JP2006345314A - Image processing apparatus and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To more plainly edit image data obtained by scanning a paper document while maintaining the contents of the image data. <P>SOLUTION: An image processing method for processing image data obtained by scanning a document including a plurality of objects having respectively different attributes is provided with: a division process for dividing the image data in each block comprising an object having a different attribute; a character processing process for vectorizing an object whose attribute is judged as a character; and a character enlarging process (step S1603) for enlarging (step S1602) a character block composed of the object so as to inscribe the character block with a cell when the vectorized object is included in the cell constituting a table block out of blocks divided by the division process and enlarging the object included in the enlarged character block correspondingly to the enlarged character block. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a processing technique for image data generated by reading a paper document using a reading device such as a scanner.

  In recent years, with the digitization and networking of equipment in offices, copiers not only function as devices for creating copies of paper documents, but also digitize faxes and paper documents, and even via networks. It has also come to function as a device for sending all image data.

  Such a multifunctional copying machine is generally called an MFP (Multi Function Peripheral), and is becoming a central support tool for carrying out business dealing with paper documents in an office. Correspondingly, various functions have been arranged so that the MFP itself can easily handle a paper document as if it were image data.

For example, Patent Document 1 describes a technique for converting a scanned paper document into editable image data.
JP 2004-265384 A

  By using the image data converted by the technique described in Patent Document 1, the user edits the image data displayed on the touch panel of the MFP with one touch, and outputs the edited image data. The applicant considers that it is possible to easily obtain a paper document after editing.

  When converting a paper document into editable image data, the MFP executes processing in the following procedure. First, image data generated by scanning a paper document is image-region-separated and divided into rectangular regions (blocks) for each attribute such as text, table, photograph, and line drawing.

  Next, for each block object obtained by dividing the area, for example, if it is a text area (text block), OCR is performed and character recognition is performed. Then, the character object is cut out from the entire image and converted into an outline vector. In this way, by converting the outline vector, the character object can be handled as non-resolution-dependent image data whose image quality does not change even when enlarged or reduced.

  As for the table object, the pixels representing the table ruled line are cut out from the entire image, converted into outline vectors, converted into non-resolution-dependent image data, and further figure recognition is performed to recognize the table structure. Similarly, photographs, line drawings, etc. are cut out from the entire image and converted individually.

  In this way, it is possible to perform edit processing such as changing the layout of individual image data that has been cut out from the entire image and individually processed optimally, and this is output as a paper document. It is also possible to arbitrarily select the extracted image data and save / send it as image data.

  In this way, in an MFP to which an editing function for editing a scanned paper document is added, characters in the paper document can be arbitrarily changed in order to reissue the material to a person with low vision such as an elderly person. Realization of a so-called “original reissue function” that easily expands in size is desired.

  As a method of easily realizing the “original reissue function”, for example, a method of scanning a paper document and arbitrarily enlarging and outputting each object obtained by separating the image area on the touch panel can be considered. Thus, it can be said that the PC-less simple method performed on the MFP is very suitable for an elderly person who dislikes complicated operation of the PC. Furthermore, automating a series of tasks for enlarging and laying out text at an arbitrary size will be a more friendly function for people with reduced vision.

  However, in realizing the original reissue function as described above, for example, after recognizing an object (for example, a character object in a text area) cut out by image area separation, the recognized text code, etc. If the enlargement process is performed using this information, if there is an erroneous recognition in character recognition, the erroneous information may be enlarged. For this reason, when performing the enlargement process, it is desirable to perform an image-based enlargement process using shape information such as a character outline.

  When performing image-based enlargement processing using such object shape information, a rectangular area (block) containing an object to be enlarged is covered with a rectangular area containing another object on a paper document. It is considered desirable to do so so that there is no gap between the rectangular regions.

  However, in the case of such enlargement processing, if the objects to be enlarged are nested (in a state where rectangular regions including objects having different attributes overlap each other), the rectangular regions including the objects are originally Therefore, there is a problem that the object cannot be enlarged unless both objects are handled as a unit. This is because, when each object is handled separately, positional displacement between the objects may occur, and the content of the image data may change.

  In the case of an object such as a table, a table ruled line object and a character object are nested. For this reason, if the table ruled line object and the character object are enlarged at different magnifications, the contents of the table may change due to characters protruding from the table cell.

