JP2010004141A - Image processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus that achieves high character reproducibility in accordance with characters included in an original. <P>SOLUTION: A character parts analysis part 18 generates character parts information describing a variety of information by the unit of the parts configuring each character on the basis of OCR information and outline information. The character parts information includes information, such as length and breadth to be reference for each parts, an optimal size of a smoothing filter in the case of length and breadth ratio to be parts reference, weighting of smoothing filter, the kind of smoothing filter, weighting of edge enhancement filter, and the kind of edge enhancement filter. A smoothing processing part 20 and an edge enhancement processing part 24 each performas image processing, based on character parts information from the character parts analysis part 18. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, and more particularly to a technique for improving character reproducibility in image forming processing.

  In an image processing apparatus such as a copying machine, it is important to improve the performance of printing characters and the like described on a document with high character quality, so-called character reproducibility. As a method for improving the character reproducibility, various processing methods using a digital image processing technique have been proposed. As a typical method of such an image processing technique, image data read by a scanner unit or the like is divided into regions according to types, for example, for character regions, filter processing suitable for the character regions, etc. There was a way to do.

Various techniques have also been proposed for identifying and extracting a character area from a manuscript. For example, Japanese Patent Laying-Open No. 2001-350763 (Patent Document 1) discloses an image search apparatus that searches only a document image including a character image that matches an arbitrary keyword input in advance from a document. Japanese Patent Laid-Open No. 2002-150306 (Patent Document 2) discloses that a specific document block is extracted from image data obtained by reading a document such as a newspaper or a magazine, and is easy to read and has a fixed size. An image processing apparatus that can obtain document data that can be efficiently pasted to a region without waste is disclosed.
JP 2001-350763 A JP 2002-150306 A

  As described above, in the conventional image processing method, filter processing suitable for a specific area (for example, for each character area) is performed, but character attributes (font type, font size, number of strokes) In some cases, the filtering process is not necessarily appropriate. As a result, there is a problem that there are characters in the character area that cannot ensure the target character quality.

  Accordingly, the present invention has been made to solve such a problem, and an object of the present invention is to provide an image processing apparatus capable of realizing high character reproducibility according to characters included in a document.

  An image processing apparatus according to an aspect of the present invention includes a character recognition unit that recognizes a character from input image data and generates first information indicating a character code of the character and attributes of parts constituting the character, and input image data An outline acquisition unit that acquires an outline of a character from the image and generates second information including position information of the character, and an image processing method according to parts constituting each character based on the first information and the second information And an image processing means for executing the corresponding image processing determined by the analysis means for the area where each part constituting each character is located with respect to the input image data.

  An image processing apparatus according to another aspect of the present invention includes a character recognition unit that recognizes a character from input image data and generates first information indicating a character code of the character and an attribute of the character; An outline acquisition unit that acquires an outline and generates second information including position information of the character; an analysis unit that determines an image processing method according to each character based on the first information and the second information; Image processing means for executing the corresponding image processing determined by the analysis means for the area where each character is located with respect to the input image data.

  Preferably, the image processing apparatus further includes storage means for storing character information for a plurality of fonts, and the character recognition means recognizes the font type of each character with reference to the storage means.

  Preferably, the character recognizing unit generates first information including a font type, a size, and a stroke number of each character, and the analyzing unit is based on the font type, size, and the number of strokes of each character included in the first information. Determine the image processing method.

  Preferably, the image processing means includes a smoothing process, and the analysis means changes the shape of the smoothing filter used in the smoothing process.

  Preferably, the image processing means includes edge enhancement processing means, and the analysis means changes at least one of an edge enhancement amount and an edge enhancement method for the character portion in the edge enhancement processing means.

  Preferably, the image processing apparatus further includes a print unit that prints out the input image data after the image processing, and the analysis unit changes the image processing apparatus according to the print characteristics of the print unit.

  Preferably, the analysis unit determines an image processing method based on the generated information even when either one of the first information and the second information cannot be generated.

