JP2020178171A - Image processing apparatus, image processing method, program, recording medium, and image forming apparatus - Google Patents

Abstract

To provide an image processing apparatus that can determine the type of a document from image data and create image data with an appropriate number of output gradations based on the type of the document from the image data.SOLUTION: In an image forming apparatus 1, an image processing apparatus 10 comprises: an image input unit that receives input of image data of a document; a document type determination unit that determines whether the document is a monochrome document or a color document based on the image data, and determines the type of the document based on the image data; a feature amount calculation operation unit that calculates the feature amount from the image data; an output gradation number setting unit that, when the document is the monochrome document, sets any one gradation number from among a plurality of output gradation number candidates as an output gradation number based on the feature amount and the type of the document; and an image processing operation unit that creates image data having the number of gradations of the image data as the set number of output gradations.SELECTED DRAWING: Figure 1

Description

The present invention relates to an image processing apparatus and the like.

Conventionally, when transmitting an image printed on a paper surface, a fax or the like has been used by a user who desires to transmit the image. Furthermore, in recent years, with the spread of digital multifunction devices (MFPs (Multifunction Peripheral Printers)) and the Internet, paper is converted into electronic data format image data that can be read by a scanner, and electronic data format image data is attached by e-mail or the like. Transmission is also very routine.

Recently, MFPs equipped with an automatic sheet feeder (hereinafter abbreviated as "ADF") for transporting paper to a scanner portion for reading a document have become widespread. By using ADF, instead of placing the original on the scanner for each sheet of paper and scanning it, multiple originals (original set) are placed on the ADF and the scanning of each original is performed by ADF. Can be done automatically and continuously. In this way, an environment is provided to reduce the burden on the user in reading the document.

However, since the use of ADF saves the trouble of placing the manuscript each time, the user does not need to be aware of the contents of the manuscript set and puts the manuscript on the ADF in a state where the manuscripts having different properties are mixed in the manuscript set. It tends to be placed. For example, there may be a case where a color image original and a monochrome image original are mixed in the original set. Further, as another example, there may be a case where a text data manuscript and a photographic data manuscript are mixed in the manuscript set.

To solve this problem, it is determined for each original whether the original included in the original set is a color original or a monochrome original (Auto Color Select: hereinafter abbreviated as "ACS"), and the determination result of the ACS determination is made. A technique for performing processing based on the above has been proposed. Further, a technique for improving the accuracy of the ACS determination has also been proposed. For example, a technique for setting a plurality of thresholds for ACS determination (see, for example, Patent Document 1), a technique for switching the criteria for ACS determination depending on the determination result of the manuscript type (for example, Patent Document 2). (See), a technique for adjusting the determination conditions used by the ACS means based on the determination results for each region according to the type of image (see, for example, Patent Document 3) has been proposed.

Japanese Unexamined Patent Publication No. 2010-21628 Japanese Unexamined Patent Publication No. 2008-35478 Japanese Unexamined Patent Publication No. 2011-142409

Here, as a method for scanning a document, there is known a method of converting a scanned image into image data in various formats, such as a color image, a grayscale image, or a black-and-white binary image. Therefore, when it is determined by the ACS determination that the original is a monochrome original, it is further determined whether the scanned original image is treated as a grayscale image or a black and white binary image (grayscale / black and white binary determination). ), And it is desirable to select an appropriate method for scanning the original.

Information such as when various originals are mixed in the original set as described above and an appropriate document scanning method is not selected for each original, the image is filled and becomes unreadable. There is a problem that the image may be greatly damaged. For example, when a document in which black characters are superimposed on a dark gray background such as a table heading is converted into image data of a black-and-white binary image, the dark gray background is filled in black (so-called black crush) and the characters. Becomes unreadable (so-called crushed characters). In such a case, it is necessary to prevent the characters from being crushed and to prevent the information from being damaged by converting the original into image data of a grayscale image.

On the other hand, there are cases where a manuscript containing only characters and a manuscript containing only photographs are mixed in the manuscript set, but when the manuscript is only photographs, a black-and-white binary image is produced by grayscale / black-and-white binary judgment. Even if it is selected as the reading processing method of, gradation loss occurs, but information loss does not occur. However, if the edge portion in the photograph is extracted as a character area, it may be erroneously determined that the character may be crushed and converted into a grayscale image. Further, for color originals, although it is not necessary to perform grayscale / black-and-white binary determination, if grayscale / black-and-white binary determination is performed for all originals, the originals included in the original set will be determined. Processing time for grayscale / black-and-white binary determination is required for all the minutes.

As described above, when it is determined by the ACS determination that the original is a monochrome original, it is necessary to further appropriately perform grayscale / black and white binary determination. However, the techniques described in Patent Document 1, Patent Document 2, and Patent Document 3 have not been able to appropriately perform grayscale / black and white binary determination.

In view of the above-mentioned problems, the present invention provides an image processing device or the like capable of determining a document type from image data and generating image data having an appropriate output gradation number based on the document type from the image data. The purpose is to do.

In order to solve the above-mentioned problems, the image processing apparatus of the present invention
Input section for inputting image data of manuscript and
A color / monochrome determination unit that determines whether the original is a monochrome original or a color original based on the image data.
A feature amount calculation unit that calculates a feature amount from the image data,
A manuscript type determination unit that determines the type of the manuscript based on the image data,
When the manuscript is a monochrome manuscript, one of a plurality of output gradation number candidates is set as the output gradation number based on the feature amount and the type of the manuscript, and the manuscript is described. When is a color document, an output gradation number setting unit that sets a predetermined gradation number as an output gradation number and
An image processing unit that generates image data in which the number of gradations of the image data is the set number of output gradations,
It is characterized by being equipped with.