  For this reason, it is conceivable to enlarge the table ruled line object and the character object at the same enlargement ratio, but generally, the enlargement of the table ruled line object is often large to some extent in the document image. In many cases, the character object in the cell cannot be enlarged sufficiently if the character object in the cell is enlarged at the same enlargement ratio.

  SUMMARY An advantage of some aspects of the invention is that image data obtained by scanning a paper document is edited more easily while maintaining the content of the image data.

In order to achieve the above object, an image processing apparatus according to the present invention comprises the following arrangement. That is,
An image processing apparatus that processes image data obtained by scanning a document including a plurality of objects having different attributes,
Dividing means for dividing the image data into blocks each made up of objects having different attributes;
A character processing means for vectorizing an object whose attribute is determined to be a character by the dividing means;
When the object vectorized by the character processing means is located in a cell constituting a table block among the blocks divided by the dividing means, the character block consisting of the object is inscribed in the cell. Character enlargement means for enlarging the object included in the character block corresponding to the enlarged character block.

  According to the present invention, it is possible to edit image data obtained by scanning a paper document more easily while maintaining the content of the image data.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings as necessary.

[First Embodiment]
1. Configuration of Document Management System A first embodiment of the present invention will be described. FIG. 1 is a diagram showing a configuration of a document management system including an image processing apparatus (MFP 100) according to the first embodiment of the present invention. As shown in the figure, the document management system is realized in an environment in which an office 120 and an office 130 are connected via the Internet 104.

  Connected to the LAN 107 constructed in the office 120 are the MFP 100, the management PC 101 that controls the MFP 100, the client PC 102, the document management server 106-1, and the database 105-1 of the document management server 106-1, respectively. 103-1.

  Similarly, the document management server 106-2 and its database 105-2 are connected to the LAN 108 built in the office 130, respectively. The LAN 107 in the office 120 and the LAN 108 in the office 130 are connected to the Internet 104 via proxy servers 103-1 and 103-2.

  As described above, the image processing apparatus (MFP 100) according to the present embodiment can function as a device constituting the document management system. Specifically, the MFP 100 is in charge of part of the image processing for the image reading unit of the paper document and the read image data, and the image processed image data is input to the management PC 101 using the LAN 109. The management PC 101 is a normal PC, and has an image storage unit, an image processing unit, a display unit, and an input unit.

2. Configuration of Image Processing Apparatus (MFP 100) FIG. 2 is a diagram showing a functional configuration of the image processing apparatus (MFP 100) according to the first embodiment of the present invention. Hereinafter, FIG. 2 will be described with reference to FIG.

  In FIG. 2, an image reading unit 201 including an auto document feeder (hereinafter referred to as ADF) uses a lens to reflect an original reflection image obtained by irradiating a paper document (original image) with a light source (not shown) onto a solid-state image sensor. By forming an image, an image reading signal on the raster is obtained as an image signal having a density of 600 DPI from the solid-state imaging device.

  In the normal copying function, the image signal is processed into a recording signal by the data processing unit 206. In the case of a plurality of copies, recording data for one page is temporarily stored in the recording device 202 and then stored in the recording device 203. Output sequentially to form an image on paper.

  On the other hand, print data output from the client PC 102 is converted into raster data that can be recorded by the data processing device 206 from the LAN 107 via the network IF 205, and then an image is formed on the paper by the recording device 203.

  As described above, the MFP 100 according to the present embodiment has a copy / output function of a normal MFP, and also has the following functions as an edit / output function.

  For example, when a vector image display on the display / input device 207 including a large touch panel is instructed (the instruction by the user is given via the display / input device 207, the same applies hereinafter), the reading process in the image reading unit 201 The image signal read in is processed into an image signal that can be edited by the data processing device 206, converted into vector data, and then displayed on the display / input device 207.

  The vector data displayed on the display / input device 207 is subjected to desired editing processing by the data processing device 206 in accordance with an instruction from the user. Then, the edited vector data is displayed on the display / input device 207 again.

  Further, in accordance with an output instruction from the user, the data processing device 207 performs image processing of vector data into a recording signal, and sequentially outputs it to the recording device 203 to form an image on paper. In accordance with a transfer instruction from the user, the vector data is sent from the network IF 208 to the document management server 106-1, the client PC 102, the document management server 106-2, etc. via the LAN 107. The present invention relates to a processing technique for image data generated by reading a paper document using an image reading unit, and the processing outline will be described in detail below.