  Preferably, the image processing unit performs image processing when a determination is made that the image processing method to be executed should be changed more than a predetermined amount of change between an area where adjacent parts are located or between adjacent characters. Suppress method changes.

  According to the present invention, high character reproducibility can be realized according to the character attribute included in the document. Therefore, it is possible to maintain high character quality of a document in which characters of different character sizes and different fonts are mixed.

  Embodiments of the present invention will be described in detail with reference to the drawings. Note that the same or corresponding parts in the drawings are denoted by the same reference numerals and description thereof will not be repeated.

  As a representative example of the image processing apparatus according to the present invention, a multi function peripheral (hereinafter also referred to as “MFP”) having a scan function in addition to an image forming function (monochrome and / or color) such as a print function and a copy function. .) Will be described below.

<Outline>
FIG. 1 shows a usage pattern of MFP 100 according to the embodiment of the present invention.

  Referring to FIG. 1, in MFP 100 according to the present embodiment, a document loaded on a scanner unit is read by a user to generate image data, and based on the generated image data, analysis of image data to be described later is performed. Processing is executed, and based on the result of the analysis processing, image processing suitable for the document (typically smoothing processing and edge enhancement processing) is selected, and the result of the selected image processing is printed out. The

<Functional configuration>
FIG. 2 is a block diagram showing a functional configuration of MFP 100 according to the embodiment of the present invention.

  Referring to FIG. 2, MFP 100 includes control unit 110, memory 120, operation panel 130, data storage unit 140, network communication unit 150, scanner unit 160, print unit 170, and image processing unit 180. And a FAX communication unit 190. These units are connected to each other via a data bus BUS so that data communication is possible.

  Control unit 110 is typically configured to include an arithmetic device such as a CPU (Central Processing Unit), and provides each process according to the present embodiment by executing a program stored in advance.

  The memory 120 typically includes a volatile memory such as a DRAM (Dynamic Random Access Memory) and a nonvolatile memory such as a flash memory. More specifically, the memory 120 temporarily expands a boot program or the like stored in advance in the nonvolatile memory in the volatile memory when the control unit 110 executes the program.

  The operation panel 130 is a user interface that presents operation information such as an operation menu and a job execution state to the user and accepts a user instruction in response to pressing (contact) by the user. More specifically, the operation panel 130 includes a key input unit 130a as an input unit and a touch panel 130b as an input unit configured integrally with the display unit.

  The key input unit 130a is configured to include a numeric keypad and a key to which each function (copy, FAX, etc.) is assigned, and outputs an instruction corresponding to the key pressed by the user to the control unit 110.

  The touch panel 130b includes a liquid crystal panel and a touch operation detection unit provided on the liquid crystal panel. More specifically, the touch panel 130b visually displays various types of information to the user, and outputs an instruction corresponding to the touch operation to the control unit 110 when detecting a touch operation by the user.

  The data storage unit 140 is typically composed of a relatively large-capacity nonvolatile memory such as a hard disk (HDD), and stores image data read by the scanner unit 160. The data storage unit 140 also stores filter information and font information used for various image processing described later.

  The network communication unit 150 performs data communication with other MFPs and PCs (personal computers) (not shown) via a network NW such as a LAN (Local Area Network). Specifically, a network FAX is transmitted to another MFP, or print data is received from a PC.

  The scanner unit 160 reads image information from a document and generates image data. This image data is stored in the data storage unit 140. More specifically, the scanner unit 160 irradiates light from a light source toward a document placed on the platen glass, and receives light reflected from the document by an imaging element or the like arranged in the main scanning direction. Then read the image information of the document. Alternatively, the scanner unit 160 may be configured to include a document feeding table, a delivery roller, a registration roller, a transport drum, and a sheet discharging table so that continuous document reading can be performed.

  The printing unit 170 forms (prints) an image data read by the scanner unit 160 or image data received by the network communication unit 150 or the FAX communication unit 190 on a paper medium. Typically, the print unit 170 includes an electrophotographic image forming unit. More specifically, the print unit 170 includes an imaging unit including a photoconductor, an exposure device, a development device, and the like, a transfer belt, a fixing device, a paper feed unit, a paper discharge unit, and the like.

