Apparatus for generating electronic form document

Abstract

The present invention relates to an electronic form document generation device, characterized by comprising: an application input unit that receives a scanned input of an application form in which items are arranged in a layout structure and the contents for each item are written, and converts the application form image into an image file and stores it; a table area detection module that detects a table area included in the application form image based on a preset deep learning-based detection algorithm and generates table line information corresponding to the outline and internal cell lines of the table area; an item recognition module that detects cell areas for a plurality of cells included in the table area based on the detection result of the table area and the application form image, and classifies each of the plurality of cells into either an item cell or a content cell according to the result of comparing the location, size, and whether a string is included in each cell area with an adjacent cell; and a form document generation unit that generates a form document corresponding to the table format in the form of an electronic document using the generated table line information and item string information corresponding to the item cell. Accordingly, simply scanning and inputting an existing paper application document allows for the automatic generation of an electronic form with blank fields, effectively enabling the immediate preparation of the desired document format without the need for separate document work.

Description

Technology Field

[0001] The present invention relates to an electronic form document generation device that generates an electronic form document in an initial state with blank content using an application form already written based on deep learning technology. Background Technology

[0002] Recently, with the widespread adoption of computers and the rapid advancement of internet technology, there has been an increase in transactions conducted between companies or between companies and customers using electronic documents created in electronic format.

[0003] In other words, business entities can move away from the method of exchanging paper documents to process business between companies or between companies and customers, and instead use electronic forms to conduct transaction activities. Electronic forms can be used by converting office documents such as Word, Hangul, or Hunminjeongeum into PDF (Portable Document Format) documents.

[0004] However, since electronic form documents are output in the form of images, there was a problem in that the process of inserting data was cumbersome and difficult because the data to be inserted had to be received in a basic XML (Extensible Markup Language) format and then converted into a PDF format when attempting to perform editing, such as inserting data.

[0005] In particular, PDF documents cannot be modified unless a separate paid editing tool with PDF editing capabilities is purchased. This presents the inconvenience of having to create a new document with the same format even when one wishes to use the existing table format and change only specific values. Prior art literature

[0006] (Patent Document 0001) KR 10-0710568 B1(Patent Document 0002) KR 10-2007-0067020 A The problem to be solved

[0007] The objective of the present invention is to solve the above problem by providing an electronic form document generation device that generates an electronic form document in an initial state with blank content using an application form already written based on deep learning technology. means of solving the problem

[0008] An electronic form document generation device according to one aspect of the present invention for achieving the above-mentioned purpose comprises: an application input unit that receives a scanned input of an application form in which items are arranged in a layout structure and the contents for each item are written, and converts the application form image into an image file and stores it; a table area detection module that detects a table area included in the application form image based on a preset deep learning-based detection algorithm and generates table line information corresponding to the outline and internal cell lines of the table area; an item recognition module that detects cell areas for a plurality of cells included in the table area based on the detection result of the table area and the application form image, and classifies each of the plurality of cells into either an item cell or a content cell according to the result of comparing the location, size, and whether a string is included in each cell area with an adjacent cell; and a form document generation unit that generates a form document corresponding to the table format in the form of an electronic document using the generated table line information and item string information corresponding to the item cell.

[0009] Preferably, the table area detection module comprises: a boundary line detection unit that detects a plurality of contour lines corresponding to the boundaries of the table area included in the application image through a preset edge detection algorithm; a vertex detection unit that applies a preset vector simplification algorithm to the detected plurality of contour lines to simplify them into a plurality of continuous line segments and then detects the intersection points between adjacent line segments as vertices; and a table line information generation unit that generates and stores table line information corresponding to line segments connecting adjacent vertices in the horizontal or vertical direction.

[0010] Preferably, the item recognition module comprises: a cell area detection unit that detects a plurality of cell areas corresponding to a plurality of cells included in the table area; a string extraction unit that extracts a text string included in each of the plurality of cell areas; and a cell that obtains the pixel coordinates of the location where each of the plurality of cell areas is detected on the application image as location information of the corresponding cell area. It is characterized by including a location acquisition unit and a cell area classification unit that classifies a cell area into one of an item cell and a content cell based on the result of comparing at least one of the length of a text string and the cell area size corresponding to each cell area and an adjacent cell area.