3. Processing in image processing apparatus
3.1 Overall Processing An overview of overall processing related to the editing / output function in the MFP 100 will be described with reference to FIG. As shown in FIG. 3, in step S301, first, the image reading unit 201 is operated to scan a single document in a raster shape and input as image information to obtain a 600 DPI-8-bit image signal.

  In step S <b> 302, the image signal is preprocessed by the data processing device 206 and stored in the storage device 202 as image data for one page. Then, the stored image data is divided into a character / line drawing part and a halftone image part, and the character part is further divided into blocks or a table composed of lines. In addition, each line drawing / drawing is separated and segmented. On the other hand, an image portion expressed in halftone is decomposed into independent objects for each desired block, such as an image portion (photograph) of a block separated into a rectangular area, a background portion, and the like.

  In step S303, the object of each block divided in step S302 is converted into vector data for each object. Conversion to vector data is performed as follows.

  First, for a block (character (TEXT) block) extracted as a character (TEXT), shape information independent of resolution is obtained by outlining the character shape obtained by binarization. On the other hand, as internal analysis information of the block, OCR (character recognition) and its text information, as well as character size, style and font are recognized, and reproducible font data is extracted from the image information obtained by scanning the document. To do.

  In addition, for the table (TABLE), line drawing (LINE), and drawing (PICTURE) block composed of lines, while extracting the graphics information independent of the resolution, the figure shape can be recognized. The shape information is acquired by the recognition process. In particular, a table can be recognized as a set of rectangles.

  Photo (PHOTO) blocks are processed as individual JPEG files as image data. As described above, each object can be handled individually by vectorization processing, and the layout can be freely changed and enlarged / reduced in units of objects.

  In step S304, the information obtained by the vectorization process in step S303 is stored as vector data in the storage device 202, and after laying out so as to reproduce one paper document by combining each object, the display / input device 207 is displayed. Display above.

  In step S304, the vector data displayed on the display / input device 207 can freely change the configuration, size, and arrangement of the object in accordance with a user input operation on the display / input device 207 ( Step S305). These user change operation results are displayed on the display / input device 207 at any time, and the layout can be changed interactively.

  In step S306, the vector data that has been subjected to user-desired editing processing in step S305 is output on paper or distributed to the document management servers 106-1 and 106-2, the client PC 102, and the like in accordance with a user instruction. . Details of each processing step will be described below.

3.2 Block selection (area division) processing (step S302)
The block selection process (step S302) is the image data shown in FIG. 4A (image data read in step S301) for each rectangular area (block) including each object as shown in FIG. 4B. Processing to recognize each block as a block of characters (TEXT) / drawing (PICTURE) / photograph (PHOTO) / line drawing (LINE) / table (TABLE), etc., and dividing the block into blocks having different attributes It is. A specific example of the block selection process will be described below.

  First, the image data is binarized into black and white, and contour tracking is performed to extract a block of pixels surrounded by a black pixel contour. For a black pixel block with a large area, the outline is also traced for the white pixels inside it, and the white pixel block is extracted. Extract the lump.

  The black pixel blocks thus obtained are classified by size and shape, and are classified into blocks having different attributes. For example, a pixel block corresponding to a character having an aspect ratio close to 1 and having a constant size is used as a character block, a portion where adjacent characters can be grouped in a well-aligned manner is a character (TEXT) block, and a flat pixel block is represented by a line. (LINE) block, the range occupied by black pixel blocks that are not less than a certain size and contain square-shaped white pixel blocks in a well-aligned manner (TALBLE) block, and an area where irregular pixel blocks are scattered (PHOTO) ) A block and a pixel block having an arbitrary shape other than the block are defined as a picture (PICTURE) block. In addition, an area that is not determined by any attribute is extracted as a background (BACKGROUND).

  An example of the block information for each block obtained by the block selection process is shown in FIG. As shown in the figure, the block information includes “attribute” and position information (“coordinate X”, “coordinate Y”) in the paper document, block shape (“width W”, “height H” for each block. ") And information about the presence or absence of" OCR information ". The information for each block is used in the vectorization process described below.

3.3 Vectorization processing (step S303)
A vectorization process performed on each block obtained in the block selection process (step S302) will be described with reference to FIG. Vectorization processing is processing that performs processing adaptively according to the attribute of each block, and realizes reusability of each block and converts it into high-compression and high-quality data.

