JP2011237840A - Document processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a document processing device which easily generates templates for Optical Character Recognition(OCR).SOLUTION: A document processing device recognizes areas 51 to 53 highlighted in a template image, and stores the coordinates of points indicating each of the areas and character strings (specified characters) within each of the areas as template information. Based on the stored template information, on a newly loaded script image data, characters, other than the specified characters, within the areas defined based on the template coordinates are recognized by OCR and stored as variable parameters together with the loaded script image data.

Description

  The present invention relates to a document processing apparatus that extracts characters from an image of a document and is used for managing the image.

  Conventionally, a form is read, then a line segment in the form is extracted, and a specific area in the form is automatically extracted based on the extracted line segment, whereby the completed form is converted into an OCR (Optical Character). It has been proposed to create a template for performing (Recognition).

JP 2005-267394 A

With the conventional technology, it is not possible to easily create an OCR template for a document other than a form, that is, an original without a line segment.
An object of the present invention is to provide a technique capable of easily creating an OCR template for various originals as well as forms.

  Hereinafter, a plurality of modes will be described as means for solving the problems. These aspects can be arbitrarily combined as necessary.

The document processing apparatus according to the first aspect of the present invention recognizes a character specified based on a designated character as a variation parameter in an image obtained from an original. The document processing apparatus includes an image reading unit, a region position information acquisition unit, a designated character acquisition unit, and a template storage unit. The image reading unit reads an image on a document as a template image. The region position information acquisition unit acquires position information of a designated region having a predetermined feature as region position information in the template image. The designated character acquisition unit acquires a designated character from a character string included in the designated area. The template storage unit stores the designated character in association with the region position information.
According to this document processing apparatus, it is possible to read a document having a specific format in advance and obtain information (position information of a designated region and designated characters) indicating a region where character recognition is to be performed on an unknown document. . Therefore, an OCR template can be easily created.

  The document processing apparatus may further include a relative position information acquisition unit that acquires relative position information indicating the relative position of the variation parameter with respect to the designated character. Further, the template storage unit may store the relative position information in association with the designated character. As a result, the position of the area where the character should be recognized can be more strictly defined.

  The area position information acquisition unit may acquire area position information by recognizing an area indicating a specific color as a specified area.

Further, the document processing apparatus according to the second aspect of the present invention recognizes a character specified based on a designated character as a variation parameter in an image obtained from an original. The document processing apparatus includes an image reading unit, a template storage unit, a template recognition unit, and a variation parameter extraction unit. The image reading unit reads an image on a document as a document image. The template storage unit stores the designated character and the region position information in association with each other. The template recognizing unit recognizes a designated character in an extraction area that is an area specified by area position information in a document image. The variation parameter extraction unit extracts characters existing in the extraction area other than the designated character as variation parameters.
As described above, based on the above-described region position information and the designated character, the region on which the variation parameter is to be acquired by performing character recognition on the document image is specified. That is, a necessary character string can be easily obtained by performing character recognition on only a specific area without performing character recognition on the entire document image. Note that the fluctuation parameter extraction unit not only extracts all the characters existing in the extraction area other than the designated characters as fluctuation parameters, but also extracts only a part of the characters existing in the extraction area other than the designated characters as fluctuation parameters. May be.

  The document processing apparatus may further include an area position information correction unit that corrects the area position information so as to correspond to a wider area when the designated character does not exist in the extraction area.

  The document processing apparatus may further include a document storage unit that stores the variation parameter in association with the document image.

  The document processing apparatus further includes a search reception unit that receives an input of a search word from a user, and a search unit that selects a document image associated with a variation parameter that matches the search word from the document image in the document storage unit. Also good.

  Further, the document processing apparatus may include both of the configurations related to the first viewpoint and the second viewpoint. That is, the document processing apparatus may include an image reading unit, a region position information acquisition unit, a designated character acquisition unit, a template storage unit, a template recognition unit, and a variation parameter extraction unit. The image reading unit reads an image on a document as a template image or a document image. The area position information acquisition unit acquires the position of a designated area having a predetermined feature as area position information in the template image read by the image reading unit. The designated character acquisition unit acquires a designated character from a character string included in the designated area. The template storage unit stores the designated character in association with the region position information. The template recognizing unit recognizes a designated character in an extraction area that is an area specified by the area position information in the document image read by the image reading unit. The variation parameter extraction unit extracts characters existing in the extraction area other than the designated character as variation parameters.

