Document preparation system and program

Abstract

To provide a document preparation system which automatically corrects even a document including defects, and prepares a document of a fair copy version.SOLUTION: A document preparation system includes: first storage means for storing a template of a document; second storage means for storing an input rule of one or more columns included in the template so that the input rule is associated with the columns; acquisition means for acquiring a content entered in the columns, from the entered document based on the template; generation means for correcting the content acquired by the acquisition means, on the basis of the input rule associated with the columns, and generating a corrected version of the content; and creation means for applying the corrected version of the content generated by the generation means to the template, and creating a fair copy version of the document.SELECTED DRAWING: Figure 10

Description

Technical Field

[0001] Embodiments of the present invention relate to a document creation system and a program.

Background Art

[0002] Conventionally, templates (also referred to as formats, styles, or prototypes) may be used for creating documents. A template defines a format that conforms to, for example, the purpose of use or the destination of submission of a document, and has an entry field indicating the content to be entered.

[0003] Examples of documents that conform to a template include work reports, inspection reports, and questionnaire forms. These documents are brought to the site where work, inspection, or questionnaire is conducted as printed materials with blank entry fields, and are also used by being handwritten in the blank fields at the site.

[0004] As described above, documents that are freely written in blank fields at the site of work or the like may contain various defects such as typographical errors and redundant expressions. For this reason, it is desired that the person in charge submit a fair copy of the document that has been recreated by correcting typographical errors and summarizing it. On the other hand, it is desired that official documents such as reports and documents be created efficiently without spending much time.

[0005] Although various techniques have been devised for extracting and processing character string data from image data (for example, Patent Document 1), there is no technique that meets the above-described demands.

Summary of the Invention

Problems to be Solved by the Invention

[0006] The problem to be solved by the present invention is to provide a document creation system and a program that automatically corrects defects in a document and creates a fair copy of the document even if the document contains defects.

Means for Solving the Problems

[0007] The document creation system according to the embodiment includes: a first storage means for storing a document template; a second storage means for storing input rules for one or more columns included in the template, associated with the columns; an acquisition means for acquiring the content entered in the columns from a completed document along the template; a generation means for correcting the content acquired by the acquisition means based on the input rules associated with the columns to generate a corrected version of the content; and a creation means for applying the corrected version of the content generated by the generation means to the template to create a fair copy of the document.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Mode for Carrying Out the Invention

[0009] (First Embodiment) Embodiments will be described with reference to the drawings. FIG. 1 is a diagram showing an example of the configuration of a document creation system according to an embodiment. The document creation system includes a server 1, an in-house DB 3, and an image input device 5.

[0010] The server 1 is an information processing device that provides a processing unit of the document creation system. The information processing device is, for example, a PC (Personal Computer) or the like.

[0011] The in-house DB 3 is an example of a first storage means and a second storage means. The in-house DB 3 is a database (DB, database) that provides information (such as a template 31 and an input rule 32) used by the document creation system for document creation, and is configured by, for example, one or more information processing devices.

[0012] The image input device 5 is a device that outputs image data obtained by reading or imaging, such as a scanner or a camera.

[0013] Note that the document creation system of the present embodiment is configured to complete processing on-premises (self-operated) for maintaining and keeping confidential information such as outside-company secrets. That is, the server 1, the in-house DB 3, and the image input device 5 included in the document creation system are all on-premises devices. However, in implementation, depending on the necessity or the confidentiality level, a configuration that does not complete on-premises may be acceptable. In that case, cloud services may be used for part or all of the devices constituting the system as appropriate.

[0014] FIG. 2 is a block diagram showing an example of the hardware configuration of the server 1. The server 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a communication unit 14, a storage unit 19, and the like.

[0015] The CPU 11 is an example of a processor and comprehensively controls each part of the server 1. The ROM 12 stores various programs. The RAM 13 is a workspace for expanding programs and various data. The CPU 11, ROM 12, and RAM 13 are connected via a bus or the like and constitute a control unit 10 of a computer configuration.