The image processing method of the present invention
Steps to enter the image data of the manuscript and
A step of determining whether the original is a monochrome original or a color original based on the image data,
The step of calculating the feature amount from the image data and
A step of determining the type of the manuscript based on the image data,
When the original is a monochrome original, the gradation number of any one of a plurality of output gradation number candidates is set as the output gradation number based on the feature amount and the type of the original, and the original If is a color document, the step of setting a predetermined number of gradations as the number of output gradations and
A step of generating image data in which the number of gradations of the image data is the set number of output gradations,
It is characterized by being equipped with.

The program of the present invention is characterized in that a computer is operated as the image processing device.

The recording medium of the present invention is characterized in that the above program is recorded.

The image forming apparatus of the present invention
A document reader that reads the document and outputs it as image data,
A color / monochrome determination unit that determines whether the original is a monochrome original or a color original based on the image data.
A feature amount calculation unit that calculates a feature amount from the image data,
A manuscript type determination unit that determines the type of the manuscript based on the image data,
When the original is a monochrome original, the gradation number of any one of a plurality of output gradation number candidates is set as the output gradation number based on the feature amount and the type of the original, and the original When is a color document, an output gradation number setting unit that sets a predetermined gradation number as an output gradation number and
An image processing unit that generates image data in which the number of gradations of the image data is the set number of output gradations,
A printing unit that prints the generated image data,
It is characterized by being equipped with.

According to the present invention, it is possible to determine the type of a document from the image data and generate image data having an appropriate number of output gradations based on the type of the document from the image data.

It is a figure for demonstrating the functional structure of the image forming apparatus in 1st Embodiment. It is a figure for demonstrating the information stored in the storage part in 1st Embodiment. It is a flow figure for demonstrating the image processing method in 1st Embodiment. It is a figure for demonstrating the effect in 1st Embodiment. It is a figure for demonstrating the functional structure of the image forming apparatus in 2nd Embodiment. It is a flow figure for demonstrating the image processing method in 2nd Embodiment. It is a figure which shows the relationship between the division of resolution and the character size in 2nd Embodiment. It is a figure for demonstrating the functional structure of the image forming apparatus in 3rd Embodiment. It is a flow figure for demonstrating the image processing method in 3rd Embodiment. It is a figure for demonstrating the weighting coefficient at the time of score calculation for character crush determination in 3rd Embodiment. It is a figure for demonstrating the relationship between the area separation result and the number of values in 3rd Embodiment. It is a figure for demonstrating the operation example in 3rd Embodiment. It is a figure for demonstrating the operation example in 3rd Embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the following description, a mode in which the image processing apparatus according to the present invention forms a part of the image forming apparatus will be illustrated.

[1. First Embodiment]
[1.1 Functional configuration description]
FIG. 1 is a block diagram showing a functional configuration of the image forming apparatus 1 according to the first embodiment. The image forming apparatus 1 is a digital multifunction device having a copy function, a printing function, a scanner function, and the like. The image forming apparatus 1 includes an image processing apparatus 10, an image reading apparatus 20, an image printing apparatus 30, and an image data transmitting apparatus 40.

An operation panel 50 is connected to the image processing device 10, the image reading device 20, the image printing device 30, and the image data transmitting device 40. The operation panel 50 includes an operation unit 52 such as a setting button and a numeric keypad for the user to set an operation mode of the image forming apparatus, and a display unit 54 composed of a liquid crystal display or the like.

Various processes executed by the image forming apparatus 1 are controlled by a control unit (a computer including a processor such as a CPU (Central Processing Unit) or a DSP (Digital Signal Processor)). The control unit 100 of the image forming apparatus 1 performs data communication with a computer and other digital multifunction devices connected to the network via a network card (not shown) and a LAN (Local Area Network) cable. Hereinafter, each part of the image forming apparatus 1 will be described in detail.

The image reading device 20 is a device that optically reads an image from a document. The image reading device 20 comprises, for example, a color scanner having a CCD (Charge Coupled Device), and the reflected light image from the document is converted into RGB (R: red, G: green, B: blue) or monochrome using the CCD. It is read as an analog signal and output to the image processing device 10. The image reading device 20 does not have to be a scanner, and may be, for example, a digital camera or the like.

The image processing device 10 includes a control unit 100, an image input unit 105, an image processing calculation unit 110, a feature amount calculation calculation unit 120, an image output unit 130, a document type determination unit 140, a storage unit 150, and an output gradation number setting unit. It has 160.

The image input unit 105 is a functional unit that inputs image data of a document by generating input image data based on an analog signal output by the image reading device 20 to the image processing device 10. The image input unit 105 is not limited to the above example, and may receive image data from the outside and input the received image data as input image data.

The image processing calculation unit 110 is a functional unit that performs image processing on the input image data and generates new image data (output image data). Specific examples of image processing include resolution change processing, smoothing processing, sharpening processing, gradation conversion processing such as gamma correction, color / monochrome conversion processing, black-and-white binary image or multi-gradation image (grayscale image, etc.). For example, conversion processing into image data having a different number of gradations.