  According to the present invention, it is possible to read a document having a specific format in advance, and acquire information (region position information and designated characters) indicating a region where character recognition should be performed on an unknown document. Therefore, an OCR template can be easily created.

1 is a block diagram showing an overview of a multifunction machine according to an embodiment of the present invention. 10 is a flowchart of template registration processing in the multifunction machine. The figure which shows the example of a template original document. The figure which shows the example of a designated area | region. The figure which shows the example of a temporary template. The figure which shows the example of the template registered. The figure which shows the other example of the template registered. The figure which shows the further another example of the template registered. 6 is a flowchart of document registration in the multifunction peripheral. FIG. 4B is a diagram schematically showing template recognition and variation parameter extraction using the template of FIG. 4A in a document image. FIG. 5 is a diagram schematically showing template recognition and variation parameter extraction using the template of FIG. 4B in a document image. The figure which shows typically template recognition and extraction of a fluctuation parameter using the template of FIG. 4C in a document image.

(1) Configuration of MFP The MFP according to the present embodiment will be described with reference to FIG. FIG. 1 is a block diagram showing the overall configuration of the multifunction machine 1.

The multifunction device 1 is a device having functions as a copier, a printer, a scanner, and a facsimile device.
As shown in FIG. 1, the multifunction device 1 includes an NCU (Network Control Unit) 11, a modem 12, a network interface 13, an image reading unit 14, an image forming unit 15, an operation panel 16, a storage unit 17, and a control device 18. . Each part in the multifunction device 1 is connected by a bus.

  The NCU 11 is connected to a PSTN (Public Subscriber Telephone Network) 101 and controls communication including outgoing and incoming calls via a telephone line.

  The modem 12 is a modem device that converts between a digital signal and an analog signal.

The network interface 13 enables the MFP 1 to connect to the Internet 102 and enables communication with devices outside the MFP 1. As the network interface 13, an interface that can be connected to a LAN (Local Area Network) is provided, and the MFP 1 may be able to communicate with other devices connected to the LAN.
The image reading unit 14 acquires image data by reading an image on a document. In the present embodiment, the image reading unit 14 can acquire color image data. Specifically, the image reading unit 14 includes an unillustrated document conveying device, a platen glass, a light source, a mirror group, a lens, a reading element, a signal processing circuit, and the like. A document transported by a document transport device or a document placed on a platen glass is illuminated by a light source. Reflected light from the document is guided onto the reading element by a mirror group and a lens. The reading element outputs an electrical signal corresponding to the received light, and the signal processing circuit performs processing such as digitization on the electrical signal, thereby obtaining image data.

  The image forming unit 15 forms an image along the image data on a sheet. As the image forming unit 15, an apparatus that forms an image by an electrophotographic method or an ink jet method is employed.

  The operation panel 16 includes a display panel and a touch panel 161 having a touch sensor; a hard key 162 and the like. The touch panel 161 displays an image on the display panel, thereby notifying the user of information such as the operation status of the image forming apparatus and the occurrence of an error, and accepting an operation from the user via a soft key. The hard keys include keys such as a numeric keypad, an execution key, and a cancel key, and can accept operations from the user. For example, the user can input an instruction to execute copying, scanning, facsimile transmission, or the like using the operation panel 16.

  The storage unit 17 is a non-volatile memory that does not erase the stored contents even when the power of the multifunction device 1 is turned off. Specifically, the storage unit 17 includes a template storage area (template storage unit) 171 and a document storage area (document storage unit) 172. The template storage area 171 stores a template described later. The document storage area 172 stores image data and variation parameters, which will be described later.

  The control device 18 is a device that controls the operation of each unit in the multi-function device 1 and executes processing such as various calculations. The control device 18 includes an arithmetic device such as an MPU (Micro Processing Unit) and a storage device such as a flash memory and an SDRAM (Synchronous Dynamic Random Access Memory). The MPU implements the function of the control device 18 by reading and executing a program stored in the flash memory. Specifically, the control device 18 controls the operation of each unit of the multifunction device 1 to execute operations such as copying, printing, facsimile transmission / reception, and scanning. The control device 18 includes a character recognition unit 20, a template acquisition unit 31 (region position information acquisition unit, designated character acquisition unit, relative position information acquisition unit), a template recognition unit 41, and a coordinate correction unit 42 (region position information correction unit). ), Function as a fluctuation parameter extraction unit 43 and a search unit 44. The function of each part will be described later. The flash memory stores various programs. The SDRAM is a readable / writable memory and functions as a work area when the MPU executes a program.

