduced by performing the determination using only the upper data of the input data. It is also possible to use both judgment by memory and interpolation by numerical operation.

 FIG. 26 shows an example of memory contents when the specific color pixel extracting section 210 is realized by using memories. In this example, each of the input R, G, and B data has a gradation of 4 bits, and represents the relationship between R, G, and FLGR when B is constant. The range of colors to be extracted can be reset by rewriting the content of the memory used as the specific color pixel extraction unit 210.

 However, in general, when a human specifies a specific color, it is difficult to directly specify a desired color range in the RGB space. Therefore, when the hue range to be extracted is arbitrarily specified, the operability can be improved by specifying the color using a color expression system that is close to human feeling.

 Figures 27A to 27C show the principle of judgment using the VCH system as an example of the color system. The VCH system expresses a color using luminance V, saturation C, and hue H. The conversion from the RGB system to the VCH system is performed, for example, by the following equation.

 V = 0.299 R + 0.287 G + 0.114 B

 C = V V {C R-V) + (B-V)}

H = tan- ¹ {(R-V) / (B-V)} Here, the saturation is a threshold value for distinguishing a color from a monochrome, and the hues 0 Ϊ and 0 2 are a range of hues representing red, 0 ₃ and

0 ₄ shows the range of the hue representing blue. Saturation threshold C i and the color phase threshold 0 3 this, theta 2, theta _3, 0 ₄ Ri by the and this for controlling the color of pixels to be extracted Ru can and this for any setting.

 This color system is widely used to represent colors in printed matter, and the standards for vermilion and stamp ink are also expressed in this system. Therefore, in systems that target documents, external colors are specified in this color system, and internal processing is performed in the same RGB system as image data. A system that allows easy specification can be realized.

 In order to realize this function, it is necessary to provide a means for converting the color range specified by the VCH system into the RGB system.

 FIG. 28 shows a configuration example of the specific color pixel extraction unit 210 for an image filing apparatus having a function of arbitrarily setting a range of colors to be extracted. In the figure, 310 is a parameter input section for specifying the range of colors to be extracted from outside, and 320 is a part of the rain element determination section 240 extracted from the input parameters. A parameter setting unit to be calculated, 330 is a control unit for controlling the writing of data from the parameter setting unit to the extracted pixel determination unit 240.

The specific color pixel extraction section 210 includes an extraction surface element determination section 240, a saturation calculation section 220, and a hue calculation section 230. The saturation judgment section 220 and the hue judgment section 230 each have three types of RGB. .

54

It is a converter that inputs multi-valued data and outputs multi-valued saturation data C and hue data H.

 The conversion from the RGB system to the VCH system can be performed by the arithmetic processing using the above-described equation, but can be more easily realized by using the memory. However, when the number of gradations of the input RGB data is large, the scale of the memory becomes enormous. In that case, enter only the upper bits of the data. Furthermore, in order not to reduce the number of gradations, it is possible to interpolate the lower bits by calculation.

When the RGB data of the pixel P (X, y) is input, the output of the saturation extraction unit 220 is C (x, y), and the output of the hue calculation unit 230 is H (x, y). The extraction pixel determination unit 240 makes a determination by comparing the saturation C (x, y) and the hue H (X, y) with respective thresholds C 0 ₂ , 6 ₃ , and 0 4. Result FLGR 1 and FLGB 1 are determined and output according to the following formula, for example.

 FLGR 1 = C (x, y) ≥ C 1

and Θ i≤ H (x, y) ≤ θ ₂

 FLGB 1 = C (χ, y) ≥ C 1

and 0 ₃ ≤ H (x, y) ≤ Q _Λ

The range of the pixels to be extracted is set by the parameter input unit 310 using the threshold C! , Θ! , 0 ₂ , Θ 3, G a Based on the input C], 0 1, 0 ₂ , Θ 3, 04, the memory contents to be recorded in the extracted pixel determination section 240 are calculated in the noram setting section 320, and the control section is operated. Under the control of 330, it is transferred to the extracted rain element determination unit 240. As a result, the pixels to be extracted can be set arbitrarily. And can be.

