JPH03263277A - Picture processor - Google Patents

Abstract

PURPOSE:To make correspondence to a color document or drawing with simple processing configuration by separately extracting a picture and a character/graphic picture by learning a mark to be directly read after write to the document/drawing and the color characteristic of the background, character, graphic and picture in the document/drawing before reading. CONSTITUTION:Scanning is successively executed with a horizontal direction as a main scanning direction and a vertical direction as a sub scanning direction. When a mode switch part 2 is set to a learning mode, a color characteristic registration part 3 learns and registers the color characteristic corresponding to the color and density of a mark, which is directly read after write to the original, and the color characteristic of the background and document. The coordinate data of a registering area set to the registration part 3 is compared with the coordinates of a scanning point and when this scanning point is positioned in the registering area, the color characteristic under registration is updated from an RGB system to an XYZ system. On the other hand, in the case of an original read mode, the color characteristic of the scanning point is compared with the registered color characteristic. When the color characteristic of the scanning point is positioned in the mark condition range, '1' and '0' are respectively written into a mark picture storage part 5 and a document/ drawing picture storage part 6, and when the scanning point is positioned in the document, graphic and picture condition range, '0' and '1' are respectively written into these storage parts. When the scanning point is positioned in the background condition range, '0' and '0' are respectively written into those storage parts.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention reads a manuscript in which area extraction marks and editing marks in a color different from that of the document 2 drawing are added, and extracts and edits the necessary area of the document 1 drawing. The present invention relates to an image processing device, and particularly to an image processing device that separates and extracts area extraction marks and editing marks of different colors from document 2 drawings.

[Prior art]

Conventionally, area extraction marks and editing marks (hereinafter referred to as '
Scan the manuscript with additional color marks (called "color marks"),
Methods for separating and extracting the color mark and document 1 drawing include the following methods (1) to (3).

(1) The first method is disclosed in Japanese Patent Application Laid-Open No. 60-31682, in which the output of an image input device such as a monochrome scan is binarized at at least two different levels: a lower level and an upper level. , By applying exclusive OR processing to both data obtained by this, a color mark image and a document are created.

This is a method of obtaining drawing images.

(2) The second method is disclosed in Japanese Patent Application Laid-open No. 63-40981, in which a document is irradiated with a white light source, and the reflected light is separated into several types of colored light using a spectral filter. The shutters are operated in synchronization with each other and whose opening/closing frequency is an integer multiple, and each color light that has passed through these shutters is refocused, and the light receiving sensor converts the focused light photoelectrically, and the resulting electrical signal is converted into a frequency The e! ! This is a method of separating and extracting the It surface and the color mark.

(3) The third method uses the output of an image input device such as a full-color scanner that is separated into the three primary colors of red, green, and blue.
This is a method of separating and extracting the color mark and document 2 drawing by calculating between red, green, and blue outputs.

[Problem to be solved by the invention]

However, in the first method (1) above, in order to support various types of color pens, the signal level range of the color mark part becomes wide, so it is difficult to set the threshold corresponding to the color mark. Even if you can set it, the minus part of the document 9 drawing image (black) and background part (white) may be mixed in the separated color mark image, so limit the color pens or use stricter filling methods. The problem was that it was necessary.

Further, the second method (2) above has the problem that the color types of color marks that can be added are limited by the spectral filter, and the structure of the image input unit and the separation/extraction process become complicated.

In addition, the third method in <3) above uses the output of the three primary colors of red, green, and blue to determine color characteristics such as lightness, two hues, and saturation. However, the existing devices are designed to process full-color natural images, and can be used to separate document 1 drawings from originals with simple marks added to document 3 drawing images. Image processing devices for separating and extracting mark images have problems in that the processing and device configurations are complicated and expensive, and the operability is poor.

In addition, if the color of the document 2 drawing and the background are not black or white, respectively, there is no means to specify them and they cannot be separated, or the writing conditions of the manuscript can be changed by interactive processing with the device or computer using a keyboard etc. There was a problem in that it was necessary to input such information, making the operation complicated.

The present invention has been made in view of the above-mentioned problems, and in order to eliminate the above-mentioned problems, the present invention learns the color characteristics of the color mark added to the document 2 drawing and/or the background 1 document 1 drawing before reading the original. , the mark image and the document 2 drawing image are separated and extracted based on the learned color characteristics, and with simple processing and configuration, the restrictions on the color type of the color pen used for the mark to be added are removed, and the color An object of the present invention is to provide an image processing device with excellent operability that can handle one drawing of a document.