[0016] The communication unit 14 is a communication interface that communicably connects the control unit 10 and other devices (for example, the in-house DB 3 and the image input device 5) via a network such as an in-house LAN (Local Area Network).

[0017] The storage unit 19 has a storage medium such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or a flash memory, and maintains the stored content even when the power is turned off. The storage unit 19 stores a program 191 executable by the CPU 11 and various information.

[0018] FIG. 3 is a diagram showing an example of a document. The document 51 in this example is a questionnaire. The document 51 includes a document ID column 511, a date column 512, and entry columns 513 to 515. Hereinafter, when it is not necessary to separately refer to the document ID column 511, the date column 512, and the entry columns 513 to 515, they may be collectively referred to simply as columns 511 to 515.

[0019] FIG. 4 is a diagram showing an example of the template 31. The template 311 corresponds to the document 51 and includes coordinate information (x11, y11) to (x51, y52) of each column 511 to 515. The coordinate information is, for example, two-dimensional coordinates when any of the four corners of the paper surface is used as the origin, and is a value indicating the position of the column.

[0020] FIG. 5 is a diagram showing an example of the input rule 32. The input rule 321 is information specifying the content to be written in each of the above columns 511 to 515. The input rule 321 is associated with each of the coordinate information (x11, y11) to (x51, y52) of the above columns 511 to 515. In FIG. 5, for convenience of explanation, the symbol of the corresponding column is attached to each coordinate information.

[0021] For example, the input rule 321 specifies that the document search identifier should be input in the document ID column 511, the date should be input in the date column 512, and the character string should be input in the entry columns 513 and 515. Further, the input rule 321 specifies that the character string to be input in the entry column 514 should be about 100 characters.

[0022] FIG. 6 is a diagram showing another example of a document. The document 52 in this example is a work report. The document 52 includes a document ID column 521, a date column 522, and entry columns 523 to 527. Hereinafter, when it is not necessary to separately mention the document ID column 521, the date column 522, and the entry columns 523 to 527, they may be collectively referred to simply as columns 521 to 527.

[0023] FIG. 7 is a diagram showing another example of the template 31. The template 312 corresponds to the document 52 and includes the coordinate information (x11, y11) to (x71, y72) of each column 521 to 527.

[0024] FIG. 8 is a diagram showing another example of the input rule 32. The input rule 322 is information specifying the content to be written in each of the above columns 521 to 527. The input rule 322 is associated with each of the coordinate information (x11, y11) to (x71, y72) of the above columns 521 to 527. In FIG. 8, for convenience of explanation, the symbol of the corresponding column is attached to each coordinate information.

[0025] For example, the input rule 322 specifies that the document search identifier should be input in the document ID column 521, the date should be input in the date column 522, and the character string should be input in the entry columns 525 and 526.

[0026] In addition, the input rule 322 specifies that what should be input into the entry field 523 is a one-line character string. Further, the input rule 322 specifies that what should be input into the entry field 524 is a date and also specifies its format. In this example, a format including year, month, day, and time zone is specified. And the input rule 322 specifies that what should be input into the entry field 527 is a character string indicating a name.

[0027] In the following description, when there is no need to distinguish between document 51 and document 52 and mention them separately, both are collectively referred to simply as document 50. Also, when there is no need to distinguish between template 311 and template 312 and mention them separately, both are collectively referred to simply as template 31. Similarly, when there is no need to distinguish between input rule 321 and input rule 322 and mention them separately, both are collectively referred to simply as input rule 32.

[0028] FIG. 9 is a block diagram showing an example of functional units included in the control unit 10. By executing the program 191, the control unit 10 functions as an information acquisition unit 101, an image processing unit 102, a front-end processing unit 105, a natural language processing unit 106, and a back-end processing unit 107.