[0011] In addition, the cell area separator is characterized by, when the first cell area is separated into the item cell, comparing the width or height of the first cell area with a plurality of second cell areas located in the same row or column based on matrix index information for the first cell area to check for a match, and separating each of the second cell areas into either the item cell or the content cell depending on whether the text string corresponding to the second cell area in the case of a match is included in the stored item name DB. Effects of the invention

[0012] According to the present invention, by simply scanning and inputting an application form in the form of a pre-existing paper document, an electronic form document with blank spaces in the contents of the application can be automatically generated, thereby providing the effect of immediately preparing a desired document form without separate document work. Brief explanation of the drawing

[0013] FIG. 1 is a block diagram showing the schematic configuration of an electronic form document generation device according to one embodiment of the present invention, and FIG. 2 is a block diagram showing in detail the detailed configuration of each of the table area detection module and the item recognition module of FIG. 1, and FIG. 3 is a diagram illustrating the process of generating a form document by detecting a table area and an item string from an application image using an electronic form document generating device according to an embodiment of the present invention. Specific details for implementing the invention

[0014] Specific details regarding the problem to be solved, the means for solving the problem, and the effects of the invention as described above are included in the embodiments and drawings to be described below. The advantages and features of the present invention, and the methods for achieving them, will become clear by referring to the embodiments described in detail below in conjunction with the accompanying drawings. Throughout the specification, the same reference numerals refer to the same components.

[0015] FIG. 1 is a block diagram showing the schematic configuration of an electronic form document generation device according to one embodiment of the present invention, FIG. 2 is a block diagram showing the detailed configuration of each of the table area detection module and item recognition module of FIG. 1, and FIG. 3 is a diagram for explaining the process of generating a form document by detecting a table area and an item string from an application image by an electronic form document generation device according to one embodiment of the present invention.

[0016] Hereinafter, an electronic form document generating device according to a preferred embodiment of the present invention will be described with reference to the aforementioned drawings.

[0017] An electronic form document generation device according to one embodiment of the present invention is configured to include, as shown in FIG. 1, an application input unit (100), a table area detection module (200), an item recognition module (300), and a form document generation unit (400).

[0018] The application input section (100) serves to receive an application form created by the user by entering items in the form of an image file.

[0019] The application input unit (100) can receive an application (11) with item-specific content written based on a table format in which multiple items are arranged in a layout structure through a user terminal (20) by scanning and inputting it, and can generate and store an application image (I) converted into an image file.

[0020] The table area detection module (200) is for detecting a table area included in an application image (I) stored by the application input unit (100).

[0021] The table area detection module (200) can detect a table area included in the application image (I) based on a preset deep learning-based detection algorithm and obtain table line information corresponding to the outline and internal cell lines of the table area.

[0022] The table area detection module (200) may specifically include a boundary line detection unit (210), a vertex detection unit (220), and a table line information storage unit (230) as shown in FIG. 2.

[0023] The boundary line detection unit (210) detects a plurality of contour lines (L) corresponding to the boundaries of the table area included in the application image through a preset edge detection algorithm. O Detects )

[0024] Here, the edge detection algorithm may include a Canny edge detection algorithm.

[0025] According to the Canny edge detection algorithm, noise in the image is removed through blurring using a Gaussian mask, and the intensity of pixel gradients in the horizontal and vertical directions is detected using a Sobel mask. Then, Non-Maximum Suppression (NMS) is performed to suppress all edges except those with large pixel intensity differences, and Hysteresis Thresholding is applied to detect the final edges.

[0026] The vertex detection unit (220) detects a plurality of contour lines (L) detected by the boundary line detection unit (210). O A preset vector simplification algorithm (Douglas-Peucker) is applied to ) to simplify it into a plurality of consecutive line segments, and then the intersection points between adjacent line segments are detected as vertices (v).

[0027] Here, the vector simplification algorithm may include the Douglas-Peucker algorithm.

[0028] In the case of the Douglas Packer algorithm, it is performed by simplifying a vector object composed of multiple vertices by identifying and removing non-important vertices based on a threshold set according to maximum precision.