[Means to solve the problem]

In order to solve the above problems, the present invention provides a document/drawing and the document/drawing.
An image processing device that reads a document containing marks of a different color from the drawings and separates/extracts the documents/drawings and the marks was configured as follows.

Marks and/or letters 9 shapes written on the study paper,
an image input section that outputs color information according to the color and shading of the background; a mode switching section that switches between learning mode and document reading mode; and when the mode switching section is set to learning mode, a color characteristic registration section that registers color information of a predetermined area of the color characteristic as a color characteristic, and when the mode switching section is set to document reading mode, the color information output from the image input section and the color characteristic registration section. An image classification unit is provided that separates and extracts characters and figures on the manuscript based on the color characteristics registered in the image classification unit.

In addition, there is an image input section that outputs color information according to the color and shade of characters, figures, and/or marks written on the manuscript, and a color characteristic registration section that registers color information of a predetermined area of the manuscript as color characteristics. Separate and extract a character 1 figure and a mark on the document based on the color information of the image input section output from an area other than the predetermined area of the document and the color characteristics registered in the color characteristic registration section. An image classification section is also provided.

Further, the color characteristic registration unit divides the first predetermined area of the study paper or the predetermined area of the manuscript into an area for specifying the color characteristics of the background and an area for specifying the color characteristics of the characters 9 shapes. ,
The mark is divided into an area for specifying the color characteristics of the mark to be added, and the color information of the divided area is registered as the color characteristics.

Furthermore, the images of the characters 9 and the background output from the image classification section were images with color information output from the image input section.

Further, the image of the mark portion outputted from the image classification section is accompanied by data to which the color characteristics of the mark are added.

[Function] In the present invention, when the mode switching section is set to the document reading mode, based on the color information output from the image input section and the color characteristics registered in the color characteristic registration section, Either by separating and extracting characters and figures on the document, or by separating and extracting characters and figures on the document, or by separating and extracting the characters and figures on the document based on the color information from the image input section output from an area other than the predetermined area of the document and the color characteristics registered in the color characteristic registration section. By separating and extracting characters 1 figures and marks, it is not only possible to remove restrictions on the color of the marks used and the paper size of the manuscript, but also to remove multimedia documents such as documents including photographs, 9 drawings, and color document drawings.

Can be adapted to drawings.

〔Example〕

Hereinafter, one embodiment of the present invention will be described based on the drawings.

FIG. 1 is a block diagram showing the configuration of an image processing apparatus according to a first embodiment of the present invention. This image processing apparatus is composed of an image input section 1, a mode switching section 2, a color characteristic registration section 3, an image classification section 4, a mark image storage section 5, and a document 1 document 9 drawing image storage section 6.

The image input unit 1 converts color information corresponding to the color and shading of a mark written on a sheet (not shown) and/or a drawing of a document 2 into a multi-value level signal of the three primary colors of red, green, and blue, for example, to a mode switching unit. Output to 2.

Mode switching unit 2 The learning mode or document reading mode is set using a switch (not shown) or the like. When the mode switching unit 2 is set to the learning mode, the output of the image input unit 1 is output to the color characteristic registration unit 3, and
A registration activation signal for starting the processing of the color characteristic registration section 3 is output. Further, when the mode switching section 2 is set to the original reading mode, the output of the image input section 1 is output to the image classification section 4, and a classification start signal is outputted to start the processing of the image classification section 4. .

The color characteristic registration section 3 starts operating in response to a registration activation signal from the mode switching section 2, selects only color information in a second predetermined area of the input paper, and registers the color information as a color characteristic. do. The color characteristics to be registered may be registered as they are as multi-level signals (RGB system) of the three primary colors of red (R), green (G), and blue (B), or other known color characteristics. You can also convert it to and register it.

In the present invention, the RGB system is converted to the XYz system using the coordinate conversion formula (1) below, and the following is obtained: X −2,7689R+ 1.7517 G + 1.
1302 BY=R+4.5907(1,+0
．． 0601BZ-0,0565G+5.5943B
(t) Calculate x, y using formula (2) based on the obtained x, y, z, x = X / (X + Y + Z) y = Y / (X + Y + Z) (2) Calculate x, y, and Y are registered as color characteristics.