  The image processing unit 180 performs image processing on the image data read by the scanner unit 160 or the image data received by the FAX communication unit 190 in accordance with an instruction from the user via the operation panel 130. The processed image data is output to the print unit 170. In particular, the image processing unit 180 analyzes input image data generated by reading a document, and executes image processing suitable for the document based on the analysis result.

  The FAX communication unit 190 is connected to a telephone line and stores FAX data (image data) received from another MFP or the like in the data storage unit 140. Further, the FAX communication unit 190 transmits the FAX data (image data) read by the scanner unit 160 to the transmission destination input by the user pressing the operation panel 130.

<Character attribute>
Next, with reference to FIG. 3 to FIG. 6, the relationship between the character attribute and the interval between the lines constituting the character and the line width will be described.

FIG. 3 is a diagram illustrating an example for comparing the influence of the number of strokes of characters.
Referring to FIG. 3, a general character (type) is a character having a large number of strokes (FIG. 3) even if it is a font such as a Gothic font in which the thickness of the lines constituting the character is substantially the same. Between (a)) and a character with a small number of strokes (FIG. 3 (b)), the interval between the lines constituting the character is different.

FIG. 4 is a diagram illustrating another example for comparing the influence of the number of strokes of characters.
Referring to FIG. 4, even if a Gothic character having the same radical “Gonbeng” is used, if the number of strokes is different, the line spacing is different (FIG. 4A). And FIG. 4 (b)). In addition, even for Gothic characters having the same radical “Kihen”, the line spacing is different if the number of strokes is different (FIG. 4 (c) and FIG. 4 (d)). .

FIG. 5 is a diagram illustrating an example for comparing the influence of the flatness of characters.
Referring to FIG. 5, even if the characters are the same, the intervals between the lines constituting the characters differ depending on the flatness ratio (ratio A / B and ratio a / b).

FIG. 6 is a diagram illustrating an example for comparing the effects of character fonts.
Referring to FIG. 6, even for the same character, the interval between lines constituting the character and the vertical and horizontal line widths differ depending on the font. In other words, the Gothic style (Fig. 6 (a)) and pop style (Fig. 6 (c)) have almost the same width and spacing between the lines constituting the character, but the Mincho style (Fig. 6 (b)). The horizontal line width is larger than the vertical line width. Further, in the line typeface (FIG. 6D), the line width and the line interval are not uniform, and many parts are formed in a curved shape.

  As described above, characters (types) have different line widths and line intervals that constitute each character depending on character attributes such as the number of strokes, font, flatness, and character size. That is, it means that the spatial frequency of each character changes depending on these character attributes. Therefore, when various image processing is performed on these characters, it is necessary to use a filter that takes into account such a spatial frequency.

<Division of characters into small areas>
FIG. 7 is a diagram for explaining division of characters into small areas according to the present embodiment.

  Referring to FIG. 7, in the present embodiment, as an example, each character is divided into nine small regions. As will be described later, this means a unit for changing the contents of the image processing, and it is possible to perform different image processing for each small area.

  As described above, when each character is divided into a plurality of small regions (1) to (9), a region where image processing is performed can be identified in association with parts (such as radicals) constituting the character.

FIG. 8 is a diagram illustrating an example of a correspondence relationship between the parts constituting the character and the small area.
Referring to FIG. 8, for example, “hen” which is a part constituting a character is located in the small areas (1), (4), and (7) (FIG. 8 (a)). In addition, “kanmuri”, which is a part constituting the character, is located in the small areas (1), (2), and (3) (FIG. 8B). Further, “ashi” which is a part constituting the character is located in the small areas (7), (8), (9) (FIG. 8C). Furthermore, “make” which is a part constituting the character is located in the small areas (3), (6), and (9) (FIG. 8D). Furthermore, “Niyo”, which is a part constituting the character, is located in the small areas (1), (4), (7), (8), and (9) (FIG. 8 (e)).