  As shown in FIG. 6, among the blocks obtained by the block selection process (step S302), the TEXT block is binarized by the binarization processing unit 601, and the character recognition unit 602 for the extracted character object. Character recognition processing is performed at, and the text code of each character is extracted. Each character shape is outlined by the outline creation unit 603 and converted into data independent of the resolution expressed by straight lines and smooth curves.

  For the TABLE block, the binarization processing unit 604 binarizes and extracts a binary image of the table frame. The extracted binary image is outlined by the outline creation unit 605, and then the table processing unit 606. Table processing is performed at, and the table frame is expressed by ruled lines. In the binarization of the TABLE block, a binary image of only the table frame can be extracted by eliminating the TEXT block in the TABLE block extracted in the block selection process and binarizing.

  As for the LINE block, like the TEXT block and TABLE block, the binarization processing unit 607 binarizes the inside of the block, extracts the LINE binary image, outlines it with the outline creation unit 608, and expresses it with smooth curves and straight lines. Then, the figure recognition unit 609 recognizes the figure and extracts information such as ruled lines, circles, ellipses, and polygons.

  The PHOTO and BACKGROUND blocks are extracted as image information by the adaptive compression unit 610 and compressed.

  Hereinafter, each process will be described in detail. The binarization processing in the binarization processing units 601, 604, and 607 and the outline conversion processing in the outline creation units 603, 605, and 608 are the same processing.

3.3.1 Binarization processing The binarization processing units (601, 604, 607) extract luminance information from image data and create a histogram of the luminance values. A plurality of threshold values are set on the histogram, and an optimal threshold value is derived by analyzing the connection of black pixels on the binary image binarized with each threshold value, and a binary image based on the threshold values is obtained.

3.3.2 Character Recognition Processing The character recognition unit 602 recognizes an image cut out in units of characters using a pattern matching technique, and obtains a corresponding text code. This recognition process compares an observed feature vector obtained by converting features obtained from a character object into a numerical sequence of several tens of dimensions with a dictionary feature vector obtained in advance for each character type, and determines the character type with the closest distance. This is a process for obtaining a recognition result. There are various known methods for extracting a feature vector. For example, there is a method of dividing a character into meshes and using a mesh number-dimensional vector obtained by counting character lines in each mesh as line elements according to directions.

  When character recognition is performed on the TEXT block extracted in the block selection process (step S302), horizontal writing and vertical writing are first determined for the corresponding block, lines are cut out in the corresponding directions, and then the character objects are cut out. To get a character image. Horizontal / vertical writing can be determined by taking horizontal / vertical projections of the pixel values in the corresponding block, and determining horizontal writing blocks when the horizontal projection variance is large and vertical writing blocks when the vertical projection variance is large. .

  For horizontal writing, character strings and characters are decomposed by cutting out lines using horizontal projection, and then cutting out characters from the vertical projection of the cut lines. For a vertically written TEXT block, the horizontal and vertical may be reversed. At this time, the character size can be detected.

3.3.3 Outline Creation Processing The outline creation unit 603 extracts blocks in the block selection process (step S302) for blocks that have been made a drawing (PICTURE), line drawing (LINE), or table (TABLE) block. The contour of the pixel block is converted into outline data expressed by straight lines and curves.

  Specifically, a point sequence of pixels forming an outline is divided by points regarded as corners, and each section is approximated by a partial straight line or curve. An angle is a point where the curvature is maximized, and the point where the curvature is maximized is that a string is placed between k points Pi-k and Pi + k that are separated from the arbitrary point Pi as shown in FIG. It is obtained as a point where the distance between the string and Pi becomes maximum when drawn. Furthermore, let R be the chord length / arc length between Pi−k and Pi + k, and a point where the value of R is equal to or less than a threshold value can be regarded as a corner. Each section after being divided by the corners can be vectorized by using a least square method for a straight line and a curve using a cubic spline function.

  When the object has an inner contour, the white pixel contour point sequence extracted in the block selection process (step S302) is used to approximate the image by a partial straight line or a curve.

  FIG. 8 shows an example of image data before being outlined and vector data that has been outlined. FIG. 8A shows image data before being outlined, and FIG. 8B shows image data after being outlined. As described above, the character shape approximated by a straight line or a curve, a table frame, or a line drawing is vector data that does not lose image quality even when enlarged or reduced, that is, does not depend on resolution.