[0029] The table line information storage unit (230) generates and stores table line information corresponding to line segments connecting adjacent vertices in the horizontal or vertical direction based on a plurality of vertices (v) detected by the vertex detection unit (220).

[0030] The item recognition module (300) is for recognizing the item portion of the table area based on the table area detected by the table area detection module (200) and the application image (I) stored by the application input unit (100).

[0031] The item recognition module (300) detects cell areas for multiple cells included in a table area and can classify the multiple cells into one of an item cell and a content cell based on the result of comparing the location and size of each cell area and whether a string is included with an adjacent cell.

[0032] The item recognition module (300) may specifically include a cell area detection unit (310), a string extraction unit (320), a cell location acquisition unit (330), and a cell area separation unit (340).

[0033] The cell region detection unit (310) detects multiple cell regions corresponding to multiple cells included in the table region.

[0034] The cell area detection unit (310) detects an area enclosed by a border line corresponding to the table area on all sides as a cell area based on a plurality of contour lines detected by the boundary line detection unit (210), and can calculate the area size of each cell area.

[0035] The string extraction unit (320) extracts text strings contained in each of the multiple cell areas.

[0036] The string extraction unit (320) can extract text strings in each cell area through an OCR (Optical Character Recognition) learning model of an artificial neural network structure that can be easily implemented using deep learning-based Python code, and can also obtain length information of the text strings according to the number of extracted strings.

[0037] The cell location acquisition unit (330) acquires the pixel coordinates of the locations where each of the multiple cell regions is detected on the application image (I) as location information of the corresponding cell region.

[0038] The cell location acquisition unit (330) can convert the location information of each cell area into matrix index information based on the table area and store them together.

[0039] The storage unit (350) is provided in the form of a predetermined database and includes an item name DB that stores name strings of multiple items included in the table format of the application form (11).

[0040] For example, the name strings for multiple items included in the table of the application form (11) shown in FIG. 1, such as ‘Name, English, Occupation, Age, Address, HP, Home, E-mail’, can be stored in the item name DB.

[0041] The cell area separation unit (340) serves to separate multiple cell areas detected by the cell area detection unit (310) into one type of item cell (CI) containing an item string and a content cell (CC) containing a content string.

[0042] The cell area separator (340) separates the cell area into one of an item cell (CI) and a content cell (CC) based on the result of comparing at least one of the length of a text string corresponding to each cell area and the cell area size corresponding to each cell area and the adjacent cell area.

[0043] Here, the adjacent cell is a cell located adjacent in the horizontal or vertical direction relative to the location of a specific cell; in other words, it may refer to a cell located immediately next to, above, or below in the same row or column.

[0044] Based on the above comparison result, the cell area separator (340) can separate the cell area corresponding to the case where the text string is longer or the cell area size is larger among the two cell areas being compared as a content cell (CC), and separate the remaining cell area as an item cell (CI).

[0045] At this time, the cell area separator (340) can immediately classify the type of the cell area as a content cell (CC) if the number of text strings extracted corresponding to a specific cell area is greater than a preset standard number, and in this case, it may not perform a comparison with the aforementioned adjacent cell area.

[0046] The cell area classification unit (340) performs type classification of adjacent surrounding cell areas based on the location of the cell area when a specific cell area is classified as an item cell (CI) according to the process described above.

[0047] If the first cell area (C1) is divided into item cells, the cell area division unit (340) checks whether there is a match by comparing the horizontal or vertical length of the first cell area (C1) with a plurality of second cell areas (C2) located in the same row or column based on matrix index information for the first cell area (C1).

[0048] The cell area separator (340) separates each second cell area (C2) into an item cell (CI) and a content cell (CC) depending on whether the text string corresponding to the second cell area (C2) that matches the above verification result is included in the stored item name DB (350).

[0049] Here, the cell area separator (340) may separate the cell area (C2) as an item cell (CI) if the text string extracted corresponding to the second cell area (C2) is included in the item name DB (350), and separate the second cell area (C2) as a content cell (CC) otherwise.

[0050] This takes into account that, generally when creating a table by item, the item name is written in a cell at the same column or row position, and the item value for that item is written in a cell located immediately to the right or in the row below the position of the cell containing the specific item.

