JP2010068381A - Document reading method and document reading system - Google Patents

Abstract

When generating image data of an entire document to be read, divided and scanned image data can be easily combined.
When scanning a document to be read 31 in a plurality of times, the document to be read is fixed to a cover sheet having position information indicating a position on a transparent sheet in a dropout color. The composite image including the dropout color and the original image not including the dropout color corresponding to the composite image are read in a plurality of times. Based on the position information extracted from the composite image, the character recognition unit 12 specifies the same position as that of the composite image as the position of the document image read so as not to include the dropout color. Thereafter, the data combining unit 13 arranges and combines the plurality of document images at the specified positions, and generates the entire image data of the reading target document 31.
[Selection] Figure 1

Description

  The present invention relates to a document reading method and a document reading system that scan a reading target document in a plurality of times and generate image data of the entire reading target document.

Along with the paperless trend, there are often attempts to digitize materials that exist as paper documents. However, the size of paper documents varies, and especially for documents that exceed the size that can be read by the scanner, after the user divides and reads them, the image is processed into a single image. It is necessary separately. Thus, for example, Patent Document 1 discloses a method of scanning in a divided manner when scanning a document original exceeding the maximum scanning range of the scanner.
Japanese Patent Laid-Open No. 2008-60666

  In order to obtain image data by scanning a document that is larger than the maximum scanning range of the scanner, the document is folded in multiple times, for example, and the images are manually joined together. Since this method involves manual work, the joint portion of the image tends to shift, and labor is required to specify the joint point. In addition, it was necessary to manually check whether the necessary images had been taken.

  Also, depending on the type and importance of the document, the document may not be folded. In addition, the method of marking a document or pasting paper makes it easy to identify the coupling point, but depending on the document, it may not be possible to describe the mark or attach paper. . And when paper is affixed for the purpose of specifying the position, there is a risk of peeling off during the work.

  In the method disclosed in Patent Document 1, it is necessary to explicitly select which part of the document to scan, and the combined part of the image is substantially manually specified. For this reason, it requires the same effort as specifying a coupling point manually.

  In view of the above, the present invention makes it possible to easily combine divided and scanned image data when it is necessary to scan the document to be read a plurality of times in order to generate the entire image data of the document to be read. The purpose is to do.

  In order to solve the above-described problems, the present invention provides a document reading method in which a scan target document is divided into a plurality of small parts and scanned, and image data of the entire scan target document is generated. Position information indicating each position of the portion is described in a dropout color, and a cover sheet that covers the entire document to be read is overlaid on the document to be read and fixed to each other, and fixed to the cover sheet A divided scanning step of reading, by the scanner, a composite image including the dropout color and a document image not including the dropout color corresponding to the composite image for each of the small portions of the document to be read; The position information is extracted from the composite image with respect to the position information extracted from the composite image. Accordingly, a position specifying step for specifying the same position as the composite image as a position of the original image read by the scanner so as not to include the dropout color, and a plurality of the original images of the small parts are respectively arranged at the specified positions. An image combining step of combining the plurality of small portions to generate image data of the entire document to be read.

  According to the present invention, in a document reading system that scans a document to be read divided into a plurality of small parts and generates image data of the entire document to be read, each position of the small parts is indicated on a transparent sheet. A cover sheet in which position information is described in a dropout color and covers the entire document to be read; a scanner capable of reading an image including the dropout color and an image not including the dropout color; and the cover sheet The position information is extracted from the composite image read together with the dropout color by the scanner for each of the small portions of the document to be read that are superimposed and fixed on the basis of the position information extracted from the composite image. The same small part as the composite image is subjected to the dropout color by the scanner. A position specifying unit for specifying the position of the document image read so as not to overlap, and the plurality of small parts of the document image are respectively arranged at the specified positions and the plurality of small parts are combined to form the entire document to be read And a data combining unit for generating the image data.

  According to the present invention, when it is necessary to scan the document to be read a plurality of times in order to generate the entire image data of the document to be read, the divided and scanned image data can be easily combined.

  An embodiment of a divided scanning method according to the present invention will be described with reference to the drawings. In addition, the same code | symbol is attached | subjected to the same or similar structure, and the overlapping description is abbreviate | omitted. Further, this embodiment is merely an example, and the present invention is not limited to this.

  FIG. 1 is a block diagram of a drawing reading system in an embodiment of a divided scanning method according to the present invention.