 In this case as well, similar to the above-mentioned judgment based on the RGB data, by performing the AND operation with the luminance binarization processing result FB / W, unnecessary pixel extraction can be avoided. You. As a result,

— FLGR ₂ and FLGB ₂ are input to the test strip 400 as the determination result.

On the other hand, in the parameter input section 310, five types of thresholds can be set freely by the operator, so it is important to express each threshold in a form that is close to the operator's feeling. FIG. 29 shows an example of the parameter input section 310. In the figure, 3 1 1 is the saturation threshold C! This is a slide switch for setting. Threshold C! Can be set independently. On the other hand, 312 is a dial for inputting the center of the hue range set as red, and 313 'is a slide switch for inputting the width Hbi of the hue to be extracted. Threshold 0] and 6 _2, Ri by the H] and H 1, is determined by the following equation.

 .Θ 1 = H i-H b!

 0 2 = H 1 + H b 1

Similarly, the blue range is set with the dial 314 and the slide switch 315. Also, if the specified two hues are ffl duplicated, such as 0 ₂ ≥ 0 ₃ , take measures such as setting the center point as a boundary or giving priority to one as necessary. . Further, by specifying 0 so as to include all hues in the specification of hue, the specific color identification section 450 can be used as the color / monochrome identification section 4100 described above. And the power s it can.

 Subsequently, the configuration of the gore removing unit 450 will be described in detail. Color ghosting is a phenomenon in which colors that do not exist in the judgment result are generated due to misalignment between RGB in the input device and differences in photoelectric conversion characteristics. For example, a pixel in the outline of a black character in a white background is determined to be red or blue.

 Ghost elimination can be realized by correcting the judgment result for the center pixel by the pattern of the judgment result for the neighboring pixels. For example, a ghost that occurs between consecutive white and black pixels can be corrected by looking at the determination results for each of the two pixels, top, bottom, left, and right.

— As an example, a case where the determination result of each of the top, bottom, left and right 2_pixels is referred to will be described. At this time, the pixel to be referred to by 9 pixels shown in 3 0 Figure, use record, Ru data becomes three _{FLGR 2, FLGB 2, FB /} W. Here, reference numeral 330 denotes a pixel for outputting a final judgment.

FIG. 31 shows an example of the configuration of the guest elimination section 4400. In the figure, 450, 460, 470, respectively, FLGR _2, FLGb _z, and temporarily stores the FB / W, data upon storage unit to simultaneously output force nine pixels Dzu' necessary determination, 480 is input 2 A ghost rewriting processor that determines the final judgment result from 7 bi-volumes of binary data.

 Here, 450, 460, and 470 can be realized with the same configuration. No.

3 2 shows an example of the configuration of the data storage unit 450. Figure Among them, 219 is a signal line for inputting FLGR ₂ from the specific color pixel extraction unit 210, 401, 411, 421, 431, 441 and 422, 423, 424, 425 are latches, 405, 415, 435, 445 are 415 Shift registers 410, 420, 430, and 440 are used to align the timing of input data. Line memory stores data for one scan line of the input image. It is. Now, when the signal FLGR 2 (x + 2, y + 3) is input from the signal line 219, the latches 401, 411, 431, and 441 respectively output FLGR 2 (x + 2, y + 2), FLGR ₂ (χ + 1, y + 2), FLGR 2 (χ-1, y + 2), FLGR ₂ (χ-2, y + 2) are output, and 421, 422, 423, 424, 425 Are FLGR ₂ (x + 2,), FLGR ₂ (x + 1,), FLGR ₂ (x-1, y), FLGR ₂ (x, y), FLGR ₂ (x-2, power ^s output From shift registers 405, 415, 435, and 445, FLGR ₂ (X, y + 2), FLGR ₂ (x, y + 1), FLGR ₂ (x, y-1), and FLGR 2 ₂ (x, y-2) is output.