  In addition to the configuration described above, the multifunction device 1 includes various devices as necessary. For example, the multifunction device 1 includes a CODEC (Coder and Decoder) (not shown). Compress (encode) / decompress (decode) the CODEC image data.

(2) Template Registration Template registration will be described with reference to FIGS. 2, 3A to 3C, and 4A to 4C. FIG. 2 is a flowchart showing the flow of template registration, FIG. 3A is a diagram showing an example of a template document, FIG. 3B is a diagram showing an example of a designated area, and FIG. 3C is a diagram showing an example of a temporary template. is there. 4A to 4C are examples of registered templates.

(2-1) Reading of template document: Step S1
As shown in FIG. 2, first, the image reading unit 14 reads an image on the template original 50 and acquires image data (step S1). As shown in FIG. 3A, the template document 50 has a first designated area 51, a second designated area 52, and a third designated area 53 to which color markers are attached.

(2-2) Acquisition of coordinates of designated area: steps S2 to S3
The template acquisition unit 31 functions as a region position information acquisition unit.

That is, the template acquisition unit 31 recognizes a designated area by recognizing an area with a color designated in advance (step S2), and obtains its coordinates as area position information (step S3). For example, when yellow is designated as the marker color in advance, the template acquisition unit 31 recognizes a yellowed area in the image of the template document 50.
The marker is only an example of the characteristics of the designated area. The designated area may be determined in advance and have a feature that can be distinguished from other areas in the template document. Features include fonts, underlines, color characters, etc., in addition to markers.

  Furthermore, as illustrated in FIG. 3B, the template acquisition unit 31 acquires the coordinates of two diagonal points of the recognized area. In the example of FIG. 3B, coordinates (X11, Y11) and coordinates (X12, Y12) are acquired as area position information corresponding to the first designated area 51. The rectangular region having the two coordinates specified in this way as diagonals is hereinafter referred to as a “template region”, and denoted by reference numeral “54” in FIGS. 3A and 3B.

  Similarly, the coordinates of the second template area 55 corresponding to the second designated area 52 and the coordinates of the third template area 56 corresponding to the third designated area 53 are acquired (FIGS. 3A and 3C).

  Note that the color designation may be set by default or may be input via the operation panel 16.

(2-3) Acquisition of temporary template: Steps S4 to S6
The template acquisition unit 31 also functions as a designated character acquisition unit and a relative position information acquisition unit. The template acquisition unit 31 acquires a temporary template, and then edits the template to acquire a final template (including coordinates, designated characters, and relative position information).

  First, acquisition of a temporary template will be described. Under the control of the template acquisition unit 31, the character recognition unit 20 performs OCR processing on the template area (step S4). That is, based on the image data, the character recognition unit 20 recognizes a character string included in the template area. As shown in FIG. 3A, the character string included in the first template area 54 is “technical document: ZZZ report”, and the character string included in the second template area 55 is “date: January 15, 2010”. The character string included in the third template area 56 is “author: Taro Suzuki”.

The character string thus recognized and the coordinates acquired in step S3 are stored as a temporary template in a storage medium such as the above-described SDRAM (step S5). If there is another area indicating the specified color in one template document 50 (No in step S6), coordinate acquisition, OCR processing, and temporary template storage are executed for the other area (step S6). S3 to S5). If the storage of the temporary template is completed for all the areas indicating the above-mentioned designated color in one template document 50 (Yes in step S6), the following processing is performed.
In the example of FIG. 3A, the above processing is executed for the three designated areas 51 to 53.

  The temporary template will be specifically described. As illustrated in FIG. 3C, the temporary template 60 includes first information 61, second information 62, and third information 63. The first information 61 has the coordinates of the first template area 54 and has “technical document: ZZZ report” which is a character string obtained from the first template area 54 as an index. The second information 62 has the coordinates of the second template area 55 and has “date: January 15, 2010” which is a character string obtained from the second template area 55 as an index. The third information 63 has the coordinates of the third template region 56 and has “Creator: Taro Suzuki” which is a character string obtained from the third template region 56 as an Index.