Here, X, Y, and Z are called tristimulus values as defined by JIS. Furthermore, as is well known, the x,
y represents hue and saturation, and Y represents lightness.

The conversion from the RGB system to the xYz system [Equation (1)] and the calculation to obtain x and y representing hue and saturation [Equation (2)] are performed by inputting R, G, and B, respectively, and calculating x, y. , z, the computation time becomes unnecessary and the processing speed can be increased.

Next, an example of color characteristic registration will be described.

The third memory is a two-dimensional memory whose addresses are x and y converted into integers.
By constructing a chromaticity diagram as shown in Figure (a) and setting one point in the two-dimensional memory determined by the obtained integer x t 3' as 1, the color characteristics corresponding to hue and saturation can be calculated. register. The initial value of the two-dimensional memory is 0''.The brightness Y is registered by updating its minimum and maximum values for each scanning point (pixel).

The image classification section 4 starts operating in response to a classification activation signal from the mode switching section 2, and selects the color characteristics calculated from the color information output from the image input section 1 and the colors registered in the color characteristic registration section 3. Based on the characteristics, the document 2 drawing and mark on the manuscript paper are separated and output to the document 9 drawing image storage section 6 and mark image storage section 5, respectively.

Note that whether the image data to be output is image data with color information or binary image data is determined depending on the purpose of the image processing apparatus.

The mark image storage unit 5 records mark images in data format output from the image classification unit 4.

The document 1 drawing image storage section 6 stores the document 2 drawing image in the data format output from the image classification section 4.

FIG. 2 is an explanatory diagram of the study paper, which is configured so that only the color characteristics of marks to be added can be entered in the color characteristics registration area. In the same figure, ml is the color characteristic registration area, and yV r
A rectangular area with one point at the position HX
Fill in the rectangular area with the same color pen as the mark to be added to the document.

As is clear from the above explanation, the study paper in Figure 2 defines only the marks to be added, so for the input manuscript, the characters, figures, and photographs written in the document 2 drawing are in black. , was designed with the assumption that the color is gray and the background is white.

In addition, if the color characteristic registration area is the background color (white) of the study paper, it is defined that the document drawing of the manuscript is an achromatic color of black gray and white, the mark to be added is an achromatic color, and the mark to be added is a chromatic color. By doing so, it is also possible to remove restrictions on the color type of the color pen to be used.

Figure 3 shows the XY color information output from the image input section 1.
This is an explanatory diagram of the color characteristics converted to the z system, which simulates the color characteristics of a document with a white background, two characters, and a black or gray photograph, and a document with a red mark added to the IK side. Y, x, and y shown in the figure are expressed by the above-mentioned formula (1), (
2). It can be seen that even if the described mark cannot be distinguished from the document 2 drawing only by the brightness Y, which has many variations in shading, it can be distinguished on the chromaticity diagram in which y is represented by the vertical axis and X is represented by the horizontal axis.

Next, the processing of the image processing apparatus having the configuration shown in FIG. 1 will be explained using FIG. 4, FIG. 2, and FIG. 3150.

Regarding the coordinate system of the plane of the paper in the figure, to avoid confusion with X, Y, and Z representing color characteristics, the main scanning direction or horizontal direction is expressed as XX, and the sub-scanning direction is expressed as YY.

Step STI, Sr1 indicates that the paper is sequentially scanned with the horizontal direction (XX) as the main scanning direction and the vertical direction (YY) as the sub-scanning direction. In this explanation, the upper left end of the paper is the scanning starting point (origin (0, 0)), and the lower right corner is the scanning end point (x
, □t3's. , ), and the mark image storage unit 5 and the document 2 drawing image storage unit 6 are also configured with the same coordinate system and operate synchronously.

In step ST3, if the setting of the mode switching section 2 is learning mode, the color characteristic registration 87!3 is performed in step ST3.
Through the processing of I O and STI 1, the color characteristics corresponding to the color and shading of the mark to be added to the document and/or the color characteristics corresponding to the color and shading of the background and document 2 drawing are learned and registered.

First, in step STIO, it is determined whether or not the scanning point is within the color characteristic registration area by comparing the coordinate data indicating the color characteristic registration area set in advance in the color characteristic registration section 3 with the coordinates of the scanning point. Determine.

In the example shown in FIG. 2, it is determined that the scanning point is within the color characteristic registration area when the coordinate values XX, YY satisfy the following equation.