[0051] The form document generation unit (400) generates a form document (12) corresponding to the table format of the application form (11) in the form of an electronic document using the table line information stored in the table line information generation unit (230) and the item string information corresponding to the item cell according to the division of the cell area division unit (340).

[0052] As shown in FIG. 3, the form document generation unit (400) can generate a form document (34) by receiving table line information (32) generated in correspondence with the table area (31) included in the application image (I) from the table line information generation unit (230) and item string information (33) extracted from the string extraction unit (320) for the cell area separated into item cells (CI) by the cell area separation unit (340), and merging them so that the pixel positions for the outlines of the corresponding table area match each other.

[0053] Although the present invention has been described in detail through preferred embodiments, the invention is not limited thereto and can be implemented in various ways within the scope of the claims.

[0054] In particular, since the foregoing has described the features and technical strengths of the present invention somewhat broadly to better understand the claims of the invention to be described below, it should be recognized by those skilled in the art that the concept and specific embodiments of the present invention described above can be immediately used as a basis for designs or modifications of other forms to perform purposes similar to those of the present invention.

[0055] Furthermore, the embodiments described above are merely one embodiment according to the present invention, and it will be understood by those skilled in the art that various modifications and changes may be implemented within the scope of the technical concept of the present invention. Accordingly, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense, and such various modifications and changes are also indicated in the claims of the present invention as falling within the scope of the technical concept of the present invention, and all differences within the equivalent scope should be interpreted as being included in the present invention. Explanation of the symbols

[0056] 10: Electronic form document generation device 20: User terminal 100: Application Input Section 200: Table area detection module 210: Boundary detection unit 220: Vertex detector 230: Table Line Information Generation Section 300: Item recognition module 310: Cell area detection unit 320: String extraction section 330: Cell location acquisition unit 340: Cell range separator 350: Storage 360: Template Creation Section

Claims

Claim 1 An application input unit that receives a scanned input of an application form in which item-specific content is written based on a table format in which multiple items are arranged in a layout structure, converts it into an image file, and generates and saves an application image; a table area detection module that detects a table area included in the application image based on a preset deep learning-based detection algorithm and generates table line information corresponding to the outline and internal cell lines of the table area; an item recognition module that detects cell areas for multiple cells included in the table area based on the detection result of the table area and the application image, and classifies each of the multiple cells into either an item cell or a content cell according to the result of comparing the location, size, and whether a string is included in each cell area with an adjacent cell; and a form document generation unit that generates a form document corresponding to the table format in the form of an electronic document using the generated table line information and item string information corresponding to the item cell; wherein the item recognition module includes a cell area detection unit that detects multiple cell areas corresponding to multiple cells included in the table area, a string extraction unit that extracts a text string included in each of the multiple cell areas, and a cell that obtains the pixel coordinates of the location where each of the multiple cell areas is detected on the application image as location information of the corresponding cell area. An electronic form document generation device comprising a location acquisition unit and a cell area classification unit that classifies a cell area into one of an item cell and a content cell based on the result of mutually comparing at least one of the length of a text string corresponding to each cell area and an adjacent cell area, respectively, and wherein the cell area classification unit classifies the cell area corresponding to the case where the text string is longer or the cell area size is larger among each cell area and the adjacent cell area as a content cell and classifies the remaining cell area as an item cell, wherein when a first cell area is classified as an item cell, the horizontal or vertical length of the first cell area is checked by mutually comparing a plurality of second cell areas located in the same row or column with the first cell area based on matrix index information for the first cell area to determine whether they match, and wherein each second cell area is classified into one of the item cell and the content cell depending on whether the text string corresponding to the second cell area in the case of a match is included in a pre-stored item name DB. Claim 2 An electronic form document generation device according to claim 1, wherein the table area detection module comprises: a boundary line detection unit that detects a plurality of contour lines corresponding to the boundaries of a table area included in the application image through a preset edge detection algorithm; a vertex detection unit that applies a preset vector simplification algorithm to the detected plurality of contour lines to simplify them into a plurality of continuous line segments and detects the intersection points between adjacent line segments as vertices; and a table line information generation unit that generates and stores table line information corresponding to line segments connecting adjacent vertices in the horizontal or vertical direction. Claim 3 delete Claim 4 delete

Images