  In this way, if the parts constituting each character can be identified, image processing suitable for each part can be executed in units of small areas where each part is located.

<Outline of image processing>
In MFP 100 according to the present embodiment, character recognition processing (OCR: Optical Character Reader) and outline acquisition processing are executed in parallel on the character area of the image data read from the original by scanner unit 160 (FIG. 2). To do. In the character recognition process, each character is recognized from the character area in the image data and converted into a character code, and the coordinates on the image data, character width, character height, font, tilt, stroke count, radical, color Such information is acquired. In the outline acquisition process, information such as the outline, coordinates, character width, character height, and color of each character is acquired. Here, the outline information of each character is stored in the form of vector information or an image.

  Then, based on the information acquired by the character recognition process and the information acquired by the outline acquisition process, the parts constituting each character included in the character area of the image data are specified, and the small area where each part is located On the other hand, image processing corresponding to each part is performed. Alternatively, image processing corresponding to each character is performed on each character region.

<Functional block>
FIG. 9 is a functional block diagram for realizing image processing according to the embodiment of the present invention. The functions shown in FIG. 9 are typically provided as processing in the image processing unit 180.

  Referring to FIG. 9, the image processing unit 180 includes a character region extraction unit 10, a character recognition unit (OCR) 12, a character information DB 14, an outline acquisition unit 16, a character part analysis unit 18, and a smoothing process. A unit 20, a smoothing filter DB 22, an edge enhancement processing unit 24, and an edge enhancement filter DB 26 are included.

  The character area extraction unit 10 is input image data such as image data read by the scanner unit 160 (FIG. 2) or image data received by the network communication unit 150 (FIG. 2) or the FAX communication unit 190 (FIG. 2). Among these, image data corresponding to the character area is extracted. Various methods are known as character region extraction methods, but typically, a method using a histogram or the like can be applied.

  The character recognition unit 12 performs a character recognition process on the character region extracted by the character region extraction unit 10, and converts the characters included in the character region into character codes. More specifically, the character recognition unit 12 further extracts an image corresponding to each character from the character region, and determines the character code by referring to the character information for each font stored in advance in the character information DB 14.

FIG. 10 is a diagram showing an example of character information stored in the character information DB 14 shown in FIG.
Referring to FIG. 10, as character information, information such as a character code, a font name, the number of parts constituting each character, the position of each part and the number of strokes is stored in advance. In particular, the character information DB 14 according to the present embodiment stores a set of character codes for each of the plurality of fonts so that the characters of the plurality of fonts can be recognized. That is, assuming that the number of recognizable character codes is n and the number of recognizable fonts is m, character information of K = n × m is stored in advance.

  Referring to FIG. 9 again, the outline acquisition unit 16 acquires outline (contour) information of each character for the character region extracted by the character region extraction unit 10. Note that edge detection or the like can be used as an outline acquisition method.

  The OCR information acquired by the character recognition unit 12 and the outline information acquired by the outline acquisition unit 16 are output to the character parts analysis unit 18. Based on the OCR information and the outline information, the character part analysis unit 18 generates character part information describing various pieces of information in units of parts constituting each character.

  FIG. 11 is a diagram showing an example of OCR information and outline information according to the present embodiment.

  Referring to FIG. 11, OCR information 32 and outline information 34 are acquired in association with each character included in the input image data.

  Each of the OCR information 32 includes information such as an ID, a character code, a coordinate on the character image data, a character width, a character height, a font, a tilt, a stroke number, a radical, and a color. Each of the OCR information 32 includes information such as ID, outline of each character, coordinates, character width, character height, and color. Here, the outline information of each character is stored in the form of vector information or an image.

FIG. 12 is a diagram showing an example of character part information according to the present embodiment.
Referring to FIG. 12, the character parts analysis unit 18 disassembles each character included in the character area into one or more parts constituting the character based on the OCR information 32 and the outline information 34 described above. Character part information is generated in association with the part.