(2-4) Template editing: Step S7
As the next processing, the temporary template is edited (step S7).
Specifically, temporary template 60 is displayed on touch panel 161, and soft keys for accepting editing of information included in the template are displayed. “Edit” includes addition, change, and deletion. Information that is not edited need not be displayed. For example, when the coordinates are set so as not to be edited, the coordinates may not be displayed as a temporary template.

  The template acquisition unit 31 acquires the final template by editing the temporary template based on the input made by the user via these soft keys. The template finally obtained may include the above-described coordinates, and may include an index name, designated characters, and relative position information.

  The index name is a character string designated by the user, and is generally a character string indicating a classification of a variation parameter described later. Specifically, the name of the index includes technical documents, annual reports, monthly reports and experimental reports, etc .; the types of dates such as the date of creation and submission; An alphabet representing the name type of the person or the like is designated.

  The designated character is a concept including Index and Delimiter. The designated character includes a symbol. The Index is a character string that is written in association with a variation parameter described later. Delimiter is a symbol indicating a delimiter between the variation parameter and another character string, for example, a symbol indicating a delimiter between the index and the variation parameter. As the symbol of Delimiter, a colon or a semicolon is used.

The relative position information is information indicating the relative position of the variation parameter with respect to the designated character, for example, information indicating the direction of the variation parameter with respect to the designated character. The direction input is performed by displaying soft keys indicating choices of “right”, “left”, “up”, and “down”, and pressing one of the soft keys.
Moreover, the template acquisition part 31 can also correct the coordinate which is area | region position information according to the instruction | indication from a user. That is, the above-mentioned coordinates can be corrected by the user who has seen the coordinates displayed on the touch panel 161 inputting a more appropriate value.

A specific example of the template will be described below.
(I) First Example The template 70 of FIG. 4A includes first template information 701, second template information 702, and third template information 703. The first template information 701, the second template information 702, and the third template information 703 are information obtained by editing the first information 61, the second information 62, and the third information 63, respectively. The template information 701 to 703 includes coordinates of the first template region 54, the second template region 55, and the third template region 56, respectively, and includes an index name, designated characters, and relative position information.

  As illustrated in FIG. 4A, the first template information 701 includes a character string “Technical Doc” corresponding to a “technical document” that is a document name, as an index name. The second template information 702 has a character string “Date” corresponding to “date”. The third template information 703 includes a character string “Originator” corresponding to “Creator”. These character strings are obtained when the user inputs them.

  As shown in FIG. 4A, as the designated character, the first template information 701 has a character string “technical document:” as an index. This character string is obtained by deleting the character string “ZZZ technical document” included in the Index in the first information 61 (FIG. 3C) of the temporary template based on a user instruction. The second template information 702 includes a character string “Date:” as an Index. This character string is obtained by deleting the character string “January 15, 2010” included in the Index of the second information 62. The third template information 703 includes a character string “creator:” as an index. This character string is obtained by deleting the character string “Taro Suzuki” included in the index of the third information. As described above, in the template information 701 to 703 in FIG. 4A, the index that is the designated character includes both the character string indicating the classification of the variation parameter and the Delimiter.

  The template document 50 includes character strings other than the designated characters (“ZZZ report”, “January 15, 2010”, and “Suzuki, which are character strings corresponding to the variation parameters described later) in the designated areas 51 to 53. Taro "). Therefore, the designated character can be obtained by deleting the character string from the temporary template 60. However, when the template document does not include a character string other than the designated character in the designated area, such deletion work is not necessary.

  As shown in FIG. 4A, the first template information 701, the second template information 702, and the third template information 703 have information of “Right” as relative position information (Location).

(Ii) Second Example The template 71 in FIG. 4B does not include an index as a designated character, and has only a delimiter. Otherwise, the template 71 of this example has the same configuration as the template 70 of the first example. Specifically, the first template information 711, the second template information 712, and the third template information 713 of the template 71 have a colon (:) as a delimiter and do not include an index.
As described above, the template acquisition unit 31 can delete “Index” included in the temporary template 60 in accordance with a user instruction, and can also add “Delimiter”.

(Iii) Third Example The template 72 of FIG. 4C has the same configuration as the template 70 of the first example, except that both the index and the delimiter are included as individual characters as designated characters.