3.3.4 Table Processing The table processing unit 606 recognizes the cells in the table and their configurations, and expresses the table frame as a set of ruled lines. FIG. 9A shows vector data obtained by outlining a binary image. As a result of the outline, the outer contour and the inner contour are obtained. For example, the cell corner points 911 and 912 are obtained based on the structural relationship between the outer contour 901 and the inner contours 902, 903, and 904. As a result of obtaining the corner points, the corner point configuration is extracted as shown in FIG. 9B, and the cell configuration and ruled line information are obtained. Next, the position is adjusted so that the determined ruled lines pass between the contour lines in FIG. 9A, and the thickness is determined from the positional relationship with the contour lines.

  With the above processing, the table can be expressed by ruled lines having a thickness. In the process of recognizing the structure of the table from such a binary image, there is a possibility that an area painted black is extracted. An example is shown in region A of FIG. If such an area is extracted, it is originally a cell, but the cell in the block may not be extracted by the binarization threshold during the binarization processing in the binarization processing unit 604. By examining the information, or by performing binarization processing and outline creation processing again to extract cells, the table can be expressed more faithfully as a vector.

3.3.5 Graphic Recognition The graphic recognition unit 609 extracts parts that can be expressed in graphics such as ruled lines, circles, ellipses, and polygons on the binary image. An example of figure recognition will be described with reference to FIG. FIG. 11 shows image data obtained by converting a binary image into an outline. For each closed curve, circle, ellipse, and polygon information is extracted from the curvature and corner point information of the outline. A circle can be determined based on whether the curvature is constant, and an ellipse can be determined using curvature transition information. Polygons can be determined from the corner points and the curvature between the corner points. For example, in FIG. 11, graphic information such as a circle and a rectangle is extracted.

  Next, the relationship between the extracted figure outline and surrounding outlines is examined. As an example, the circle outline 1103 will be described in detail. From the relationship between the curvature between the corner points 1111 to 1116 of the outer outline outline 1104 of the circle outline 1103 and the distance between the corner points, the outline 1101 is graphically expressed as a circular curve having a thickness. The

  At this time, the part which does not comprise the circle with the corner points 1111 to 1112, 1113 to 1114, and 1115 to 1116 is divided as an outline. In addition, if the binary image is thinned in advance, it is possible to efficiently extract the coupling portions 1121 to 1123 of the circle and the line. Finally, a straight line portion of each outline is detected, and it is determined whether there is a straight line paired with the extracted straight line. If there is a straight line, the line is replaced with a ruled line having a thickness.

  As described above, the figure shape is extracted from the outline. In addition, what was not recognized as a figure is described with an outline as it is.

3.3.6 Adaptive Compression The adaptive compression unit 610 processes, as raster data, individual JPEG files for blocks and BACKGROUND (background) determined as PHOTO blocks by block selection processing (step S302). At this time, the background means a portion which is not necessary in the paper document, and in some cases, image data is created with low resolution or high compression. This makes it possible to reduce the amount of data when finally creating vector data as application data.

3.4 Vector data display process (step S304)
As described above, the image data for one page is subjected to the block selection process (step S302) and the vectorization process (step S303), thereby generating an intermediate data format file as shown in FIG. Such a data format is called a document analysis output format (DAOF). FIG. 12 shows an example of the data structure of DAOF.

  In FIG. 12, reference numeral 1201 denotes a header, which holds information related to image data to be processed. In the layout description data section 1202, the attribute and position of each block recognized for each attribute such as TEXT (character), LINE (line drawing), PICTURE (drawing), TABLE (table), PHOTO (photo), etc. in the image data. Keep information.

  The character recognition description data portion 1203 holds a character recognition result obtained by character recognition of the TEXT block. The table description data portion 1204 stores details of the structure of the TABLE block. The image description data portion 1205 cuts out image data of blocks such as PICTURE and LINE from the image data of the paper document and holds them.

  On the other hand, a document structure tree in a paper document is created from block information. Here, the document structure tree will be described with reference to FIG. FIG. 13A is a diagram showing an example of each block obtained from the block information, each attribute, and a grouping block. About each block, the relationship between blocks, such as the distance between blocks being near, and the block width being substantially the same, is determined and grouped. For example, a grouping block V1 is generated from T3, T4, and T5 due to the relationship between the blocks, and a grouping block V2 is generated from T6 and T7. By repeating such grouping, FIGS. ) Is created.