3'＋＋β≦YY≦3'＋”dy+−βx1+β≦XX
≦x 1+ d y 1-β 〈3) Here, β
is a constant determined based on the setting tolerance of the scanner and paper, mark writing conditions, etc.

In step 5TII, the RGB color information of the image input unit 1 input through the mode switching g2 is converted to the XYZ color information using the above-mentioned formula (1), and x, which is obtained using the formula (2), is Convert y to an integer and obtain the corresponding Y, x.

y is used to update the color characteristics being registered as described in the explanation of FIG.

In step ST3, if the setting of the mode switching section 2 is the original reading mode, the operation of the image classification section 4 is started.

In step STI 2, the color characteristic of the scanning point is compared with the color characteristic registered in the color characteristic registration section 3, and if the color characteristic of the scanning point is within the mark condition range, that is, the color characteristic of the scanning point is inputted via the mode switching section 2. The RGB system color information of the image input unit 1 is converted to the XYZ system using the above-mentioned formula (1), and the brightness range in which the Y value representing brightness is registered, for example, the mark shown in FIG. 3(b). When the brightness is within the brightness range of For example, if it is determined that the area is within the diagonally upward-left diagonal line shown in FIG. In 6, write r□, which represents the background part (step 5T13).If the document drawing requires an image with color information, write the color corresponding to the background part registered in the color characteristic registration part 3. The information is written into the document 9 drawing image storage section 6.

If it is necessary to classify mark images according to their color information, the color information output from the image input section 1 may be written directly into the mark image storage section 5, or the mark images may be classified into two categories. The color information or the color hood corresponding to the color information may be registered in a value image format in a mark table (not shown) provided separately from the mark image storage section 5. Note that it is preferable to write data in which the coordinates of any one point of the mark images to be connected, for example, the scanning point where the mark image is first detected, are paired with the color code in the mark table. .

In step 5T12, if the color characteristics of the scanning point are within the document 2 figure/photo condition range, "0" representing the background is written in the mark image storage section 5, and the document 2 drawing is written in the document 2 drawing image storage section 6. Write “1” to indicate (Step 5
T14).

Note that if a document drawing requires an image with color information, the color information output from the image input section 1 can be directly input to the document 1.
It is written into the drawing image storage section 6.

As for the mark image, the background usually has no meaning, so the explanation in step 5T14 is followed and the mark table mentioned above is not updated.

In step 5T12, if the color characteristics of the scanning point are within the background condition range, r□ representing the background is written in both the mark image storage section 5 and the document 2 drawing image storage section 6 (step 5T15). If an image requiring color information is required, the same processing as described above is performed.

In addition, in the process of step 5T12, if the color characteristic of the registered mark part is a chromatic color, and if the color characteristic of the scanning point is other than the diagonally lined mark area diagonally upward to the right in FIG. It is determined that there is.

FIG. 6 is a diagram showing another example of the study paper, and the same figure (a
) is an example of a study paper that can be defined for the color characteristics of the background, additional marks, and characters and figures. In the figure, mll is a description area where the color characteristics of the background part of the document can be defined, m12 is the description area where the color characteristics of the add mark mark of the document can be defined, and m13
is a description area in which characteristics of characters, drawings, and/or photographs of the manuscript can be defined. In this way, in the case of a study paper, all the written information on the manuscript, that is, the background of the manuscript, 1 document, 8 drawings.

Since it is possible to set color characteristics for marks, it is possible to handle documents such as one drawing containing a photograph and two drawings of multimedia documents such as color document drawings, and it is also possible to deal with poor quality paper such as newspapers. It is also possible to remove the pen and paper limitations.

It is also possible to separate the postscript mark on the colored transparent cover from the document 2 drawing inside the cover. FIG. 6(b) is a study paper with multiple registration areas for color characteristics of additional marks, where m14 is an area defining the color characteristics of the first mark to be added, and m15 is an area for defining the color characteristics of the first mark to be added. The area m16 that defines the color characteristics is an area that defines the color characteristics of the third mark to be added.

In the case of such a study paper, it is possible to classify and extract the postscript marks, so by assigning a meaning to each postscript mark, it is possible to expand the editing and area extraction processing functions.

Also, by changing the color of the registration area on the study paper,
It is possible to use the same manuscript for different editing and area extraction processes, and even if adding a mark fails, a different color will be used as the additional mark, and the registration area on the study paper will also be in the different color mentioned above. This makes it possible to deal with it. Color noise can also be removed for the same reason.