That is, the first template information 721 of the template 72 includes “technical document” as an index and a colon as a delimiter. The second template information 722 includes “date” as an index and a colon as a delimiter. The third template information 723 includes “creator” as an index and a colon as a delimiter.
As described above, the template acquisition unit 31 can delete a part of the Index included in the temporary template 60 and can also add “Delimiter”.

(2-5) Template registration: Steps S8 and S9
After the editing is completed (Yes in step S8), the latest template after editing is stored in the template storage area 171 of the storage unit 17 under the control of the template acquisition unit 31 (step S9). Thus, the coordinates of the template area and the designated character are stored in association with each other.

  Completion of editing is instructed, for example, by pressing a start key provided as a hard key, pressing a “done” key provided as a soft key, pressing a “save” key provided as a soft key, or the like.

(3) Document Registration The multifunction device 1 can register an image (document) of a document including characters in association with a variation parameter, and use the variation parameter for document management. That is, the variation parameter is used as an index.
Hereinafter, document registration will be described with reference to FIG. FIG. 5 is a flowchart showing the flow of document registration.

(3-1) Template selection: S11
The multi-function device 1 accepts selection of a template used for the following processing from the user via the operation panel 16 (step S11).

(3-2) Document reading: S12 to S20
The image reading unit 14 reads an image from a document set by the user (step S12).

(3-3) Fluctuation parameter extraction: Steps S13 to S19
The template recognition unit 41 extracts a region specified by the coordinates included in the template selected in step S11 from the obtained image (step S13).

  The character recognition unit 20 performs OCR processing on the extracted region (hereinafter referred to as “extraction region”) under the control of the template recognition unit 41 (step S14).

  The template recognition unit 41 determines whether or not the character string obtained in this way includes a designated character (Index and / or Delimiter) in the template (step S15).

If a character that matches the designated character is included (Yes in step S16), the variation parameter extracting unit 43 determines, for the designated character, a character string that exists at the position specified by the relative position information in the template, Extracted as a variation parameter (step S17).
The variation parameter extraction unit 43 may extract not only all characters existing in the extraction area as variation parameters, but also extract only a part of characters existing in the extraction region other than the designated characters as variation parameters. For example, a limit value may be provided for the number of characters to be extracted as the variation parameter, and when the character string existing in the extraction region exceeds the limit value, the variation parameter extraction unit 43 matches the limit value. A number of character strings may be extracted.
The characters extracted by the variation parameter extraction unit 43 are displayed on the touch panel 161, so that the user can check the extracted contents. The variation parameter extraction unit 43 may further modify the variation parameter (deletion and change of some character strings, etc.) in accordance with a user instruction or the like.

  If a character that matches the designated character is not included (No in step S16), the coordinate correction unit 42 corrects the coordinates so that the extraction area becomes wider (step S18). Specifically, the coordinates are corrected so that the difference between the X components and / or the difference between the Y components of two coordinates in one template information becomes large. Based on the corrected coordinate information, the processing after the region extraction is performed again.

  The above processing is performed until the acquisition of the variation parameter is completed for all the extraction regions specified by one template (step S19).

  A specific example of fluctuation parameter extraction will be described with reference to FIGS. 6A to 6C. 6A to 6C are diagrams schematically showing template recognition and variation parameter extraction when the above-described templates 70 to 72 are respectively used in images obtained from the same document. This will be described below.

(I) First Example When the template 70 of FIG. 4A is selected in step S12, as shown in FIG. 6A, the coordinates of the first template information 701, the second template 702, and the third template 703 in the document image 80 are displayed. Based on the above, the first extraction area 81, the second extraction area 82, and the third extraction area 83 are extracted (step S13).

  In the first extraction area 81, the character string 811 that matches the index “technical document:” of the first template information 701 is recognized, and the character that is the variation parameter 812 present on the right side of the character string 811 based on the relative position information. The column “experiment report” is extracted (steps S14 to S17).

  In the second extraction area 82, the character string 821 that matches the index “date:” of the second template information 702 is recognized, and the character string that is the variation parameter 822 present on the right side of the character string 821 based on the relative position information. “March 15, 2010” is extracted (steps S14 to S17).