  Note that V0 is the highest layer representing the entire page. Incidentally, the actual data of each block of the document structure tree is stored in the DAOF and is associated with the document structure tree. For example, in the case of a TEXT block, various types of images can be generated by flowing outline data, text code, or DAOF actual data into a document structure tree. In step S304, the generated vector data is displayed. The TEXT block displays a character shape outline in order to avoid image loss due to erroneous recognition of characters.

3.5 Vector editing process (step S305)
In step S305, the displayed vector data is edited according to a user instruction. In the present embodiment, the function of editing vector data in a TEXT block intended for a blind person is described, but the object of the present invention is not particularly limited to this. The editing process in step S305 can be performed by a user's desired editing process such as changing the layout of each block or changing internal data of the block. Editing of character object vector data described below is an example. .

  The vector data is edited by changing the DAOF data when editing the block, and by changing the document structure tree and DAOF when changing the layout.

  In addition, for example, switching of information inside the block such as a character shape and a text code in the TEXT block can be performed by switching of DAOF data flowing into the document structure tree.

  The TEXT block editing function will be described below. FIG. 14 is a diagram illustrating an example of a UI for realizing simple character enlargement for a predetermined block in a paper document.

  In the figure, reference numeral 1401 denotes a scanned document display unit, which automatically displays vector data scanned and vectorized. The vectorized data is displayed so that the attribute obtained by the block selection process can be seen for each block, and can be selected for each block on the UI.

  Here, when the “enlarge character” button 1402 is pressed, the selected block is displayed again in an enlarged layout. When the user determines that the display result is appropriate, the user presses an “output” button 1404 to execute output processing. If it is determined that the desired result is not obtained, a “return” button 1403 is pressed. As a result, the enlargement operation can be canceled and returned to the original state.

  Here, the layout method by character enlargement will be described in detail. FIG. 15A shows a vectorized paper document used in the description of this embodiment. In order to simplify the following description, the character shape to be enlarged is indicated by a dotted rectangle, but it is assumed that the outlined character shape is contained inside the dotted line. FIG. 15B shows information obtained by subjecting FIG. 15A to block selection processing. In the figure, reference numerals 1511 and 1512 denote blocks (solid line rectangles) determined to be TABLE, and the other blocks are blocks (dotted line rectangle) determined to be TEXT.

  FIG. 16 is a flowchart for creating vector data obtained by enlarging characters in a paper document.

  In step S1601, the character (TEXT) / table (TABLE) block is enlarged. In the enlargement of the character / table block in step S1601, the blocks overlap. For example, for the character blocks in the table block, the outer table block and the inner character object are considered as a set, and the outer table block is enlarged. Map within the table block with the same magnification. Therefore, in the enlargement of the character / table block, the enlarged layout of the block shown by the thick frame in FIG. 17A among the objects in FIG. 15B is considered.

  Next, an enlargement method will be described. The aspect ratio of each block is kept constant, the block is expanded, and when the block overlaps another block or another object that is not a text block, the expansion ends, and the region is set as the block expansion region. As a result, each area in FIG. 17A is enlarged as shown in FIG. At this time, the position of the frame of the block may be slightly shifted as long as the positional relationship with other peripheral blocks and objects does not change. It should be noted that the character shape of each character object, the ruled line of the table object, and the like are enlarged by linear transformation as the block is enlarged.

  In step S1602, the character block is enlarged. In the enlargement of the character block in step S1602, only the frame of each character block obtained by the block selection process is enlarged, and the character shape inside the block is not enlarged in accordance with the enlargement of the frame. In the enlargement of the character block frame, the aspect ratio is not constant, and it is determined whether the four sides of the block can be enlarged. At this time, if it is a character block in the table object, since the cell configuration and the faithful position of the table ruled line are extracted in step S606, the area of each cell whose table frame is expanded in step S1601 is the character block as it is. It becomes an enlarged area of the frame. As a result, the result shown in FIG. 18 is obtained for the paper document shown in FIG.

  In step S1603, the character shape in each character block enlarged in step S1602 is enlarged. Here, each character shape is cut out in character units by the character recognition processing in step S602, and each character can be handled as a character string. The enlargement of the character shape will be described with reference to FIG.