The feature amount calculation calculation unit 120 is a functional unit that calculates a feature amount to be used when determining an appropriate output gradation number based on image data. The feature amount may be simple numerical data, or may be feature amount image data that handles a group of numerical data indicating the feature amount for each pixel of the image data as image data. Further, the feature amount does not necessarily have to be one component, and may be a vector having a plurality of components. That is, it is also possible to obtain feature amount image data having the same number of pixels as the first evaluation image and the second evaluation image, which will be described later, and having a plurality of components for each pixel.

The image output unit 130 is a functional unit that outputs the output image data generated by the image processing device 10 to the outside of the image processing device 10 in the form of electronic data.

The manuscript type determination unit 140 is a functional unit that determines the type of the manuscript (manuscript type determination) and outputs the result of determining the type of the manuscript (manuscript type determination result). As a method for determining the type of manuscript, any known method may be used. For example, analysis of the maximum concentration value for each category using a concentration histogram disclosed in JP-A-2002-218232 is used. Therefore, it is possible to use a method of determining whether the scanned original is a photographic original or a character original.

The storage unit 150 is a functional unit that stores information on a plurality of output gradation number candidates and various other setting information in advance, and temporarily stores image data created by image processing calculation during image processing. is there. The storage unit 150 can use a volatile memory, a non-volatile memory, or a combination thereof. Further, a magnetic storage medium such as a hard disk device may be used.

The image processing calculation unit 110, the feature amount calculation calculation unit 120, the document type determination unit 140, and the output gradation setting unit 160 operate as each functional unit when the control unit 100 executes a program. , Each functional part may be realized by hardware.

FIG. 2 is an example of information stored in the storage unit 150 in the present embodiment. FIG. 2A is information (hereinafter, referred to as “manuscript type information”) indicating whether or not grayscale / black and white binary determination is performed for each type of manuscript. Specifically, as shown in FIG. 2A, the manuscript type information includes a manuscript type determination result (for example, “character”) indicating the type of the manuscript to be determined by the manuscript type determination unit 140 and the type of the manuscript. This is information associated with the presence / absence of grayscale / black / white binary determination (for example, “ON”) indicating whether or not to perform grayscale / black / white binary determination for each.

The type of manuscript is a classification of manuscripts based on the nature of the image displayed on the manuscript. In the present embodiment, the type of the manuscript determined by the manuscript type determination unit 140 is any of the following five types.
(1) Characters Most of the originals are characters composed of characters with clear edges.
(2) Character / photographic paper photo The manuscript is a paper (for example, copy paper) to which a photo of photographic paper or glossy paper is attached, and the manuscript is a manuscript in which photographs and characters are mixed.
(3) Photographic paper The manuscript is photographic paper, and most of the manuscript is a manuscript composed of photographs expressed in continuous gradation such as silver halide photographs.
(4) Printed Photograph The manuscript is paper other than photographic paper (for example, printed matter), and most of the manuscript is a manuscript composed of photographs expressed by a halftone method such as halftone dots or a multi-line screen. ..
(5) Character / Printed Photograph The original is paper other than photographic paper (for example, copy paper), and the original is a manuscript in which photographs and characters are mixed.

For example, in the example of FIG. 2A, if the document type determination result output by the document type determination unit 140 is "character", "character / photographic paper photograph", or "character / print photograph", grayscale / black and white binary determination is performed. Indicates that If the original type determination result output by the original type determination unit 140 is "photographic paper photograph" or "printed photograph", the grayscale / black and white binary determination is not performed and the default setting (output set as the initial value) is performed. Indicates that the output is based on the number of gradations).

Further, FIG. 2B is information showing the parameter set used in the grayscale / black and white binary determination for each type of manuscript (hereinafter, referred to as “parameter set information”). Specifically, as shown in FIG. 2B, the parameter set information includes a manuscript type determination result (for example, “character”) indicating the type of the manuscript to be judged by the manuscript type determination unit 140, and grayscale. This is information associated with a parameter set (for example, "Param_Text") used in black-and-white binary determination.

For example, as a parameter set used when the document type determination result is "character", since the entire area of the input image data is a character area, it is used for determination whether or not the edge included in the image is a character edge. Store the parameter set that lowers the threshold. On the other hand, as a parameter set used when the document type determination result is "character / photographic paper photograph", in order to prevent an edge included in the photographic area from being erroneously detected as a character edge, the image is included. Stores a parameter set that raises the threshold used to determine whether the included edge is a character edge. By doing so, it is possible to switch the parameter set for extracting characters according to the document type determination result, and improve the accuracy of grayscale / black and white binary determination.

In addition to the above-mentioned information, the storage unit 150 stores a plurality of output gradations, which are the number of gradations when the input image data is output, as output gradation number candidates. For example, the output gradation value for a color image when the color original is output as the image data of the color image and the output gradation value for the monochrome image when the monochrome original is output as the image data of the grayscale image are stored. To do.

The output gradation number setting unit 160 is the most appropriate for representing input image data among a plurality of output gradation number candidates stored in the storage unit 150 based on the feature amount calculated by the feature amount calculation calculation unit 120. This is a functional unit that determines the number of gradations and sets it as the number of output gradations. The image processing calculation unit 110 converts the number of gradations of the input image data according to the number of output gradations set by the output gradation number setting unit 160 to generate the output image data.

The image printing device 30 is a device that prints an image on paper based on the output image data. The image printing device 30 prints an image based on the output image data on paper by using, for example, a general electrophotographic multifunction device or a printer.