  In the third extraction area 83, the character string 831 that matches the index “creator:” of the third template information 703 is recognized, and the character that is the variation parameter 832 present on the right side of the character string 831 based on the relative position information. The column “Jiro Tanaka” is extracted (steps S14 to S17).

(Ii) Second Example In this example, the same processing as that in the first example is performed except that the used template is different and the designated characters are different, and the extracted variation parameters are also the same.

  When the template 71 of FIG. 4B is selected in step S12, as shown in FIG. 6B, the first extraction area 81, the second extraction area 82, and the third extraction area 83 are extracted from the document image 80 (step S12). S13).

  In the first extraction area 81, the symbol 841 that matches the Delimiter of the first template information 711 is recognized, and the variation parameter 812 is extracted based on the relative position information (steps S14 to S17).

  In the second extraction area 82, the symbol 842 that matches the Delimiter of the second template information 712 is recognized, and the variation parameter 822 is extracted based on the relative position information (steps S14 to S17).

  In the third extraction region 83, the symbol 843 that matches the Delimiter of the third template information 713 is recognized, and the variation parameter 832 is extracted based on the relative position information (steps S14 to S17).

(Iii) Third Example In this example, the same processing as that in the first example is performed except that the used template is different and the designated characters are different, and the extracted variation parameters are also the same.

  When the template 72 of FIG. 4C is selected in step S12, as shown in FIG. 6C, the first extraction area 81, the second extraction area 82, and the third extraction area 83 are extracted from the document image 80 (step S12). S13).

  In the first extraction area 81, the character string symbol 851 that matches the Index “technical document” of the first template information 721 and the symbol 841 that matches the Delimiter are recognized (steps S14 to S16). Based on the relative position information, the fluctuation parameter 812 existing on the right side of the symbol 841 is extracted (steps S14 to S17).

  In the second extraction area 82, the character string 852 that matches the index “date” of the second template information 722 and the symbol 842 that matches Delimiter are recognized (steps S14 to S16). Based on the relative position information, the fluctuation parameter 822 existing on the right side of the symbol 842 is extracted (steps S14 to S17).

  In the third extraction area 83, the character string 853 that matches the index “creator” of the third template information 723 and the symbol 843 that matches Delimiter are recognized (steps S14 to S16). Then, based on the relative position information, the fluctuation parameter 832 present on the right side of the symbol 843 is extracted (steps S14 to S17).

(3-4) Document registration: Step S20
After the acquisition of the variation parameter is completed (Yes in step S19), the image data of the document and the variation parameter are associated and stored in the document storage area 172 (step S20). Specifically, the variation parameter is stored as XML (Extensible Markup Language) data.

(4) Document Search The variation parameters registered in this way are used for document classification and search, for example.
When the user inputs a search word via the operation panel 16 and instructs the execution of the search, the search unit 44 searches the document storage area 172 for an image having a variation parameter that matches the search word. The search result is displayed on the touch panel 161. In this way, the user can easily obtain the target image.
In the search unit 44, a search method such as partial match or complete match may be set by default, or the search method may be changeable by user designation.

  As mentioned above, although one Embodiment of this invention was described, this invention is not limited to the said embodiment, A various change is possible in the range which does not deviate from the summary of invention.

(5) Features The multi function device 1 recognizes a character specified based on a designated character as a variation parameter in an image obtained from a document. The multi-function device 1 includes an image reading unit 14, a template acquisition unit 31, and a template storage area 171. The image reading unit 14 reads an image on a document. The template acquisition unit 31 acquires the coordinates of the designated areas 51 to 54 having predetermined characteristics in the template image read by the image reading unit 14. The template acquisition unit 31 acquires a designated character from the character string included in the designated areas 51 to 54. The template storage area 171 stores designated characters in association with coordinates.

  According to the multi function device 1, it is possible to read a document having a specific format in advance, and acquire information (region position information and designated characters) indicating an area where character recognition should be performed on an unknown document. Therefore, an OCR template can be easily created.

  In the multi function device 1, the template acquisition unit 31 also functions as a relative position information acquisition unit that acquires relative position information indicating the relative position of the variation parameter with respect to the designated character. Further, the template storage area 171 may store the relative position information in association with the designated character. As a result, the position of the area where the character should be recognized can be more strictly defined.

  In the multi function device 1, the template acquisition unit 31 recognizes a region showing a specific color.