 By having three similar circuits, FLGBa and FB / W can also be output simultaneously for nine pixels.

 The rewriting processing unit 480 in FIG. 31 determines the judgment results FLGR and FLGB for the center pixel based on the input binary data of 27 bit. When the number of pixels to be referred to is small, the ghost rewriting unit 480 can be realized by memory, but here, an example in which the ghost is removed by a logic circuit is shown. .

Now, if the rewriting condition is defined as (red of width i pixels, sandwiched between two or more white pixels and two or more black pixels), Four types of patterns shown in Figure D Are targeted. In the present embodiment, only the case where a pixel of FLGR ₂ = 1 is input and FLGR = 0 ¾r is output is described as an example.However, when FLGB ₂ = 0 is input in a similar configuration, FLGB ₂ = 0 is input. Outputting = 1 can also be handled. Also, when specifying extracted hue, and theta _2, also properly the theta ₃ 0 ₄ Dochi et al or in one, Do set the hue to be extracted cormorants by including all hues mules, This guest elimination unit can be applied as it is to the above-described color / monochrome mouth identification.

 Next, confirmation of the determination result will be described. If a display device that can display a color image, such as a color CRT, is connected to the system, displaying the accumulated color binary image allows the judgment result to be confirmed. You. However, even if the connected display device is black and white, it is possible to confirm the judgment result by blinking the extracted portion (blinking).

FIG. 34 shows an example of the configuration of a device for checking the judgment result on a black-and-white display unit. Here, for the sake of simplicity, an example will be described in which only the portion determined to be red blinks. In the figure, reference numeral 800 denotes an image conversion unit for creating a one-screen display screen from three pieces of binary image data, reference numeral 810 denotes a blink controlling blinking timing, and reference numeral 820 denotes a blinking control. A logic element for obtaining exclusive OR, 830 is a selector for selecting two types of input data. The input data is the luminance data and red plane data stored in the memories 610 and 620. At the time of image display, by selecting the selector 710, binary data of luminance is output from the signal line 711, and the above-mentioned binary data is output from the signal line 821. An image excluding the part judged to be red is output. The selector 830 switches between these two types of binary image data under the control of the clock 810, and displays them alternately.

 In the present embodiment, a case has been described in which a pixel represented in a specific color is extracted from a document image in order to represent and store a color document in black and one or two specific colors. However, although it is not actually represented by a specific color, the part of the image that should be recorded as a specific color can be similarly specified by specifying from the outside. Can be efficiently accumulated. In this case, the region to be extracted as a specific color in the image is designated from the outside, so that the specific color extraction unit can be omitted.

 FIG. 35 shows a configuration example of the image filing apparatus in this case. In the figure, 670 is a known coordinate designation device such as a mouse, 650 is an area designation section that reads the coordinates of the area designated by the coordinate designation device 670, and 660 is an area designation Image extraction unit that reads out image data in the area specified by the unit 650 from the image memory 610 and transfers it to the image memory 620 or 630. It is.

The document is input as multi-valued image data from the image input unit 100, binarized, recorded in the image memory 610, and simultaneously displayed on the image display unit 750. You. The operator can use the image display 7 Specify an arbitrary area with the mouse from the image displayed on the screen and the position of the mouse. The coordinates of the area specified by the mouse are read out by the area specifying unit 650. Here, the image clipping unit 660 reads the content of the brightness image data corresponding to the specified area from the image memory 610, and the image memory 620 Or Copy to image memory 630. As a result, it is possible to record a color document image by performing a manual operation without using the specific color extraction function.

 In the present embodiment, by storing the coordinates of the specified area, it is possible to successively input each document in the once specified area when continuously inputting unified format documents. It is possible. .

 Even when a black-and-white document image is input, it is possible to store the specified portion as a color lane by specifying a portion to be colored with a specific color.