[0029] The information acquisition unit 101 acquires information such as the template 31 and the input rule 32 from the in-house DB3 according to the requests of other functional units.

[0030] The image processing unit 102 acquires image data obtained by scanning or imaging the document 50 via the image input device 5, and extracts the information included in the document 50 by analyzing the acquired image data. In the above-mentioned analysis, for example, OCR (Optical Character Recognition / Reader, optical character recognition) processing and image recognition processing are performed. The information extracted from the document 50 is, for example, text data obtained by reading the text described in the document 50 or data of the illustration drawn in the document 50.

[0031] In addition, the image processing unit 102 acquires coordinate information on the display positions of text data and illustration data on their pages from the image data of the document 50. Then, the image processing unit 102 associates the coordinate information with the text data and illustration data and stores them in a predetermined storage unit.

[0032] The front-end processing unit 105 is an example of acquisition means for acquiring the content entered in one or more columns included in the template 31 from the completed document 50 along the template 31. For example, the front-end processing unit 105 acquires the completed document 50 as image data, and acquires the content entered in the columns by performing image processing on the image data. Further, the front-end processing unit 105 acquires, for example, text data as the content entered in the columns. In addition, the front-end processing unit 105 does not transfer to the natural language processing unit 106 the content determined not to be the target of revised version generation by the natural language processing unit 106 among the content entered in the columns, and transfers it to the back-end processing unit 107.

[0033] Specifically, the front-end processing unit 105 determines the materials for the document to be created this time by the user's operation. More specifically, for example, the front-end processing unit 105 provides a UI (User Interface) that can be displayed on the terminal operated by the user, and receives the operation via the UI, thereby inputting information as the above materials to the natural language processing unit 106.

[0034] The above UI may include, for example, a web page displayed by a web browser, or may be realized by application software installed on the user's terminal or the server 1. In the above example, the user performs an operation on the above UI to specify the image data of the desired document 50 among the image data input from the image input device 5 as the above materials.

[0035] In response to the user's operation, the front-end processing unit 105 first obtains, via the information acquisition unit 101, a template 31 suitable for the document 50 and an input rule 32 corresponding to the template 31 from the in-house DB 3.

[0036] When selecting a template 31 suitable for the document 50, the front-end processing unit 105 determines a template 31 suitable for the document 50 based on the information (text data and illustration data) extracted from the document 50 and their coordinate information. In addition, for this determination, the title described at the beginning of the document or the identification information (document search identifier) shown in the document ID column 521 may be used.

[0037] Next, the front-end processing unit 105 distributes information that serves as material for creating a fair copy of the document 50 to the natural language processing unit 106 and the back-end processing unit 107.

[0038] In the distribution, the front-end processing unit 105 distinguishes, based on the input rule 32, text data to be the object of natural language processing by the natural language processing unit 106 and text data that is not. Subsequently, the front-end processing unit 105 transfers the text data that is the object of natural language processing and its input rule 32 to the natural language processing unit 106. Furthermore, the front-end processing unit 105 transfers the text data that is not the object of natural language processing, the template 31 corresponding to the document 50, and the input rule 32 to the back-end processing unit 107.

[0039] The text data to be the object of natural language processing is, for example, what is entered in columns written in free form such as the entry columns 513 to 515 and the entry columns 525 and 546. The sentences entered in these columns may contain typos and omissions or may be redundant sentences. Therefore, for those sentences, the natural language processing unit 106 corrects typos and omissions and performs summarization.

[0040] The natural language processing unit 106 is an example of a generation unit that modifies the content acquired by the acquisition unit based on the input rule 32 associated with the column to generate a revised version of the content. The natural language processing unit 106 generates a revised version of the content by correcting typographical errors and summarizing the content based on the input rule 32.