  The text part information includes the vertical and horizontal sizes (reference size X, reference size Y) that are the reference of each part, the smoothing filter size (X, Y) that is optimal in the case of the aspect ratio that is the part reference, It includes information such as the weight of the smoothing filter, the type of smoothing filter, the size (X, Y) of the edge enhancement filter, the weight of the edge enhancement filter, and the type of edge enhancement filter.

  Typically, the character part analysis unit 18 reduces the filter size as the part or character having a larger number of strokes, that is, as the spatial frequency is higher, or as the character size is smaller.

  Referring again to FIG. 9, the smoothing processing unit 20 performs, based on the character part information from the character part analysis unit 18, the small region where each part of each character included in the input image data is located. Smoothing is performed according to each part. Or you may make it the smoothing process part 20 perform the smoothing process according to each character with respect to the area | region of each character. This is the same as the process when the character is composed of one part.

  More specifically, the smoothing processing unit 20 selects an appropriate smoothing filter from the smoothing filter DB 22 based on the font type of each character, the number of characters or parts, the character size, etc. Using the smoothing filter, the corresponding small area or character area is processed.

  In addition to these factors, the character part analysis unit 18 switches between the smoothing strength and the smoothing method according to the aspect ratio in the print characteristics of the print unit 170 (FIG. 2). May be. Further, the character parts analysis unit 18 is instructed to switch between changing the smoothing filter type more than a predetermined threshold value between the adjacent small areas or the adjacent character areas. Such switching operation of the smoothing filter may be suppressed.

  FIG. 13 is a diagram illustrating an example of a smoothing filter group stored in the smoothing filter DB 22 illustrated in FIG. 9.

  Referring to FIG. 13, the smoothing filter can change the content of the smoothing process according to the smoothing range, the weighting that is the contribution ratio of the surrounding pixels, the filter shape, and the like. A plurality of smoothing filters are prepared in advance, and one of the plurality of smoothing filters is selected, and an appropriate smoothing process corresponding to the character is executed.

  Referring to FIG. 9 again, the edge enhancement processing unit 24, based on the character part information from the character part analysis unit 18, for the small area where each part of each character included in the input image data is located. Edge enhancement processing according to each part is performed. Alternatively, the edge enhancement processing unit 24 may perform edge enhancement processing according to each character on each character region. This is the same as the process when the character is composed of one part.

  More specifically, the edge enhancement processing unit 24 selects an appropriate edge enhancement filter from the smoothing filter DB 22 at any time based on the font type of each character, the number of characters or parts strokes, the character size, and the like. Using the edge enhancement filter, the corresponding small area or character area is image-processed. By selecting such an edge enhancement filter, the edge enhancement amount and the edge enhancement method for the character portion are switched.

  In addition to these factors, the character part analysis unit 18 may switch between the edge enhancement amount and the edge enhancement method according to the aspect ratio in the printing characteristics of the printing unit 170 (FIG. 2). Good. Further, when the character parts analysis unit 18 is instructed to switch between the adjacent small regions or the adjacent character regions so as to change the type of the edge enhancement filter more than a predetermined threshold value, Such an edge enhancement filter switching operation may be suppressed.

  FIG. 14 is a diagram illustrating an example of the edge enhancement filter group stored in the edge enhancement filter DB 26 illustrated in FIG. 9.

  Referring to FIG. 14, the edge enhancement filter can change the content of edge enhancement processing according to the type of directional filter, the weight of surrounding pixels to be enhanced, the filter shape, and the like. A plurality of edge enhancement filters are prepared in advance in the DB 26, and one of the plurality of edge enhancement filters is selected, and appropriate edge enhancement processing corresponding to the character is executed.

<Effects of image processing method according to the present embodiment>
FIG. 15 is a diagram for describing the effect of the image processing method according to the present embodiment.

  According to the image processing method according to the present embodiment, characters included in the input image data as shown in FIG. 15A are broken down into two parts as an example. Then, appropriate image processing (smoothing processing or edge processing) is performed on each part. Therefore, for example, the filter 1 is used for the left part shown in FIG. 15A and the filter 2 is used for the right part shown in FIG.

