KR20220132213A - Methods and systems for recognizing tables - Google Patents

Abstract

The present invention relates to a method and system for correcting an image including a table by performing analysis on the image. According to the present invention, the method for correcting an image may include: a step of receiving an image including a table; a step of extracting a plurality of lines forming the table from the image; a step of calculating a degree of rotation of the plurality of lines based on a reference line; and a step of correcting the image to perform rotation for the table based on the degree of rotation.

Description

METHODS AND SYSTEMS FOR RECOGNIZING TABLES

The present invention relates to a method and system for correcting an image including a table by performing an analysis on the image.

The dictionary meaning of artificial intelligence is a technology that realizes human learning ability, reasoning ability, perceptual ability, and natural language understanding ability through computer programs. Such artificial intelligence made a leap forward due to deep learning, which added a neural network that mimics the human brain to machine learning.

Deep learning is a technology that allows a computer to judge and learn like a human, and to cluster or classify objects or data through this. It is actively used in the industrial field.

With the development of artificial intelligence, various automations are being made in the field of office automation. In particular, in the field of office automation, much effort is being made to convert content printed on paper into data based on image data analysis technology using artificial intelligence. As part of that, in the field of office automation, the content included in the document is converted into data through image analysis technology (or analysis technology for image data) that images the paper document and analyzes the content included in the image. It is necessary to have a technique for analyzing images according to the characteristics of the content included in the .

For example, when dataizing a document containing a table (table), accurate information about various elements surrounding the table, such as the format of the table, the content of the text included in the table, and the location of the text included in the table, etc. analysis is needed

Accordingly, Korean Patent Registration No. 1907029 (Table Generating Apparatus and Method for Automating Forms) discloses a method for recognizing a table from an image and reproducing the recognized table.

As such, various developments are being made on techniques for reproducing the recognized table, and in order to accurately recognize various elements surrounding the table, it is also very important to improve the recognition performance of the table included in the image.

An object of the present invention is to provide a method and system for correcting an image including a table that can increase the recognition performance of the table included in the image.

More specifically, the present invention is to provide a method and system for correcting an image including a table that can improve the recognition performance of the table in order to reproduce a table having the same configuration as the table included in the image.

In order to solve the above problems, the method of correcting an image including a table according to the present invention includes the steps of receiving an image including a table, and extracting a plurality of lines constituting the table from the image The method may include calculating the degree of rotation of the plurality of lines based on a reference line, and correcting the image so that rotation of the table is performed based on the degree of rotation.

Furthermore, the system for correcting an image including a table according to the present invention includes a storage unit, a receiver for receiving an image including a table, and a control unit for extracting a plurality of lines constituting the table from the image Including, the control unit may calculate the degree of rotation of the plurality of lines based on a reference line, and correct the image so that the rotation of the table is performed based on the degree of rotation.

Furthermore, the program that is executed by one or more processes in the electronic device and can be stored in a computer-readable recording medium includes the steps of: receiving an image including a table; extracting lines of , calculating the degree of rotation of the plurality of lines based on a reference line, and correcting the image so that rotation of the table is performed based on the degree of rotation It may contain instructions.

As described above, in the image correction method and system including a table according to the present invention, when the table included in the image is rotated, the recognition performance of the table can be improved by correcting the rotation.

More specifically, the image correction method and system including the table according to the present invention can extract a plurality of lines constituting the table, and calculate the degree of rotation thereof. Furthermore, the table recognition method and system according to the present invention determine how much the table is rotated relative to the reference line based on this degree of rotation, and based on the determined result, the table rotation is corrected for the image. Calibration may be performed.

Such image correction may correspond to a pre-processing process for increasing the recognition rate of information constituting the table, and as a result, the image correction method and system including the table according to the present invention has the recognition performance of the table included in the image. can increase Furthermore, when the recognition performance for a table is improved by the image correction method and system including the table according to the present invention, it is possible to create a table having the same structure as the table included in the image.

1 is a conceptual diagram for explaining an image correction system including a table according to the present invention.
2 is a conceptual diagram for explaining a table.
3 is a flowchart illustrating an image correction method including a table according to the present invention.
4, 5, 6A, 6B and 7, 8, and 9 are conceptual diagrams for explaining an image correction method including a table according to the present invention.

Hereinafter, the embodiments disclosed in the present specification will be described in detail with reference to the accompanying drawings, but the same or similar components will be given the same reference numerals regardless of reference numerals, and redundant descriptions thereof will be omitted. The suffixes "module" and "part" for components used in the following description are given or mixed in consideration of only the ease of writing the specification, and do not have distinct meanings or roles by themselves. In addition, in describing the embodiments disclosed in the present specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in this specification, the detailed description thereof will be omitted. In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in this specification, and the technical idea disclosed herein is not limited by the accompanying drawings, and all changes included in the spirit and scope of the present invention , should be understood to include equivalents or substitutes.