[0041] Specifically, the natural language processing unit 106 performs natural language processing on the information passed from the front-end processing unit 105 as a material to generate a sentence for writing in the fair copy of the document 50. More specifically, the natural language processing unit 106 performs natural language processing based on the text data extracted from the image data of the document 50 to generate a sentence that conforms to the input rule 32. Regarding the natural language processing itself, existing technologies are used and detailed explanations are omitted.

[0042] The natural language processing unit 106 passes the generated sentence to the back-end processing unit 107.

[0043] The back-end processing unit 107 is an example of a creation unit that creates the fair copy of the document 50 by fitting the revised version of the content generated by the generation unit to the template 31. The back-end processing unit 107 creates a document as the fair copy of the document 50 by fitting the sentence passed from the natural language processing unit 106 and the data passed from the front-end processing unit 105 to the template 31.

[0044] At this time, the back-end processing unit 107 performs a process of correcting the data that has not passed through the natural language processing unit 106 according to the input rule 32 as necessary. For example, the date notation format is adjusted to a specified one (e.g., "YYYY / MM / DD" or "MMMM D, YYYY").

[0045] FIG. 10 is a flowchart showing an example of the flow of processing executed by the control unit 10. When the control unit 10 receives the input of image data from the image input device 5 (step S1), it performs image data analysis as the image processing unit 102 (step S2). Subsequently, the control unit 10, as the front-end processing unit 105, acquires input data according to the user's operation (step S3). Here, the input data is the material of the document to be created as described above. For example, in addition to the text etc. extracted from the document 50 and their coordinate information, it is the template 31 and the input rule 32 suitable for the document 50.

[0046] After this, the control unit 10 repeats the processing of steps S4 to S9 for each column of the document 50, and appropriately arranges text in each column. Then the control unit 10 finally outputs the document data as the fair copy of the document 50 (step S10) and ends the processing. Note that the reading of the input data (step S4) is performed, for example, for each column, and the reading order may be in the order from the column located above to the column located below the document 50, or may be random.

[0047] A typical example of the processing in steps S4 to S9 is as follows. For example, in the case of the document 51, when the control unit 10 reads the input data related to the column 514 (No in step S4), it refers to the content associated with the coordinate information of the column 514 in the input rule 321 and obtains the specification of "string, about 100 characters" (step S5). From the content of this specification, the control unit 10 (front-end processing unit 105) determines that the column 514 is the target of natural language processing, and creates the input data to the natural language processing unit 106 (step S6). The input data is the text data read from the column 514 and the input rule 321 of the column 514.

[0048] Subsequently, the control unit 10 as the front-end processing unit 105 transfers the input data created in step S6 to the natural language processing unit 106. The control unit 10 as the natural language processing unit 106 performs natural language processing on the text data read from column 514 according to the input rule 321, thereby creating a text in which typos and omissions are corrected and summarized to about 100 characters (step S7).

[0049] Subsequently, the control unit 10, as the back-end processing unit 107, refers to the template 311 (step S8) and arranges the text created in step S7 in a predetermined column 514 of the template 311 (steps S8 to S9).

[0050] The above processing is performed for all columns. However, for columns that are not subject to natural language processing, steps S6 and S7 are skipped, and the processing proceeds from step S5 to step S8.

[0051] When the reading of all input data is completed in step S4 (Yes in step S4), the control unit 10 outputs a document as the fair copy of the document 50 (step S10) and ends this processing.

[0052] Thus, according to the present embodiment, even if the original document contains defects such as typos and omissions or redundant expressions, they can be automatically corrected by using natural language processing to eliminate the defects and redundant expressions, etc., and a fair copy document can be created.

[0053] That is, according to the present embodiment, for example, a printed matter in which each column of the template 31 is printed blank is brought into the work or inspection site, text data is extracted from the document 50 created by a worker or inspector filling it in by hand, and the text is automatically arranged, etc., so that a fair copy document can be created.

[0054] As a result, conventionally, after work or inspection, workers and inspectors no longer need to perform cumbersome post-work tasks such as manually reading handwritten characters visually using a PC or the like and then manually inputting them into the appropriate columns of the template 31 while arranging the text.