The image data transmission device 40 is a device that transmits output image data to the outside via a network. The image data transmission device 40 transmits the output image data to the outside by using, for example, a general e-mail.

[1.2 Processing flow]
Next, with reference to FIG. 3, a method of setting the number of output gradations in the image forming apparatus 1 having the above configuration will be described. The process shown in FIG. 3 is executed by the control unit 100 of the image forming apparatus 1 controlling the functional unit included in the image forming apparatus 1.

[1.2.1 Initial setting of output gradation number]
First, the initial value of the number of output gradations is set and stored in the storage unit 150 (step S102). For example, in order to output appropriate image data when the original is read after the user manually sets either "grayscale" or "black and white binary" as the output gradation. The number of output gradations corresponding to the user's setting is stored in the storage unit 150 as an initial value. Specifically, "256 gradations" and "16 gradations" are stored as initial values of the number of output gradations corresponding to the "grayscale" setting, and the output floor corresponding to the "black and white binary" setting is stored. "2 gradations" are stored as the initial value of the key. By doing so, when the output gradation is manually set by the user, the number of output gradations corresponding to the gradation set by the user (for example, the output gradation) regardless of the type of the original. Image data can be output in 256 gradations or 16 gradations if is "gray scale", and 2 gradations if is "black and white binary"). In addition, the output scale used as a reference value when it is necessary to specify the output gradation of grayscale or black-and-white binary when there is no setting by the user and the judgment of grayscale / black-and-white binary is not performed. The key value is stored in the storage unit 150 as the initial value of the output gradation number.

[1.2.2 Manuscript type judgment]
Subsequently, the control unit 100 reads the originals one by one from the original set placed on the ADF via the image reading device 20 (step S104). At this time, the image input unit 105 inputs the image data of the original by generating the input image data based on the analog signal output from the image reading device 20. Subsequently, it is determined whether or not the original is a monochrome original (whether it is a color original) based on the input image data by the ACS determination (step S106). When it is determined that the original is a monochrome original as a result of the ACS determination (step S106; Yes), the original type determination unit 140 determines the original type for specifying the type of the original to be processed based on the input image data. (Step S108). As described above, when the original is a monochrome original, character crushing may occur due to handling as a black and white binary image depending on the type of the original. Therefore, when the document type determination unit 140 determines the type of the document, the document that may cause character crushing is targeted for grayscale / black and white binary determination. By doing so, it is possible to efficiently process only the originals that may cause character crushing.

As the processing content of the manuscript type determination, for example, a method of classifying and analyzing the concentration having the maximum frequency using a density histogram as disclosed in JP-A-2002-218232 can be used. .. According to the manuscript type judgment, a printed manuscript composed of photographs in which most of the manuscript is represented by halftone dots, a manuscript of characters with sentences written on it, or a photo of both text and photographic paper are on the same page. Each manuscript is judged as a type of manuscript such as "printed photograph", "character" or "character / photographic paper photograph". As described above, the manuscript type determination unit 140 outputs any of the manuscript type determination results of the manuscript type information stored in the storage unit 150 as the result of the manuscript type determination.

[1.2.3 Grayscale / black and white binary judgment and output gradation control]
Subsequently, the control unit 100 determines whether or not the original type determination result is a “photographic paper photograph” or a “printed photograph”, that is, whether or not the original is a photographic original that does not cause character crushing (step S110). ). If the manuscript is a photographic manuscript, the characters will not be crushed. Therefore, the control unit 100 does not execute the grayscale / black and white binary determination for the output gradation number setting unit 160, and outputs the “grayscale” or “black and white binary” set as the initial value in step S102. Control is performed to set the initial value of the output gradation number to the output gradation number of the image data to be output so that the image data is output in the gradation number (step S110; Yes → step S112). If "gray scale" is set as the initial value when the output gradation set by the user or the output gradation is not set, the output gradation number setting unit 160 stores in step S102. 256 gradations and 16 gradations are set as the number of output gradations. Further, if the output gradation set by the user or the "black and white binary value" is set as the initial value when the output gradation is not set, the output gradation number setting unit 160 may perform step S102. The two gradations stored in the above are set as the number of output gradations.

On the other hand, in step S110, when the manuscript type determination result is not a photographic manuscript such as "printed paper photograph" or "printed photograph", that is, "character", "character / photographic paper photograph", and "character / printed photograph". If this is the case, there is a possibility that a document that causes crushed characters is included. Therefore, the control unit 100 controls the output gradation number setting unit 160 so as to perform grayscale / black and white binary determination. Therefore, first, the feature amount calculation calculation unit 120 calculates the feature amount of the input image data used for the grayscale / black and white binary determination (step S110; No → step S114). Then, the output gradation number setting unit 160 executes grayscale / black-and-white binary determination based on the type and feature amount of the original (step S116). When executing the grayscale / black and white binary determination, the output gradation number setting unit 160 may determine using the parameter set corresponding to the document type determination result among the parameter set information stored in the storage unit 150. Good. Further, the output gradation number setting unit 160 appropriately sets the output gradation number based on the determination of the grayscale / black-and-white binary determination so that the characters are not crushed (step S118). That is, the output gradation number setting unit 160 sets as the output gradation number the number of gradations corresponding to the determination of grayscale / black and white binary determination from the output gradation number candidates stored in the storage unit 150. To do.

If the original is a color original as a result of the ACS determination in step S106, the output gradation number setting unit 160 may set the output gradation number to the gradation number for the color image (step S106; No → step S120).