Claims

The scope of the claims

1. An input means for inputting an image corresponding to a document as multi-valued data;

 Input data conversion means for converting the multi-valued data into binary data,

 Area determination means having at least a color identification means for extracting a pixel area to be displayed with a specific color included in the image from the multi-valued data;

 Binarizing means for binarizing the luminance information of the image from the multi-valued data, including a pixel area to be displayed in the specific color, and

 Selection means for selecting the output of the binarization means in accordance with the output of the area determination means

Input data conversion means comprising:

 An image memory means for temporarily storing the binary data,

 A first memory for storing the binarized luminance information (Y; Gc / G (i)) including a pixel area to be displayed in the specific color as a first plane; Means, and

 Second memory means for storing the binarized luminance information of the pixel area to be displayed in the specific color as a second plane

Image memory means comprising:

Sign the binary data stored in the image memory means Image storage means for decoding the stored data and outputting the decoded data to the image memory means;

 Output data conversion means for outputting desired image data based on the binary image data of the image memory means;

 Image display means for displaying a desired image based on the image data from the output data conversion means;

Image filing device equipped with

 2. The binarization means includes a binarization processing means for binarizing the multi-valued data with a predetermined threshold value and a dither processing means for pseudo-multiplying the multi-valued data (hereinafter, intermediate processing to binarize). The image filing apparatus according to claim 1, wherein the image filing apparatus has:

 3. The area determining means has a character Z photograph area determining means for determining whether each part of the surface image belongs to a character area or a photograph area,

 The selection means stores one of the output of the binarization processing means and the output of the dither processing means as the luminance information stored as the first plane in the character / photograph area determination means. Select according to output,

3. An image filing apparatus according to claim 2, wherein:

4. The selection means includes, as the luminance information to be stored as the second plane, the binarization processing means for a pixel area to be displayed in the specific color. One of the output of the dither processing means and the output of the dither processing means is selected in accordance with the output of the character Z photograph area determination means. Judgment means output Select "0" regardless of power,

4. The image filing apparatus according to claim 3, wherein:

 5. The color identification means is

 G, R, and B types of multi-valued data are input as the multi-valued data, and it is determined whether or not the pixel is a pixel to be displayed in the specific color based on the relative relationship between the values. Specific color pixel extracting means,

 For each pixel, rewrite the above determination result for the pixel from the combination of the determination result of the specific color pixel extraction means for the neighboring pixels and the binary luminance information obtained from the above-mentioned binarization means. The image filing apparatus according to any one of claims 1 to 4, further comprising: a stall removing means.

 6. The above color identification means

 Parameter input means for externally specifying the specific color and its hue range,

 The color specified by the above parameter input means

 The image filing apparatus according to claim 5, further comprising: a parameter setting unit configured to convert the image into an RGB 3 primary color system.

 7. The specific color pixel extracting means is as follows:

 Saturation calculating means for calculating the saturation from the three types of multi-valued data of G, R, and B;

Hue calculation means for calculating the hue from the three types of multi-valued data of G, R, and B; Extracted pixel determining means for comparing the calculated saturation and hue with the output of the parameter setting means and determining whether or not the pixel is to be displayed in the specific color.

7. The image filing apparatus according to claim 6, comprising:

 8. The image storage means has a data storage device for storing the data of the first and second planes in the order in which the data of the first plane can be read first when reading. The image filing apparatus according to any one of claims 1 to 7, wherein:

 9. The image filing according to any one of claims 1 to 8, wherein the image display means displays the binary data of the first plane as a black and white image. apparatus.

 10. The input device according to claim 1, wherein the input means comprises a color scanner for inputting an image corresponding to the document as three types of multi-value data of GR and B. 10. The image filing device described in the above item.

 11. The image filing apparatus according to claim 10, wherein, as said luminance information, a binarization processing output of said binarization means for multi-value data of G is used.

 12. The area determining means determines whether each pixel of the input image is a monochrome area described in black and white or a color area including color information. It has a nozzle identification means.

The image memory means further has a third memory means. And

 The above-mentioned selection means provides the above-described G, R, and B three-valued multi-valued data to the above-mentioned binarization means for a pixel determined as a color area by the above-mentioned color / monochrome identification means. Is selected by the pseudo-intermediate processing, and is output to the first, second and third memory means.