  The drawing reading system according to the present embodiment scans the reading target document 31 by dividing it into a plurality of small parts, and creates the entire image data of the reading target document. The reading target document 31 is fixed to the cover sheet 32 and scanned.

  The drawing reading system includes a main body 10, an operation screen 11, a temporary storage unit 18, and a scanner 19. The main body 10 includes a character recognition unit 12, a data combination unit 13, a file creation unit 14, a user IF creation unit 15, a scanner control unit 16, and an event control unit 17.

  The operation screen 11 is an interface between the drawing reading system and the user. The user can give an instruction to the drawing reading system by operating the user interface displayed on the operation screen 11. The operation screen 11 is a display screen having a touch panel, for example. The operation screen 11 may be a combination of a display device such as a liquid crystal display and input means such as a mouse for moving and clicking a pointer displayed on the display device.

  The user IF creation unit 15 creates an image to be displayed on the operation screen 11. The event control unit 17 controls each event.

  The scanner 19 includes a reading window 23, reads an image of a document placed on the reading window 23, and outputs it as image data. The scanner 19 includes, for example, two types of light sources, and is formed so as to be able to read an image including a dropout color and an image not including the dropout color. A dropout color is a color that does not appear as an image when read by a scanner. For example, when the document to be read 31 is a monochrome monochrome image, the dropout color can be green. The scanner control unit 16 controls the scanner 19 and acquires image data output from the scanner 19.

  The file creation unit 14 creates a temporary file of scanned image data. This temporary file is stored in the temporary storage unit 18. The temporary storage unit 18 is an area on the hard disk, for example. The data combining unit 13 combines a plurality of image data read by the scanner 19. The character recognition unit 12 recognizes a character in the image data and specifies the position of the character in the image.

  FIG. 2 is a flowchart in the present embodiment.

  In the divided scanning method of the present embodiment, the preparation step S1, the divided scanning step S2, the position specifying step S3, and the image combining step S4 are sequentially performed. In the preparation step S1, the user prepares for the divided scan.

  FIG. 3 is a perspective view showing the cover sheet in the present embodiment together with the document to be read. FIG. 4 is a partially enlarged plan view of the cover sheet in the present embodiment. FIG. 5 is a plan view of a state in which the document to be read is sandwiched and fixed on the cover sheet in the present embodiment. FIG. 6 is a cross-sectional view showing a state in which the document to be read is sandwiched and fixed on the cover sheet according to the present embodiment.

  The user fixes the reading target document 31 to be scanned to the cover sheet 32 as preparation for divided scanning. The cover sheet 32 is larger than the reading target document 31 so that the entire reading target document 31 can be covered.

  The cover sheet 32 is obtained by laminating transparent sheets 33 and 34 in two layers and bonding one side 35. One transparent sheet 34 of the cover sheet 32 has grid-like position information 37 written in a dropout color. The position information 37 is a grid 39 drawn with a constant interval width, and an identifier 38 representing a character-recognizable position written in each grid. As the identifier 38, “A1”, “B2” or the like in which the row position is written in alphabet and the column position is written in numbers is used. Based on this position information 37, each position of a small portion of the document to be read 31 to be divided and scanned is specified.

  The user moves the reading target document 31 in the direction of the arrow 40, arranges it between the two transparent sheets 33 and 34 of the cover sheet 32, and closes the transparent sheets 33 and 34, thereby reading the reading target document. 31 is fixed to the cover sheet 32. A fixing tool 36 is provided on the side facing the side 35 to which the cover sheet 32 is bonded. The user fixes the side opposite to the side 35 to which the cover sheet 32 is bonded using the fixing tool 36 so that the document to be read 31 and the cover sheet 32 do not shift. Here, the fixing tool 36 is used, but other methods may be used as long as the relative positions of the document to be read 31 and the cover sheet 32 are not shifted, such as the cover sheet 32 having adhesiveness.

  FIG. 7 is a flowchart of the divided scanning process in the present embodiment.

  In the divided scanning step S2, the drawing reading system scans the reading target document 31 in a plurality of times according to a user instruction. First, the user places the document to be read 31 sandwiched between the cover sheets 32 on the document table of the scanner 19 (step S21). Since the reading target document 31 is larger than the reading window 23 of the scanner 19, a part of the reading target document 31 can be read from the reading window 23.