Subsequently, the image processing calculation unit 110 generates output image data obtained by converting the number of gradations of the input image data into the number of output gradations set in any of the processes of step S112, step S118, and step S120 ( Step S122). The image processing calculation unit 110 may perform processing such as smoothing processing, sharpening processing, and gamma correction based on information such as the type of original, the feature amount, and the number of output gradations. Then, if the processing is performed on all the input image data, that is, if all the originals placed on the ADF are read out, the processing ends (step S124; Yes). If all the input image data has not been processed, the process returns to step S104 (step S124; No → step S104).

According to the present embodiment, it is possible to determine whether to output a monochrome document as black-and-white binary image data or as grayscale image data based on an information loss evaluation value in which information loss is quantified. it can. At this time, when a photographic original with less possibility of character crushing is read, it is possible to set the number of output gradations without performing grayscale / black and white binary determination. Therefore, the time required for processing execution can be reduced, the man-hours for image processing until the output image is obtained can be shortened, and the speed can be increased.

Further, it is possible to change the parameter set used for the grayscale / black and white binary determination based on the determination result of the original type. Therefore, the accuracy of grayscale / black and white binary determination can be improved.

For example, when a document as shown in FIG. 4A is read, it is conventionally output as a character document and as image data as shown in FIG. 4B. That is, some places are crushed by black and the manuscript cannot be read. However, according to the present embodiment, since the feature amounts are also calculated and output as grayscale, the image processing apparatus 10 outputs the image data in the state of FIG. 4A.

[2. Second Embodiment]
Next, the second embodiment will be described. Recently, documents are read with an optimum reading resolution. For example, when scanning a document that does not contain small characters, the resolution can be set low so that scanning can be performed without unnecessarily increasing the file size. Further, when reading a document containing small characters, the resolution can be increased so that the characters can be read without impairing the legibility of the characters. In order to realize such a thing, a technique of switching the reading resolution according to the original has been proposed.

For example, based on the pixel values of the pixels that make up the image of the original, the character candidates related to the image are specified, the character string is specified using the character candidates, and the image is output based on the character candidates or the character strings. A technique of an image processing device for determining the output resolution has been proposed. For example, any known method can be used as a method for determining the output resolution related to the image in consideration of the influence of the non-character region contained in the image. For example, Japanese Patent Application Laid-Open No. 2017-55221 can be used. Use the disclosed method.

In the second embodiment, one of the methods is used to estimate the resolution to determine whether or not a small character is included as the size of the character contained in the document. Further, in the grayscale / black and white binary determination, the parameters used for determining the character area and the like are switched according to the resolution. In the second embodiment, FIG. 1 of the first embodiment is replaced with FIG. 5, and FIG. 2 is replaced with FIG. The same functional parts are designated by the same reference numerals, and the description thereof will be omitted.

[2.1 Functional configuration]
FIG. 5 shows a block diagram showing a functional configuration of the image forming apparatus 2 in the present embodiment. The resolution estimation unit 170 identifies the character candidates constituting the image, specifies the character string using the character candidate, and determines the output resolution related to the output of the image based on the character candidate or the character string. It is a functional part of. The resolution estimation unit 170 operates when the control unit 100 executes a program, but it may be realized separately by hardware.

[2.2 Processing flow]
[2.2.1 Initial setting of minimum font size]
The processing flow in this embodiment will be described with reference to FIG. First, before performing the grayscale / black-and-white binary determination, the resolution and the minimum character size suitable for the resolution are set in advance and stored in the storage unit 150 as initial values (step S102 → step S202). The minimum character size is the minimum value of the edge length included in the area treated as the character area. For example, regarding the area including the edge, if the length of the short side of the area is equal to or larger than the minimum character size, it is treated as a character area, and if the length of the short side of the area is less than the minimum character, it is treated as a character area. Does not handle. Here, for example, when the original is read after the resolution is manually specified by the user, the reading is performed at the resolution specified by the user, so that the minimum character size suitable for the resolution specified by the user is obtained. The value is referred to and determined as an initial value, and stored in the storage unit 150. Further, it is used when the response of the resolution estimation unit 170 is not properly performed, such as when the resolution is not manually specified by the user and the resolution estimation by the resolution estimation unit 170 described above is difficult. It is also used as a provisional value. For example, when 300 dpi is set as the initial resolution, a value (for example, 8 pixels) according to the initial resolution is stored in the storage unit 150.

[2.2.2 Execution of resolution estimation]
Next, when the type of the scanned manuscript is not a photographic manuscript such as "printed paper photograph" or "printed photograph", that is, when it is "character", "character / photographic paper photograph", or "character / printed photograph". The resolution estimation unit 170 estimates the resolution of the input image data (step S110; No → step S204). As an outline of the resolution estimation process, first, the resolution estimation unit 170 extracts edges in the image, and extracts a character-like area as a character rectangle from the extracted edges. Next, the resolution estimation unit 170 determines whether or not the pixels included in the character rectangle constitute a character string for the extracted character rectangle, and determines that the pixels constituting the character rectangle are character strings. Extracts a character rectangle as a character string. Then, the resolution estimation unit 170 estimates the optimum output resolution as the resolution of the input image data for the character rectangle extracted as the character string. As a method of estimating the resolution of the input image data, there is a method of making a determination by deriving from a correspondence table between the rectangle size and the character size and the output resolution.