  In addition, the multifunction machine 1 includes a template recognition unit 41 and a variation parameter extraction unit 43. The template recognition unit 41 recognizes a designated character in the extraction areas 81 to 83 that are areas specified by coordinates in the document image 80. The variation parameter extraction unit 43 extracts characters existing in the extraction areas 81 to 83 as variation parameters other than the designated character.

  As described above, based on the above-described coordinates and the designated character, a region on which the variation parameter is to be acquired by performing character recognition on the document image is specified. That is, a necessary character string can be easily obtained by performing character recognition on only a specific area without performing character recognition on the entire document image.

The multifunction device 1 further includes a coordinate correction unit 42 that corrects the region position information so as to correspond to a wider region when there is no designated character in the extraction regions 81 to 83.
The multi-function device 1 further includes a document storage area 172 for storing the variation parameter in association with the document image.

  In the multi function device 1, the operation panel 16 receives an input of a search word from the user. The search unit 44 selects a document image associated with the variation parameter that matches the search word from the document images in the document storage unit.

  The present invention can be used for an apparatus having an image reading function, for example, a facsimile machine, a scanner, a multifunction machine, and the like.

DESCRIPTION OF SYMBOLS 1 Multifunction machine 12 Modem 13 Network interface 14 Image reading part 15 Image forming part 16 Operation panel 17 Memory | storage part 18 Control apparatus 20 Character recognition part 31 Template acquisition part (Area position information acquisition part, designated character acquisition part, relative position information acquisition part )
41 Template recognition unit 42 Coordinate correction unit 43 Fluctuation parameter extraction unit 44 Search unit 50 Template document 51 First designated region 52 Second designated region 53 Third designated region 54 First template region 55 Second template region 56 Third template region 60 Temporary template 70 Template 71 Template 72 Template 102 Internet 161 Touch panel (search reception part)
162 Hard key 171 Template storage area (template storage unit)
172 Document storage area

Claims (8)

A document processing apparatus that recognizes, as a variation parameter, a character specified based on a specified character in an area specified by area position information in an image obtained from a document.
An image reading unit that reads an image on a document as a template image;
In the template image, an area position information acquisition unit that acquires a position of a specified area having a predetermined characteristic as the area position information;
A designated character obtaining unit for obtaining the designated character from a character string included in the designated region;
A template storage unit for storing the designated character in association with the region position information;
A document processing apparatus comprising: A relative position information acquisition unit that acquires relative position information representing a relative position of the variation parameter with respect to the designated character;
The document processing apparatus according to claim 1, wherein the template storage unit stores the relative position information in association with the designated character. The document processing apparatus according to claim 1, wherein the area position information acquisition unit acquires the area position information by recognizing an area indicating a specific color as the designated area. A document processing apparatus that recognizes, as a variation parameter, a character specified based on a specified character in an area specified by area position information in an image obtained from a document.
An image reading unit that reads an image on a document as a document image;
A template storage unit that associates and stores the designated character and the region position information;
A template recognizing unit that recognizes the designated character in an extraction area that is an area specified by the area position information in the document image;
A variation parameter extraction unit that extracts characters existing in the extraction region other than the designated character as a variation parameter;
A document processing apparatus comprising: The document processing apparatus according to claim 4, further comprising an area position information correction unit that corrects the area position information so as to correspond to a wider area when the designated character does not exist in the extraction area. The document processing apparatus according to claim 4, further comprising a document storage unit that stores the variation parameter in association with the document image. A search reception unit that receives input of search terms from the user;
A search unit that selects the document image associated with the variation parameter that matches the search term from the document image in the document storage unit;
The document processing apparatus according to claim 6, further comprising: A document processing apparatus that recognizes, as a variation parameter, a character specified based on a specified character in an area specified by area position information in an image obtained from a document.
An image reading unit that reads an image on a document as a template image or a document image;
In the template image, an area position information acquisition unit that acquires the position of a specified area having a predetermined feature as area position information;
A designated character obtaining unit for obtaining a designated character from a character string included in the designated region;
A template storage unit for storing the designated character in association with the region position information;
A template recognizing unit that recognizes the designated character in an extraction area that is an area specified by the area position information in the document image;
A variation parameter extraction unit that extracts characters existing in the extraction region other than the designated character as a variation parameter;
A document processing apparatus comprising:

Images