  Thereafter, the user instructs the drawing reading system to read the drawing by touching the operation screen 11 (step S22). When drawing reading is instructed, the scanner control unit 16 of the drawing reading system first causes the scanner 19 to acquire image data including a dropout color (step S23). Immediately thereafter, the scanner control unit 16 changes the wavelength (color) of the light source and acquires image data that does not include the dropout color (step S24). Thereby, the drawing reading system can read image data including the dropout color and image data not including the dropout color for the same area of the document to be read 31. Here, image data including the dropout color is referred to as a composite image, and image data not including the dropout color is referred to as an original image. Note that the original image may be read before the composite image.

  The composite image and the document image acquired for the same area of the document to be read 31 are associated with each other by the file creation unit 14 and stored as a temporary file in the temporary storage unit 18 (step S25). The composite image and the document image are associated with each other by, for example, a file name. For association by file name, for example, the file name of the composite image is <system time> _ <serial number> _D.bmp, and the file name of the original image is <system time> _ <serial number> _N.bmp. Good.

  When the composite image and the original image are acquired for a certain area of the document 31 to be read and the temporary file is saved, the drawing reading system displays the image created by the user IF creation unit 15 on the operation screen 11. The user is notified of the completion of drawing reading (step S26). When the drawing reading completion is notified, the user determines whether or not the entire scanning of the reading target document 31 is completed (step S27). When the entire scan of the reading target document 31 is not completed, the user moves the reading target document 31 with respect to the reading window of the scanner 19 (step S28), and instructs the drawing reading again (step S22). When the entire scan of the reading target document 31 is completed, the process proceeds to the position specifying step S3. In this way, the reading target document 31 is divided into a plurality of small portions and scanned, and the steps S22 to S27 are repeated until these small portions cover the entire reading target document 31.

  FIG. 8 is a flowchart of the position specifying step in the present embodiment.

  In the position specifying step S3, the position of each small portion scanned by the character recognition unit 12 of the drawing reading system is acquired. First, the user instructs the drawing reading system to combine drawings using the mouse 22 (step S31). The character recognizing unit 12 recognizes the identifier 38 indicating the position in the lattice included in the composite image (step S32), and specifies which position in the entire drawing the composite image corresponds to (step S33). Next, it is determined whether or not the specification of the positions of all the composite images has been completed (step S34), and steps S32 and S33 are repeated until completion.

  FIG. 9 is an enlarged plan view of the position information described on the cover sheet in the present embodiment.

  For character recognition, for example, when an image data file is given, an OCR control package that returns a recognized character string for each group of character strings and the position of the character string can be used. In this case, for example, when the image data file shown in FIG. 9 is given, the package recognizes the character string “B2” and returns the coordinates of the layout area 41. Since “B2” is one of the identifiers 38, it can be seen that this character string is the identifier 38.

  The coordinates of the layout area 41 are expressed by the upper left coordinates (xs, ys), the width w of the layout area 41, and the height h. The position of the identifier 38 is the center of the layout area 41, for example. That is, the position of the identifier 38 “B2” is (xs + w / 2, ys + h / 2). Using the position of the identifier 38 thus obtained on the composite image, the position of the composite image and the corresponding original image on the reading target document is specified.

  The drawing reading system stores the position of each identifier 38 described on the cover sheet 32 on the cover sheet 32 for each large paper size. Here, it is assumed that the paper size is larger than the range that can be scanned at once by the reading window 23 of the scanner 19.

  Therefore, the original images are combined by moving all the original images by a vector from the position of the identifier 38 on the composite image corresponding to each original image to the position on the cover sheet 32 of the identifier 38. Thus, image data corresponding to the document to be read 31 can be generated. This vector is called a mapping vector, and the mapping vector is calculated by the character recognition unit 12. That is, the character recognition unit 12 functions as a position specifying unit that extracts position information from the composite image and specifies the position of the document image corresponding to the composite image based on the position information.

  FIG. 10 is a diagram schematically showing the correspondence between the composite image and the cover sheet in the present embodiment. FIG. 11 is a diagram schematically showing a mapping vector in the present embodiment.

  FIG. 10 shows a case where four identifiers 38 of “J2”, “D4”, “G6”, and “Z9” can be recognized on the composite image 43. In FIG. 10, the positions of these identifiers 38 on the composite image 43 are represented as “j2”, “d4”, “g6”, and “z9”.