  FIG. 19A shows the character block frame and the character shape in the character block in step S1602. In FIG. 19B, reference numeral 1901 shows that the character width of the character block frame (FIG. 19A) obtained in step S1602 is enlarged by Δα. By expanding Δα, the leftmost character 2300 protrudes from the character block frame, but by treating it as a character string, the protruding character shape as shown in FIG. 19C can be reflected in the next line. . When creating a new line by shifting to the next line, the line spacing is determined according to the character size. The line spacing may be changed according to the character size. If the next line extends beyond the character block frame due to the addition to the next line, the line is added again to the next line. The above expansion of Δα is repeated as long as the character string fits within the character block frame.

  By executing the above processing, a character enlarged image as shown in FIG. 20 can be generated from the paper document shown in FIG.

  In this embodiment, an example of “character enlargement” that enlarges a character to the maximum with one touch of the “character enlargement” button 1402 has been described. However, the present invention is not limited to this. For example, a “character enlargement” button It is also possible to previously set each enlargement ratio in steps S1601 to S1603 in one touch of 1402, and obtain a desired result by touching the user a plurality of times.

3.6 Vector Output In step S306, vector data is output. Vector data is temporarily converted into application data when it is output. Application data can be generated using the DAOF created in step S304 and the document tree structure. Note that the DAOF and the document tree structure are changed by the vector editing process in step S305.

  The application data generation process for the document tree structure in the case of FIG. 13 will be described. In FIG. 18, since there are two blocks T1 and T2 in the horizontal direction, H1 has two columns, and after T1 internal information (refer to DAOF, text of character recognition result, image, etc.) is output, the column is changed, and T2 The internal information is output, and then S1 is output. Since H2 has two blocks V1 and V2 in the horizontal direction, it outputs as two columns, V1 outputs its internal information in the order of T3, T4, T5, then changes the column, and outputs the internal information of T6, T7 of V2 To do. The application data is generated as described above, and the application data is output on a paper surface or output on a network.

  As is clear from the above description, according to the present embodiment, the table object and each cell ruled line position information is extracted from the paper document, and each character shape is expanded without breaking the layout in the paper document. It is possible to greatly enlarge the characters in the table in the paper document that originally has many small characters. Since the character shape is outlined by vectorization processing, it does not depend on the resolution and can be enlarged without loss of image compared to image-based enlargement.

  Although the characters in the table can be enlarged greatly, if multiple character objects existing in the table are enlarged at different magnifications, there is a possibility that the image will be damaged as a whole. Since the configuration is recognized in advance, when the characters are enlarged, various ideas such as making the enlargement ratios of all the characters in the table uniform are possible.

[Second Embodiment]
In the first embodiment, the enlargement magnification is maximized, but the present invention is not particularly limited to this. In the table, even if the characters are enlarged at the same magnification in the same cell, if the magnification is different for each cell, it is often difficult to see the characters in the paper document. Therefore, when enlarging characters, the magnifications may be set for all characters in the table extracted by the block selection process. In addition, since there is no problem in making the magnifications uniform for each item, means for setting for each column or row may be provided.

[Third Embodiment]
In the first embodiment, an outline character shape is used as an enlarged layout of characters, but the present invention is not particularly limited to this. The outlined character shape is used because image loss due to erroneous character recognition is taken into account. If the accuracy of the character recognition unit 602 and the accuracy of font recognition are sufficient, the character code that recognizes the character shape is used. It may be replaced with font information.

  Note that the font recognition processing is performed after character recognition in the character recognition unit 602, and a plurality of dictionary feature vectors corresponding to the number of character types used for character recognition are prepared for the character shape type, that is, the font type. The font type may be output together with the character code at the time of matching. Thereby, the recognition process of a character font is realizable.

[Other Embodiments]
Note that the present invention can be applied to a system including a plurality of devices (for example, a host computer, an interface device, a reader, and a printer), and a device (for example, a copying machine and a facsimile device) including a single device. You may apply to.

  Another object of the present invention is to supply a storage medium storing software program codes for implementing the functions of the above-described embodiments to a system or apparatus, and the computer (or CPU or MPU) of the system or apparatus stores the storage medium. Needless to say, this can also be achieved by reading and executing the program code stored in the.

  In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the storage medium storing the program code constitutes the present invention.

  As a storage medium for supplying the program code, for example, a floppy (registered trademark) disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD-R, magnetic tape, nonvolatile memory card, ROM, or the like is used. be able to.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an OS (operating system) operating on the computer based on the instruction of the program code. It goes without saying that a case where the function of the above-described embodiment is realized by performing part or all of the actual processing and the processing is included.