  FIG. 15B and FIG. 15C are diagrams showing a change in concentration along the cross-sectional line LN shown in FIG.

  In the prior art, specific smoothing processing and edge enhancement processing are uniformly executed for all character regions. Therefore, as shown in FIG. 15B, even among characters having a large number of strokes or the same character, In a portion with a large number of strokes, sufficient character quality may not be obtained because sufficient smoothing and edge enhancement effects cannot be obtained.

  On the other hand, according to the image processing method according to the present embodiment, the contents of the image processing are changed for each part constituting each character or for each character as described above, so as shown in FIG. In addition, even if a character with a large number of strokes or a portion with a large number of strokes exists in the same character, smoothing and edge enhancement effects can be exhibited, so that sufficient character quality can be ensured.

<Processing procedure>
FIG. 16 is a flowchart showing an image processing procedure according to the embodiment of the present invention.

  Referring to FIG. 16, first, control unit 110 determines whether or not any document has been read by scanner unit 160 (step S100). If no original is read by scanner unit 160 (NO in step S100), the process returns to step S100.

  If any document is read by the scanner unit 160 (YES in step S100), the image processing unit 180 extracts a character area from the input image data generated by the reading (step S102).

  Further, the image processing unit 180 performs a character recognition process on the extracted character region, and generates OCR information including a character code (step S104). In parallel, the image processing unit 180 performs outline acquisition processing on the extracted character region to generate outline information (step S106).

  After completing the processing of steps S104 and S106 for all characters included in the extracted character area, the image processing unit 180 performs various types of information for each part constituting each character based on the OCR information and the outline information. Is generated (step S108).

  Subsequently, based on the character part information, the image processing unit 180 performs smoothing processing and / or edge enhancement processing to be performed on any character included in the character region or each part included in each character. The type is determined (step S110). Further, the image processing unit 180 executes the determined smoothing process and / or edge enhancement process on the target character area or the small area corresponding to the target part (step S112).

  Further, the image processing unit 180 determines whether or not smoothing processing and / or edge enhancement processing has been executed for all characters included in the extracted character region (step S114). If there is a character for which smoothing processing and / or edge enhancement processing has not been performed (NO in step S114), image processing unit 180 performs the process of step S112 again on the unprocessed character. Execute.

  On the other hand, if smoothing processing and / or edge enhancement processing has been executed for all characters included in the extracted character region (YES in step S114), image processing unit 180 performs post-processing image data. Is output to the printing unit 170 (step S116), and the printing unit 170 prints out the image data on a paper medium (step S118). Then, the process returns.

  If either one of the OCR information and outline information is not successfully acquired, the part of each character or the image processing method for each character may be determined based only on the acquired information. Good.

<Effects of the present embodiment>
According to the embodiment of the present invention, OCR information and outline information are acquired for each character in the character area included in the input image data, and based on these information, the attribute of each part constituting each character Or the attributes of each character are parsed. Based on these attributes, an appropriate image processing method is determined, and the determined image forming method is sequentially applied to the corresponding region of the input image data.

  By adopting such a configuration, it is possible to perform finer image processing as compared with a configuration in which an image processing method having a characteristic is uniformly applied to a character region.

  As a more specific example, even for characters with the same font and character size, but with a large number of strokes, by using a smaller image processing filter (such as a smoothing filter or an edge enhancement filter) High character reproducibility can be realized.

  As described above, according to the present embodiment, it is possible to maintain high character quality of a document in which characters of different character sizes and different fonts are mixed.

[Other embodiments]
A program executed by a control unit or the like according to the present embodiment is a program module provided as a part of a computer operating system (OS) and calls and processes necessary modules at a predetermined timing in a predetermined arrangement. It may be executed. In that case, the program itself does not include the module, and the process is executed in cooperation with the OS. A program that does not include such a module can also be included in the program according to the present invention.