 10. The image filing device according to claim 10, wherein:

13. The selection means determines whether the pixel belongs to a character area or a photograph area for a pixel determined to be a monaural opening area by the color / monochrome identification means. Judgment: For the pixels belonging to the photographic area, select the dither processing output obtained by binarizing the above three types of multi-valued data by the pseudo-intermediate processing by the above-mentioned binarizing means. Output to 2nd and 3rd memory means

The image filing apparatus according to claim 12, characterized in that:

14. The selecting means sets the multi-level data of G to a predetermined threshold value for the pixels belonging to the character area by the binarizing means for a full-color document and an image specified in the mode. And outputs the same value to the first, second and _third memory means.

14. The image filing apparatus according to claim 13, wherein:

15 5. The above-mentioned selection means is divided into a monochromatic area and a color area. In the case of a document in which characters are mixed and an image whose mode is specified, the binary data is binarized by the above-mentioned binarization means for the pixels belonging to the above-mentioned character area using a predetermined threshold. And outputs “◦” to the second and third memory means while outputting to the first memory means.

14. The image filing device according to claim 13, wherein:

16. The image memory means described above uses the color Z monok mouth identification means for the document in which both the monochrome area and the color area are mixed and the mode-specified image. 16. The image filing apparatus according to claim 15, further comprising a fourth memory means for storing the information.

 1 7. The output data conversion means

 A selector for selecting desired data from the binary data stored in the image memory means,

 Multi-value processing means for generating multi-value image data for each color from the data of the selectors

The plane image filing apparatus according to any one of claims 1 to 16, characterized by having:

 18. The image display means combines the multi-value image data for each color generated by the multi-value processing means and displays the multi-value image data. Term

17. The image filing device according to item 7.

1 9. The multi-value processing means is

 Means for determining whether the pixel belongs to a character area or a photograph area based on the binary data from the selector;

 Means for generating multi-valued image data from the binary data from the selector,

 Means for shifting the binary data from the selector,

 Means for selecting one of the output of the multi-value data generation means and the output of the shift means based on the result of the judgment by the judgment means

 The image filing apparatus according to claim 17, wherein the image filing apparatus has: .

 20. The above output data conversion means,

 The binary luminance information stored as the first plane in the image memory means and the pixel area to be displayed in the specific color stored as the second plane are stored in the image memory means. Means for calculating an exclusive OR of the binary luminance information and

 Means for alternately selecting the calculated binary image and the binary image of the luminance information

The image filing apparatus according to any one of claims 1 to 16, characterized by having:

2 1. Inputting the image corresponding to the document as multi-valued data. A step of converting the multi-valued data into binary data,

 An area determination step including at least a color identification step for extracting a pixel area to be displayed in a specific color included in the image from the multi-valued data,

 A step of binarizing the luminance information of the image from the multi-valued data, including a pixel area to be displayed in the specific color, and

 A step of selecting the output of the binarizing means in accordance with the result of the region extraction

A data conversion step consisting of

 The binarized luminance information including the pixel area to be displayed in the specific color is defined as a first plane, and the binarized luminance in the pixel area to be displayed in the specific color is set as the first plane. A step of temporarily storing the information as a second plane in the image memory means,

 Encoding the binary data stored in the image memory means and storing the encoded data in a data storage device;

Image filing method with

 2 2. a step of decoding the data stored in the data storage device and outputting it to the image memory means;

 Converting the binary image data of the image memory means into desired image data and outputting the image data; and

A step of displaying a desired image based on the desired image data 22. The image filing method according to claim 21, comprising:

 23. The binarization step consists of a simple binarization processing step of binarizing the multi-valued data with a predetermined threshold and a binarization by intermediately processing the multi-valued data. The image filing method according to claim 21, comprising a dither processing step for performing dithering.

 24. The area extraction step includes a character / photo area determination step for determining whether each part of the image belongs to a text area or a photo area.