  Further, after the program code read from the storage medium is written into a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing.

1 is a block diagram illustrating a configuration example of a document management system including an image processing apparatus (MFP 100) according to a first embodiment of the present invention. 1 is a diagram illustrating a functional configuration of an image processing apparatus (MFP 100) according to a first embodiment of the present invention. FIG. 2 is a diagram showing an overview of overall processing in MFP 100. FIG. It is a figure for demonstrating the outline | summary of a block selection process. It is a figure which shows an example of the block information with respect to each block obtained by the block selection process. It is a flowchart which shows the flow of the vectorization process performed with respect to each block obtained by the block selection process (step S302). It is a figure for demonstrating the outline | summary of the process in an outline preparation part. It is a figure which shows the example of the image data before making it outline, and the vector data made outline. It is a figure for demonstrating the outline | summary of the process in a table process part. It is a figure for demonstrating the outline | summary of the process in a table process part. It is a figure for demonstrating the outline | summary of the process in a figure recognition part. It is a figure which shows an example of the data structure of DAOF. It is a figure for demonstrating a document structure tree. FIG. 3 is a diagram illustrating a UI showing an example of realizing simple character enlargement in a paper document. It is a figure for demonstrating the layout method by character expansion. 6 is a flowchart for creating a vector image in which characters in a paper document are enlarged. FIG. 6 is a diagram illustrating an example when an object area in a document is enlarged. It is a figure which shows an example in the case of enlarging a character area using a cell ruled line. It is a figure which shows an example in the case of carrying out enlarged mapping of each character within a character object area | region. It is a figure which shows an example at the time of performing a character expansion layout.

Claims (12)

An image processing apparatus that processes image data obtained by scanning a document including a plurality of objects having different attributes,
Dividing means for dividing the image data into blocks each made up of objects having different attributes;
A character processing means for vectorizing an object whose attribute is determined to be a character by the dividing means;
When the object vectorized by the character processing means is located in a cell constituting a table block among the blocks divided by the dividing means, the character block consisting of the object is inscribed in the cell. An image processing apparatus, comprising: a character enlargement unit that enlarges the object included in the character block in correspondence with the enlarged character block. Table processing means for vectorizing an object whose attribute is determined to be a table by the dividing means;
The table block enlarging means for enlarging the object by enlarging the table block made of the object vectorized by the table processing means until it circumscribes another block. The image processing apparatus described. The character block enlarging means for enlarging the object by enlarging the character block consisting of the object vectorized by the character processing means until it circumscribes another block. Image processing device. The character block enlarging means is
When the object vectorized by the character processing means is located in a cell constituting a table block among the blocks divided by the dividing means, according to the enlargement by the table block enlargement means, The image processing apparatus according to claim 3, wherein the object is enlarged. The character block enlarging means enlarges the character object by determining a character line folding position within a range that does not protrude from the character block of the character object included in the enlarged character block. The image processing apparatus according to claim 1. An image processing method in an image processing apparatus for processing image data obtained by scanning a document including a plurality of objects having different attributes,
A division step of dividing the image data into blocks each made up of objects having different attributes;
A character processing step of vectorizing an object whose attribute is determined to be a character by the dividing step;
When the object vectorized by the character processing step is located in a cell constituting a table block among the blocks divided by the dividing step, the character block made up of the object is inscribed in the cell. And a character enlargement step of enlarging the object included in the character block corresponding to the enlarged character block. A table processing step of vectorizing an object whose attribute is determined to be a table by the dividing step;
The table block enlarging step of enlarging the object by enlarging the table block made of the object vectorized by the table processing step until it circumscribes the other block. The image processing method as described. The character block enlargement step according to claim 7, further comprising a character block enlargement step of enlarging the object by enlarging a character block made up of the object vectorized by the character processing step until it circumscribes another block. Image processing method. The character block expansion step includes
When the object vectorized by the character processing step is located in a cell constituting a table block among the blocks divided by the division step, according to the enlargement by the table block enlargement step, The image processing method according to claim 8, wherein the object is enlarged. In the character block enlargement step, the character object included in the enlarged character block is enlarged by determining a wrapping position of the character line within a range that does not protrude from the character block. The image processing method according to claim 6. A storage medium storing a control program for realizing the image processing method according to claim 6 by a computer. A control program for realizing the image processing method according to claim 6 by a computer.

Images