  The drawing reading system stores the position 45 of each identifier 38 in the document arrangement area 44 in which the position information 37 of the cover sheet 32 is described. Based on the position information, the drawing reading system calculates V1, V2, V3, and V4 for each vector from the position 42 of the identifier 38 on the composite image 43 to the position 45 of the identifier 38 in the document arrangement area 44. To do.

  If character recognition is performed on one composite image and one or more identifiers 38 can be recognized, the position of the composite image 43 can be determined. Further, if a plurality of identifiers 38 can be recognized in one composite image 43, the position determination accuracy of the composite image 43 can be improved.

  For example, when the read target document 31 includes the same character string as the identifier 38, the position of the composite image 43 may be erroneously determined. However, if three or more identifiers 38 can be recognized from one composite image, only one mapping vector is different, so that a misrecognized character string can be excluded from position determination.

  In the example of FIG. 10, the vector V4 from the position 42 of the identifier 38 “Z9” recognized on the composite image 43 to the position 45 of the identifier 38 “Z9” in the document arrangement area 44 is the other vector V1, Significantly different for V2 and V3. Therefore, this vector V4 is not used for calculating the mapping vector, and the average vector V of the vectors V1, V2 and V3 is used as the mapping vector. Whether a vector is significantly different from other vectors, for example, the difference vector from the average of other vectors, or the length of the longest vector of differences from other vectors is longer than the predetermined length. It can be determined by whether or not it is larger.

  Next, the position of the composite image 43 in the document placement area 44 can be specified by moving the composite image 43 by the mapping vector V thus obtained. Such character recognition is repeated for the number of composite images 43, and the positions of all composite images and document images are determined. When character recognition is impossible and the position cannot be specified, the composite image is ignored.

  When the composite image 43 and the original image are rotated relative to the cover sheet 32, a vector from the identifier 38 recognized on the composite image 43 to the position of the identifier 38 on the cover sheet 32 is used. The rotation can be corrected. With this correction, even when the cover sheet 32 is tilted together with the document to be read 31 sandwiched between the images when scanning an image, it can be mapped to the correct position.

  In the image combining step S4, the data combining unit 13 of the drawing reading system combines a plurality of document images generated by a plurality of scans. The drawing reading system arranges and combines all the original images at the specified position in the document arrangement area 44. For the arrangement of the document image, a mapping vector of the composite image 43 corresponding to the document image is used.

  In the area where the document images overlap, for example, a document image having a large number of recognized identifiers 38 is preferentially used. Alternatively, an average of a plurality of document images may be taken in a region where document images overlap. In this way, image data in which all document images are combined becomes image data of the entire document to be read 31.

  The drawing reading system notifies the user of the completion of drawing combining after completion of combining, for example, stores the entire image data as an electronic file, or outputs it to another device or the like. If the document image is insufficient as a result of the combination, the user is notified of a warning that the document image is insufficient.

  As described above, when the drawing reading system according to the present embodiment is used, when the user creates image data of a document that is larger than the maximum scanning range of the scanner, the user manually combines the scanned image data after scanning in multiple times. There is no need. Further, the user does not need to specify the reading position. For this reason, the entire image data of the document to be read can be easily created, and the labor required for work and the shift at the time of joining are suppressed. Another advantage is that it does not depend on the size of the scanner.

  Moreover, even if the reading target document is scanned without being aware of the reading order and the reading position, the drawing reading system can automatically create the entire image data. In this respect as well, the burden on the user is reduced. Further, whether or not the entire scan of the document to be read has been completed can be determined by the drawing reading system. As long as the cover sheet 32 having a corresponding size can be created, any sheet of paper such as a design drawing, a layout drawing, a wiring diagram, and a newspaper can be divided and read and automatically combined.

  In order to determine whether or not scanning of the entire document to be read 31 has been completed, the drawing reading system may display a region where an image has been acquired on the operation screen 11. As a result, reading omission can be prevented, and scanning can be performed efficiently. Further, by inputting the size of the document to be read 31 in advance into the drawing reading system, it is possible to cause the drawing reading system to determine whether or not the reading of the entire document to be read 31 has been completed.