[2.2.3 Regulation of minimum font size based on estimated resolution]
Regarding the estimated resolution of the image data, considering the relationship between the resolution and the resolution of the characters, the relationship between the resolution division of the image input device and the minimum character size is obtained, for example, the relationship as shown in FIG. Become. In the example of FIG. 7, the resolution of the image data is x, and the value of x is divided into three levels of less than 150 dpi, 150 dpi or more and less than 300 dpi, and 300 dpi or more. It is a figure which shows that 6 or 8 pixels are defined as the number of pixels of the character of the minimum character size. Although the minimum character size is shown by the number of pixels in FIG. 7, the minimum character size is not limited to the number of pixels, and may be shown by, for example, the number of points or a series.

Then, in the grayscale / black-and-white binary determination, the minimum number of character pixels is used in the character area determination process for determining whether or not the character area is an attribute of the pixel of interest. For example, when the resolution of the input image data is determined to be 150 dpi, 6 pixels is the minimum character size. Therefore, in the character area determination, when the number of pixels of the minimum character including the peripheral pixels of the attention pixel is used as one of the determination criteria, the pixel value (for example, the area including the peripheral pixels of the attention pixel is short). Characters whose side length) is less than 6 pixels can be regarded as characters. By doing so, it is possible to instantly determine whether or not the character is not a character, although it is only from the viewpoint of the number of pixels, without using the result of other determination processing.

Further, since the edge portion extracted in the area other than the original character area in the image with less than 6 pixels is not judged as a character, the character edge is not classified by the minimum number of pixels of the character. It is possible to reduce the edges in the photograph that were erroneously detected as.

Information indicating the resolution and the resolution of characters as shown in FIG. 7 is stored in, for example, the storage unit 150. Then, the control unit 100 defines (estimates) the minimum character size in the input image data based on the information indicating the resolution estimated in step S204, the resolution stored in the storage unit 150, and the character resolution (step). S206). Then, the output gradation number setting unit 160 executes grayscale / black-and-white binary determination in consideration of the number of pixels of the minimum character (step S208). When the original is read after the resolution is specified by the user, the control unit 100 skips step S204 and step S206, and based on the minimum character size suitable for the resolution specified by the user, Grayscale / black and white binary judgment is executed.

When the above steps are carried out and all the input image data are processed, the process ends (step S124; Yes). If all the input image data has not been processed, the process returns to step S104 (step S124; No → step S104).

According to the present embodiment, it is possible to switch the determination conditions for grayscale / black and white binary determination according to the resolution determination result. By doing so, for example, a small character edge region falsely detected in the photographic region can be excluded from the region to be determined in the grayscale / black and white binary determination.

[3. Third Embodiment]
Next, a third embodiment will be described. In the third embodiment, the input image data, particularly the input image data when the original is read by a scanning device, is classified smaller than the relatively large classification of images such as "characters" and "photographic paper photographs". It is an embodiment to carry out. For example, when the black characters (black characters) included in the input image data are further classified, whether they are black characters in the character area without a background or halftone dots, black characters with a background (black characters with a background), or black characters on the halftone dots. It is classified as to whether it is (black letters on halftone dots). By classifying the characters into small pieces in this way, it is possible to adaptively determine whether or not there is character crushing in each area on the document, instead of making the entire surface of the document a uniform paper surface. In the third embodiment, FIG. 1 of the first embodiment is replaced with FIG. 8, and FIG. 2 is replaced with FIG. The same functional parts are designated by the same reference numerals, and the description thereof will be omitted. In the third embodiment, FIG. 1 of the first embodiment is replaced with FIG. 8, and FIG. 2 is replaced with FIG. The same functional parts are designated by the same reference numerals, and the description thereof will be omitted.

[3.1 Functional configuration]
FIG. 8 shows a block diagram showing a functional configuration of the image forming apparatus 3 in the present embodiment. The area separation determination unit 180 calculates a feature amount such as a difference from peripheral pixels for each pixel constituting the image, and uses statistical determination to determine the relationship with the surroundings of the pixel. This is a functional unit for determining what kind of characteristics the pixel of interest has in view of the above. The area separation determination unit 180 operates when the control unit 100 executes a program, but it may be realized separately by hardware.

[3.2 Processing flow]
[3.2.1 Initial setting of parameters for character crush judgment]
The processing flow in this embodiment will be described with reference to FIG. First, the initial value of the determination parameter used in the area separation determination unit 180 that determines whether or not the character is crushed is stored in the storage unit 150 (step S302). For example, it is assumed that the determination result in the area separation determination unit 180 is classified into "character", "halftone dot", "base", or "other". Further, it is assumed that the region of black characters is extracted by the region separation determination unit 180. At this time, among the black characters, the black characters in the area classified as "ground" (black characters with background) and the black characters in the area classified as "halftone dots" (black characters on halftone dots) are subject to character crushing. .. Therefore, in the areas belonging to the "base" and "halftone dots", which are areas where character crushing is likely to occur, the possibility of character crushing is higher than in other areas, so it is easy to determine that character crushing has occurred. Give weight like this. For example, as a determination method, when performing a threshold value process for determining whether or not a predetermined value is exceeded, the weighting coefficient of the region belonging to these "base" and "halftone dot" regions is set large. In this way, in the "base" and "halftone dot" areas, the numerical value used for the threshold processing tends to be larger than in the other areas.