  Furthermore, part or all of the functions realized by the program according to the present invention may be configured by dedicated hardware.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

It is a diagram showing a usage pattern of the MFP according to the embodiment of the present invention. It is a block diagram showing a functional configuration of the MFP according to the embodiment of the present invention. It is a figure which shows an example for comparing the influence of the stroke number of a character. It is a figure which shows another example for comparing the influence of the stroke number of a character. It is a figure which shows an example for comparing the influence of the flatness of a character. It is a figure which shows an example for comparing the influence of a character font. It is a figure for demonstrating the division | segmentation into the small area | region of the character according to this Embodiment. It is a figure which shows an example of the correspondence of the part which comprises a character, and a small area | region. It is a functional block diagram for implement | achieving the image processing according to embodiment of this invention. It is a figure which shows an example of the character information stored in character information DB shown in FIG. It is a figure which shows an example of OCR information and outline information according to this Embodiment. It is a figure which shows an example of the character parts information according to this Embodiment. It is a figure which shows an example of the smoothing filter group stored in smoothing filter DB shown in FIG. It is a figure which shows an example of the edge enhancement filter group stored in edge enhancement filter DB shown in FIG. It is a figure for demonstrating the effect by the image processing method according to this Embodiment. It is a flowchart which shows the image processing procedure according to embodiment of this invention.

Explanation of symbols

  10 character region extraction unit, 12 character recognition unit, 14 character information DB, 16 outline acquisition unit, 18 character part analysis unit, 20 smoothing processing unit, 22 smoothing filter DB, 24 edge enhancement processing unit, 26 edge enhancement filter DB , 32 OCR information, 34 outline information, 100 MFP, 110 control unit, 120 memory, 130 operation panel, 130a key input unit, 130b touch panel, 140 data storage unit, 150 network communication unit, 160 scanner unit, 170 print unit, 180 Image processing unit, 190 communication unit, BUS data bus, NW network.

Claims (9)

Character recognition means for recognizing a character from input image data and generating first information indicating a character code of the character and an attribute of a part constituting the character;
An outline acquisition unit that acquires an outline of a character from the input image data and generates second information including position information of the character;
Analyzing means for determining an image processing method according to parts constituting each character based on the first information and the second information;
An image processing apparatus comprising: an image processing unit that executes, for the input image data, a corresponding image process determined by the analysis unit for a region where each part constituting each character is located. Character recognition means for recognizing a character from input image data and generating first information indicating a character code of the character and an attribute of the character;
An outline acquisition unit that acquires an outline of a character from the input image data and generates second information including position information of the character;
Analyzing means for determining an image processing method corresponding to each character based on the first information and the second information;
An image processing apparatus comprising: an image processing unit that executes, for the input image data, a corresponding image process determined by the analysis unit with respect to a region where each character is located. It further comprises storage means for storing character information for a plurality of fonts,
The image processing apparatus according to claim 1, wherein the character recognition unit recognizes a font type of each character with reference to the storage unit. The character recognition means generates the first information including the font type, size, and number of strokes of each character,
The image processing apparatus according to claim 1, wherein the analysis unit determines an image processing method based on a font type, a size, and a number of strokes of each character included in the first information. The image processing means includes a smoothing process,
The image processing apparatus according to claim 1, wherein the analysis unit changes a shape of a smoothing filter used in the smoothing process. The image processing means includes edge enhancement processing means,
The image processing apparatus according to claim 1, wherein the analysis unit changes at least one of an edge enhancement amount and an edge enhancement method for a character portion in the edge enhancement processing unit. A print unit that prints out the input image data after image processing;
The image processing apparatus according to claim 1, wherein the analysis unit changes the image processing apparatus according to printing characteristics of the print unit.   8. The method according to claim 1, wherein the analysis unit determines an image processing method based on the generated information even if either the first information or the second information cannot be generated. The image processing apparatus according to item 1.   The image processing unit determines whether the image processing method to be executed is to be changed more than a predetermined change amount between an area where the adjacent part is located or between adjacent characters. The image processing apparatus according to claim 1, wherein the image processing apparatus suppresses the change.

Images