  And the positional information described in the cover sheet 32 may be other than a character, and if the position can be specified, the lattice 39 may not be described. Further, the dropout color is not limited to the visible light wavelength range. For example, as the dropout color, a “color” that reflects light having a wavelength in the ultraviolet region other than the visible light wavelength range may be used. In this case, the identifier described in the dropout color can be read by using a light source that emits light of the ultraviolet wavelength as the light source of the scanner. In this way, by setting the “color” outside the visible light wavelength range as a dropout color, it is also possible to read a color reading target document such as a photograph.

1 is a block diagram of a drawing reading system in an embodiment of a divided scanning method according to the present invention. It is a flowchart in one Embodiment of the division | segmentation scanning method which concerns on this invention. It is a perspective view which shows the cover sheet in one Embodiment of the division | segmentation scanning method which concerns on this invention with the original to be read. It is a partial enlarged plan view of the cover sheet in one embodiment of the divided scanning method according to the present invention. FIG. 5 is a plan view of a state in which a document to be read is sandwiched and fixed on a cover sheet in an embodiment of the divided scanning method according to the present invention. FIG. 6 is a cross-sectional view of a state in which a document to be read is sandwiched and fixed to a cover sheet in an embodiment of a divided scanning method according to the present invention. It is a flowchart of the division | segmentation scanning process in one Embodiment of the division | segmentation scanning method which concerns on this invention. It is a flowchart of the position specific process in one Embodiment of the division | segmentation scanning method which concerns on this invention. It is an enlarged plan view of the positional information described on the cover sheet in one embodiment of the divided scanning method according to the present invention. It is a figure which shows typically the correspondence of the composite image and cover sheet in one Embodiment of the division | segmentation scanning method which concerns on this invention. It is a figure which shows typically the mapping vector in one Embodiment of the division | segmentation scanning method which concerns on this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Main body, 11 ... Operation screen, 12 ... Character recognition part, 13 ... Data combination part, 14 ... File preparation part, 15 ... User IF preparation part, 16 ... Scanner control part, 17 ... Event control part, 18 ... Temporary storage , 19 ... scanner, 23 ... reading window, 31 ... document to be read, 32 ... cover sheet, 33, 34 ... transparent sheet, 36 ... fixture, 37 ... position information, 38 ... identifier, 41 ... layout area, 43 ... Composite image, 44 ... Document placement area, 45 ... Position

Claims (5)

In a document reading method for scanning a document to be read into a plurality of small parts and generating image data of the entire document to be read,
A preparatory step in which position information indicating each position of the small portion is described in a drop-out color on a transparent sheet, and a cover sheet that covers the entire document to be read is superimposed on the document to be read and fixed to each other;
A divided scanning step of reading, with the scanner, a composite image including the dropout color and an original image not including the dropout color corresponding to the composite image for each of the small portions of the document to be read fixed to the cover sheet. When,
A document in which the position information is extracted from the composite image for each of the small portions, and the same position as the composite image is read by the scanner so as not to include the dropout color based on the position information extracted from the composite image A position specifying step for specifying the position of the image;
An image combining step of combining the plurality of small portions by respectively arranging a plurality of the original images of the small portions at specified positions and generating image data of the entire document to be read;
A document reading method comprising:   2. The position specifying step includes a step of specifying an average of the position information as the position of the document image when two or more pieces of the position information are extracted from the composite image. Document reading method.   In the position specifying step, when three or more pieces of the position information are extracted from the composite image, a position that is separated by a predetermined size or more from the position of the document image specified from the other position information is specified. 3. The document reading method according to claim 1, wherein the position information is not used for specifying the position of the document image.   2. The position specifying step includes a step of correcting an inclination of the original image based on the position information when two or more pieces of the position information are extracted from the composite image. Item 4. The document reading method according to any one of Items 3 to 4. In a document reading system that scans a document to be read into a plurality of small parts and generates image data of the entire document to be read,
A cover sheet that covers the whole of the document to be read in which position information indicating each position of the small portion is described in a dropout color on a transparent sheet;
A scanner capable of reading an image including the dropout color and an image not including the dropout color;
The position information is extracted from the composite image read together with the dropout color by the scanner for each of the small portions of the document to be read that is superimposed and fixed on the cover sheet, and the position information extracted from the composite image A position specifying unit for specifying, as the position of the original image read by the scanner so as not to include the dropout color, based on
A data combining unit configured to combine the plurality of small parts by respectively arranging a plurality of the original images of the small parts and specifying the whole image data of the document to be read;
A document reading system comprising:

Images