FIGS. 5(b) and 5(c) show a document 2 separated and extracted from the original shown in FIG. 5(a) using a learning paper that defines the color characteristics of only the additional marks shown as slanted marks that rise diagonally to the right. Shows drawing images and mark images.

FIG. 7 shows a mark image separated and extracted from the original shown in FIG. 5(a) using a study paper in which three types of postscript marks as shown in FIG. 6(b) are defined. The document 1 drawing image is the same as that shown in FIG. 5(b).

Note that the mode switching unit 2 can be removed by setting a part of the document as a learning area and learning the color characteristics of the registration mark area existing within the area.

Furthermore, the current state and size of the identification mark and color characteristic registration area for learning are not limited to the above-described forms.

〔Effect of the invention〕

As described above in detail, according to the present invention, the color characteristics of the mark to be added to the document 2 drawing, the document 9 drawing, and the background are learned before reading the document, and the mark image and the document are combined based on the learned color characteristics. Since the two drawing images are separated and extracted, the following excellent effects can be obtained.

(1) Not only can restrictions on the color of marks used and the paper size of manuscripts be removed, but also multimedia documents such as documents including photographs, drawings in color, and color documents.

It is possible to realize an image processing device with high versatility and excellent operability that can be applied to drawings with simple processing and configuration.

(2) The image processing device of the present invention is an OCR, a document reading device,
By implementing the present invention in an LA collection device, a FAX, an electronic whiteboard, etc., it becomes possible to make the device highly versatile and easy to operate.

[Brief Description of the Drawings] Figure 1 is a block diagram showing the configuration of an image processing device according to the first embodiment of the present invention, Figure 2 is an explanatory diagram showing an example of a study paper, and Figure 3 is a signal level diagram. (a) is an explanatory diagram of the chromaticity range of the information written in the manuscript, (b) is an explanatory diagram of the brightness range of the information written in the manuscript, and Fig. 4 is an illustration of the image processing of the present invention. FIG. 5 is an explanatory diagram of the image classification section. FIGS. 6(a) and (b) are diagrams showing other examples of learning sheets, and FIG. ) is a diagram showing a mark image separated and extracted from the manuscript shown in FIG. 5(a) using the study paper shown in FIG. In the figure, 1... image input section, 2... mode switching section, 3... color characteristic registration section, 4... image classification section, 5
. . . Mark image storage unit, 6 . . . Document 2 drawing image storage unit.

Claims (5)

[Claims] (1) In an image processing device that reads a manuscript in which a document/drawing and a mark of a different color from the document/drawing are written, and separates and extracts the document/drawing and the mark, the mark written on the study paper is used. and/or characters, figures,
an image input section that outputs color information according to the color and shading of the background; a mode switching section that switches between a learning mode and a document reading mode; and when the mode switching section is set to the learning mode, a color characteristic registration unit that registers color information of a predetermined area of 1 as a color characteristic; and when the mode switching unit is set to a document reading mode, the color information output from the image input unit and the color An image processing device comprising: an image classification section that separates and extracts characters and figures on a document based on color characteristics registered in a characteristic registration section. (2) In an image processing device that reads a manuscript in which a document/drawing and a mark of a different color from the document/drawing are written, and separates and extracts the document/drawing and the mark, the mark written on the document/drawing is read. an image input unit that outputs color information according to the color and shading of characters, figures, and/or marks; a color characteristic registration unit that registers color information of a predetermined area of the original as a color characteristic; an image classification unit that separates and extracts characters, figures, and marks on a document based on color information from the image input unit output from an area other than the area and color characteristics registered in the color characteristic registration unit; An image processing device characterized by being provided with. (3) The color characteristic registration unit specifies a first predetermined area of the study paper or a predetermined area of the manuscript as an area for specifying color characteristics of the background and color characteristics of characters and figures. Claim (1) characterized in that the mark is divided into an area for specifying the color characteristics of the mark to be added and an area for specifying the color characteristics of the mark to be added, and the color information of the divided area is registered as the color characteristics.
Or the image processing device described in (2). (4) The image information of characters, figures, and background output from the image classification section is image information accompanied by color information output from the image input section. ) or the image processing device described in (3). (5) The image of the mark portion outputted from the image classification section is characterized in that data to which color characteristics of the mark are added is included. 4) The image processing device described above.