For example, a weighting coefficient having a relationship as an initial value as shown in FIG. 10 is stored in the storage unit 150 as an initial value. This setting can be used as a provisional value used when the response is not properly performed, such as when the region separation processing in the region separation determination unit 180 described above is difficult. By doing so, even if the region separation determination unit 180 fails, the result of the region separation is not reflected, and the operation with the same accuracy as when the region separation is not performed can be guaranteed. That is, as the weighting coefficient, the "other" weight of 1.0 is applied to all the pixels, so that the weighting coefficient does not change.

[3.2.2 Execution of area separation processing]
Next, when the type of the scanned manuscript is not a photographic manuscript such as "printed paper photograph" or "printed photograph", that is, when it is "character", "character / photographic paper photograph", or "character / printed photograph". The area separation determination unit 180 performs the area separation process (step S110; No → step S304). The area separation determination unit 180 separates the input image data into a plurality of areas including a character area, a halftone dot area, and a photographic area as the area separation process. As an outline of the process of separating the regions, statistics such as the maximum / minimum density value and the maximum density difference are obtained in a predetermined area for determination, and statistics are obtained for each pixel based on these statistics. It is determined what kind of area the predetermined area is on the condition of the magnitude relation of the amount. By doing so, it is possible to discriminate areas such as "base" and "photographic paper", or "characters" and "halftone dots" and "others".

By executing the area separation here, when the attributes for each pixel of the scanned document are classified into five types as shown in FIG. 10, for example, the number of values corresponding to the result of the area separation attribute. An example of is shown in FIG. The two-dimensional array data corresponding to the total number of pixels having this number of values is used for the processing of switching the weighting coefficient in the judgment of the next grayscale / black and white binary judgment.

[3.2.3 Grayscale / black and white binary judgment processing using weight coefficient for character collapse judgment]
Subsequently, a grayscale / black-and-white binary determination using the weighting coefficient in the character collapse determination is performed (step S306). Here, a grayscale / black-and-white binary determination using a weighting coefficient in a character crush determination will be described. FIG. 12 is an example of a grayscale image when the character “mountain” above the halftone dots is read. The rectangle indicates the pixel, and the numerical value in the rectangle indicates the pixel value. Assuming that the result of region separation at this time is obtained as shown in FIG. 13, the weighting coefficient in the attribute shown in FIG. 10 is applied to the weighting coefficient in the discriminant using the corresponding coefficient for the determination.

For example, the range of 10 × 10 pixels surrounded by the thick frame on the upper left of FIG. 12 is defined as the block α. In this case, in the calculation formula of the score value (evaluation value) of the pixel at the position A, which is the pixel of the halftone dot, the weight coefficient of a at the same coordinates on the two-dimensional array data of FIG. The weight of 2.00 of the number of values 16 (= halftone dots) in the correspondence table of FIG. 10, which is the weight coefficient of the position, is used.

Here, the feature amount of the block α is a value calculated before executing the character crush determination process, and is calculated by the feature amount calculation calculation unit 120. If the feature amount of the block α is calculated to be 100, the score value of the pixel at the position A, which is the pixel of the halftone dot, is calculated by the following formula.
ScoreA = Feature amount of block α × Weight coefficient of pixel position A
= 100 x 2.00
= 200

Similarly, in the calculation formula of the score value of the pixel at the position b corresponding to the pixel of the character on the actual halftone dot in FIG. 13, the weighting coefficient at the position B in FIG. 12 is shown on the halftone dot from the correspondence table of FIG. Black letters 10.0 are used. Therefore, the score value of the pixel at the position B, which is the pixel of the character on the halftone dot, is calculated by the following formula.
ScoreB = Feature of block α × Weight coefficient of pixel position B
= 100 x 10.00
= 1000
The score value is calculated for each pixel in the block as described above.

For each pixel, the weighting coefficient changes according to the attributes of the pixel. Therefore, when the weighting coefficient is assigned as shown in FIG. 10, in the same block, the "halftone dot" pixel is doubled with respect to the "other" pixel, and the "halftone dot character" pixel or In the case of "black characters with background" pixels, the score value fluctuates 10 times. Therefore, it is possible to make it easier to distinguish "black characters with a background" and "black characters on halftone dots" that may be crushed in black.

If the score value calculated for each pixel in this way is larger than a predetermined value for the sum of the score values for the total number of pixels in the block, it is determined as a block candidate in which character crushing occurs.

Then, when the above steps are carried out and all the input image data are processed, the process ends (step S124; Yes). If all the input image data has not been processed, the process returns to step S104 (step S124; No → step S104).

According to the present embodiment, in grayscale / black and white binary determination, it is possible to control the weight of the parameter used for character detection and switch the calculation method for the portion where character crushing is likely to occur. .. Further, in the grayscale / black and white binary determination, it is possible to reduce the influence of the region unnecessary for the determination.

[4. Modification example]
Each functional unit included in the image processing apparatus 10 described in the description of the embodiment may be realized by software, or may be realized in whole or in part by hardware (circuit or device). In any case, it suffices to set an appropriate number of output gradations based on the feature amount of the image data and the type of the original, and generate the input image data as the image data of the set number of output gradations. ..

The present invention may be applied to a digital copier. Further, it may be applied to a digital color multifunction device having a copy function, a printer function, a facsimile transmission function, a scan to e-mail function and the like. The digital color multifunction device may further include a communication device including, for example, a modem or a network card. When transmitting a facsimile, the communication device performs a transmission procedure with the other party with a modem, and when a state in which transmission is possible is secured, image data compressed in a predetermined format (image read by a scanner). Data) is read from the memory, necessary processing such as changing the compression format is performed, and the data is sequentially transmitted to the other party via the communication line.