 In the selection step, one of the simple binarization processing result and the dither processing result is used as the luminance information stored as the first plane as the character / photograph area determination result. Select according to.

 The image filing method according to claim 23, characterized in that:

 25. The selection step is the luminance information stored as the second plane, and the simple binary value for the territory of the pixel to be displayed in the specific color is used as the luminance information. One of the digitization processing result and the dither processing result is selected according to the character / photograph area determination result, and the area other than rain element to be displayed in the specific color is determined by the character / photograph area determination result. Select "0"

The image filing method according to claim 24, characterized by this.

2 6. The above color identification steps G, R, and B three types of multi-valued data are input as the multi-valued data, and a determination is made based on the relative relationship between the values to determine whether the pixel is a pixel to be displayed in the specific color. Color pixel extraction step and

 For each pixel, rewrite the above-mentioned judgment result for the relevant pixel from the combination of the above-mentioned specific color pixel extraction result for the neighboring pixel and the above-mentioned binary luminance information obtained in the above-mentioned binarization step. Step

The image filing method according to any one of claims 21 to 25, comprising:

 27. The above color identification steps are

 A parameter input step for externally specifying the specific color and its hue range,

 27. The image filing method according to claim 26, further comprising a parameter setting step for converting a color specified in the parameter input step into an RGB 3 primary color system.

 2 8. The above specific color pixel extraction step

 A godsend operation step for calculating the saturation from the three types of multi-valued data of G, R, and B;

 A hue calculation step for calculating the hue from the three types of multi-value data of G, R, and B;

An extracted pixel determination step of comparing the calculated saturation and hue with the result of the parameter setting step to determine whether or not the pixel should be displayed in the specific color. 28. The image filing method according to claim 27, comprising:

 29. The image storage steps are stored in the data storage device in the order in which the data of the first plane can be read first when read out. 30. The image filing method according to claim 21, wherein data of the image is stored.

 30. The image according to any one of claims 21 to 29, wherein the image display step displays the binary data of the first plane as a black and white image. The firing method.

 31. The input step is characterized in that an image corresponding to the document is input as three types of multi-valued data of G-, R, and B by a color scanner. 21. The image filing method according to any one of claims 21 to 30, wherein

 32. The result of the binarization processing on the G serial data is used as the luminance information.

31 The image filtering method described in 1.

 33. The above-mentioned area determination step determines whether each pixel of the input image is a black-and-white area or a color area including color information. Has a color identification step,

The selection step is performed on the G, R, and Β pixels in the binarization step for a pixel determined to be a color area based on the result of the color / monochrome port identification. Suspect three types of multi-valued data The dither processing result binarized by the similar intermediate processing is selected and output to the second, third, and third memory means in the plane image memory means.

31. The image filing method according to claim 31, wherein:

 3 4. The above selection step is performed for a pixel determined to be a monochrome area by the above-mentioned color identification area, the pixel is a character area and a photograph area. Of the G, R, and Β3 types of multi-valued data are binarized by pseudo intermediate processing in the binarization step for the pixels belonging to the photographic area. The image processing result is selected and output to the first, second, and third memory means in the image memory means.

 34. The image filing method according to claim 33, wherein:

 35. The above selection step is for the full color one document and the image specified in the mode. A simple binarization processing result binarized by a predetermined threshold value is selected and the same value is output to the first, second and third memory means in the image memory means.

 The image filing method according to claim 34, characterized by this.

36. The above selection step is used for documents in which both the monochrome area and the color area are mixed, and the mode-specified image. For the pixels belonging to the character area, select the simple binarization processing result obtained by binarizing the multi-valued data of G with the predetermined threshold in the binarization step and select the first binarization result. Output "0" to the second and third memory means while outputting to the memory means.

The image filing method according to claim 34, characterized by this.

 37. For a document in which both a monochrome area and a color area are mixed and an image specified in the mode, the identification result of the above color identification step is displayed in the above image memo. 37. The image filing method according to claim 36, wherein the image is stored in a fourth memory means in the memory means.