[0055] In addition, according to the present embodiment, for example, text data and illustration data are extracted from a non-handwritten document 50 input in each column of the template 31, and the text in each column is automatically arranged and the data format and size of the illustration data are adjusted in accordance with the input rule 32, so that a fair copy version of the document can be created.

[0056] As a result, a document 50 containing defects such as typographical errors, images whose size exceeds the allowable range, and redundant expressions can be automatically arranged to conform to the input rule 32, and a fair copy version of the document can be created. This can reduce the labor required for fair copying of the document.

[0057] The above-described embodiment can also be appropriately modified and implemented by changing a part of the configuration or function of each of the above-described devices.

[0058] The program executed by each device of the above-described embodiment is provided by being pre-installed in a ROM or the like. The program executed by each device of the above-described embodiment may be configured to be recorded on a computer-readable recording medium such as a CD-ROM, a flexible disk (FD), a CD-R, or a DVD (Digital Versatile Disk) in an installable format or an executable format file and provided.

[0059] Furthermore, the program executed by each device of the above-described embodiment may be configured to be stored on a computer connected to a network such as the Internet and downloaded via the network for providing. Also, the program executed by each device of the above-described embodiment may be configured to be provided or distributed via a network such as the Internet.

[0060] Embodiments of the present invention are presented as examples and are not intended to limit the scope of the invention. The embodiments can be implemented in various other forms, and various omissions, replacements, changes, and combinations can be made without departing from the gist of the invention. These embodiments and their modifications are included in the scope and gist of the invention and are included in the invention described in the claims and its equivalent scope.

Explanation of Reference Numerals

[0061] 1... Server 10... Control Unit 101... Information Acquisition Unit, 102... Image Processing Unit 105... Front-End Processing Unit (an example of acquisition means) 106... Natural Language Processing Unit (an example of generation means) 107... Back-End Processing Unit (an example of creation means) 11... CPU, 12... ROM, 13... RAM 14... Communication Unit 19... Storage Unit, 191... Program 3... In-house DB (an example of the first storage means and the second storage means) 31, 311, 312... Templates 32, 321, 322... Input Rules 5... Image Input Device 50, 51, 52... Documents 511... Document ID Column, 512... Date Column, 513 - 515... Filling Columns 521... Document ID Column, 522... Date Column, 523 - 527... Filling Columns

Prior Art Documents

Patent Documents

[0062]

Patent Document 1

Claims

1. A first storage means for storing a document template; A second storage means for storing input rules for the columns in association with one or more columns included in the template; An acquisition means for acquiring the content entered in the column from the completed document along the template; A generation means for generating a corrected version of the content by correcting the content acquired by the acquisition means based on the input rules associated with the column; A creation means for creating a fair copy of the document by applying the corrected version of the content generated by the generation means to the template; A document creation system comprising:

2. The generation means generates a corrected version of the content by performing at least one of correcting typos and omissions in the content and summarizing the content based on the input rules. The document creation system according to Claim 1.

3. The acquisition means acquires the completed document as image data, and acquires the content by performing image processing on the image data. The document creation system according to Claim 1.

4. The acquisition means acquires text data as the content by the image processing. The document creation system according to Claim 3.

5. The acquisition means does not transfer to the generation means what is determined not to be the target of generating a corrected version by the generation means among the content, and transfers it to the creation means. The document creation system according to Claim 1.

6. A computer included in a document creation system comprising a first storage means for storing a document template and a second storage means for storing input rules for the columns in association with one or more columns included in the template, An acquisition means for acquiring the content entered in the column from the completed document along the template; A generation means for generating a corrected version of the content by correcting the content acquired by the acquisition means based on the input rules associated with the column; A creation means for creating a fair copy of the document by applying the corrected version of the content generated by the generation means to the template; A program for causing the computer to function as such.

Images