When receiving the facsimile, the CPU receives the image data transmitted from the other party and inputs it to the image processing device while performing the communication procedure, and the image processing device compresses the received image data (not shown). Stretching is performed in the stretching processing section. The stretched image data is subjected to rotation processing and resolution conversion processing, output gradation correction, and gradation reproduction processing, if necessary, and is output from the image printing device 30.

In addition, data communication is performed with a computer or other digital multifunction device connected to the network via a network card or LAN cable.

Although the color multifunction device has been described above, it may be a monochrome multifunction device. In the present invention, an image including an object taken with a background is input to a computer-readable recording medium on which a program to be executed by a computer is recorded, and image data of only the target area is input. It is also possible to record the processing method to be extracted.

As a result, a recording medium on which the program code (executable program, intermediate code program, source program) that performs the above processing is recorded can be freely provided.
In the present embodiment, as the recording medium, a memory (for example, a ROM) itself (not shown) may be a program medium because the processing is performed by the microcomputer, or is shown in the figure. Although it is not provided, a program reading device may be provided as an external storage device, and the program media may be readable by inserting a recording medium into the program reading device.

In either case, the stored program may be configured to be accessed and executed by the microprocessor, or in either case, the program code is read and the read program code is the microcomputer's. A method may be used in which the program is downloaded to a program storage area (not shown) and the program is executed. It is assumed that this download program is stored in the main unit in advance.
Here, the program media is a recording medium configured to be separable from the main body, and is a tape system such as a magnetic tape or a cassette tape, a magnetic disk such as a flexible disk or a hard disk, a CD-ROM / MO / MD / DVD, or the like. Disk system of optical disk, card system such as IC card (including memory card) / optical card, or semiconductor by mask ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash ROM, etc. It may be a medium that fixedly carries the program code including the memory.

Further, in the present embodiment, since the system configuration is such that a communication network including the Internet can be connected, the medium may be a medium that supports the program code in a fluid manner so as to download the program code from the communication network. When the program is downloaded from the communication network in this way, the download program may be stored in the main unit in advance or may be installed from another recording medium. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave, in which the program code is embodied by electronic transmission.

The above-mentioned image processing method is executed by reading the recording medium by a program reading device provided in the image processing device.

1, 2, 3 Image forming device 10 Image processing device 100 Control unit 105 Image input unit 110 Image processing calculation unit 120 Feature amount calculation calculation unit 130 Image output unit 140 Manuscript type determination unit 150 Storage unit 160 Output gradation number setting unit 170 Resolution estimation unit 180 Area separation judgment unit 20 Image reading device 30 Image printing device 40 Image data transmission device 50 Operation panel 52 Operation unit 54 Display unit

Claims (8)

Input section for inputting image data of manuscript and
A color / monochrome determination unit that determines whether the original is a monochrome original or a color original based on the image data.
A feature amount calculation unit that calculates a feature amount from the image data,
A manuscript type determination unit that determines the type of the manuscript based on the image data,
When the original is a monochrome original, the gradation number of any one of a plurality of output gradation number candidates is set as the output gradation number based on the feature amount and the type of the original, and the original When is a color document, an output gradation number setting unit that sets a predetermined gradation number as an output gradation number and
An image processing unit that generates image data in which the number of gradations of the image data is the set number of output gradations,
An image processing device characterized by being equipped with. The image processing apparatus according to claim 1, wherein the output gradation number setting unit sets an output gradation number using a parameter set according to the type of the document. A character size estimation unit for estimating the minimum character size of characters included in the image data is further provided.
The image processing apparatus according to claim 1 or 2, wherein the output gradation number setting unit sets an output gradation number based on the feature amount, the type of the original document, and the character size. .. Further, an area separation unit for separating the image data into a plurality of areas including a character area, a halftone dot area, and a base area is provided.
The image processing apparatus according to claim 1 or 2, wherein the output gradation number setting unit sets the output gradation number so that characters on the halftone dot region and the background region do not collapse. Steps to enter the image data of the manuscript and
A step of determining whether the original is a monochrome original or a color original based on the image data,
The step of calculating the feature amount from the image data and
A step of determining the type of the manuscript based on the image data,
When the original is a monochrome original, the gradation number of any one of a plurality of output gradation number candidates is set as the output gradation number based on the feature amount and the type of the original, and the original If is a color document, the step of setting a predetermined number of gradations as the number of output gradations and
A step of generating image data in which the number of gradations of the image data is the set number of output gradations,
An image processing method characterized by being provided with. A program for operating the image processing device according to any one of claims 1 to 4 on a computer. A computer-readable recording medium on which the program according to claim 6 is recorded. A document reader that reads the document and outputs it as image data,
A color / monochrome determination unit that determines whether the original is a monochrome original or a color original based on the image data.
A feature amount calculation unit that calculates a feature amount from the image data,
A manuscript type determination unit that determines the type of the manuscript based on the image data,
When the original is a monochrome original, the gradation number of any one of a plurality of output gradation number candidates is set as the output gradation number based on the feature amount and the type of the original, and the original When is a color document, an output gradation number setting unit that sets a predetermined gradation number as an output gradation number and
An image processing unit that generates image data in which the number of gradations of the image data is the set number of output gradations,
A printing unit that prints the generated image data,
An image forming apparatus characterized by being provided with.