Terms including ordinal numbers such as first, second, etc. may be used to describe various elements, but the elements are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

When a component is referred to as being “connected” or “connected” to another component, it may be directly connected or connected to the other component, but it is understood that other components may exist in between. it should be On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that the other element does not exist in the middle.

The singular expression includes the plural expression unless the context clearly dictates otherwise.

In the present application, terms such as “comprises” or “have” are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It is to be understood that this does not preclude the possibility of the presence or addition of numbers, steps, operations, components, parts, or combinations thereof.

As discussed above, with the development of artificial intelligence, office automation is being implemented in various forms. The need for digitalization (dataization) as it is is gradually increasing.

For example, various types of documents such as receipts contain tables, and insurance companies, hospitals, etc. need to process tables included in large amounts of paper documents as digitized data and computerize them. have.

Based on these needs, development of a technique for generating or reproducing a table having the same configuration as a table included in an image is being actively developed.

On the other hand, it is very important to accurately recognize various elements surrounding the table (eg, text, text position, cell configuration, cell position, relation between cells, etc.) do.

To this end, it is important to improve the recognition performance of the table included in the image itself, and the present invention provides an image correction method and system including a table that can increase the recognition performance of the table included in the image.

Hereinafter, an image correction method and system including a table capable of increasing table recognition performance will be described in more detail with accompanying drawings.

1 is a conceptual diagram for explaining an image correction system including a table according to the present invention, and FIG. 2 is a conceptual diagram for explaining the table. Furthermore, FIG. 3 is a flowchart for explaining an image correction method including a table according to the present invention, and FIGS. 4, 5, 6A, 6B, 7, 8 and 9 are tables according to the present invention. These are conceptual diagrams for explaining the included image correction method.

First, as shown in FIG. 1 , the table recognition system 100 according to the present invention recognizes a table included in an image, extracts a line or a line constituting the recognized table, and extracts the extracted line Based on the , it can be determined whether the table included in the image is rotated.

As shown, the image 1000 may include one or more tables 1100 , 1200 , and 1300 , and in the present invention, regardless of the number of tables included in the image 1000 , the image 1000 may include It is possible to recognize all tables.

In the present invention, the image 1000 may be an image obtained by scanning a paper document, an image obtained by photographing, or an image obtained by various other methods.

Meanwhile, as shown in FIG. 1 , the table recognition system 100 according to the present invention may include at least one of a reception unit 110 , a storage unit 120 , an OCR unit 130 , and a control unit 140 . .

The receiver 110 is a means for receiving the image 1000 including the tables 1100, 1200, and 1300, and is configured to include at least one of a communication unit, a scan unit, and an input unit, or can receive other images 1000 It can be done by any means available.

The table recognition system 100 analyzes the tables 1100 , 1200 , and 1300 included in the image 1000 received through the receiver 110 , and the tables 1100 , 1200 , 1300 included in the image 1000 and A preprocessing process may be performed so that the same table is created or data (or information) included in the tables 1100 , 1200 , and 1300 is accurately extracted.

As such, the operation of creating the same table as the tables 1100, 1200, 1300 included in the image 1000 or extracting the information included in the tables 1100, 1200, 1300 into data is “reproduce the table, Reconstruction or restoration (reconstruction, restoration)” can be expressed.

In this specification, recognizing the tables 1100, 1200, and 1300 included in the image 1000, and correcting the image 1000 or the tables 1100, 1200, 1300 based on the recognition result, is to the image 1000. It may be understood as a part of a process for restoring the included tables 1100 , 1200 , and 1300 .

That is, the table recognition system 100 according to the present invention may be understood as a concept included in a table creation (or restoration) system.

On the other hand, the restoration of the tables 1100, 1200, and 1300 included in the image 1000 is performed on the image 1000 or the tables 1100, 1200, and 1300 through the table recognition system 100 according to the present invention. This can be done after the correction has been made.

Next, the storage unit 120 may be configured to store various information according to the present invention. The types of the storage unit 120 may be very diverse, and at least some of them may mean an external server 150 (at least one of the cloud server 151 and the database (DB, 152 )). That is, it can be understood that the storage unit 120 is sufficient as long as a space in which information related to the present invention is stored, and there is no restriction on the physical space.

In the storage unit 120, i) the image 1000 including the tables 1100, 1200, 1300 and related data, ii) the components (lines) of the tables 1100, 1200, and 1300 extracted from the image 1000 , corner, etc.) and related data, and iii) content (eg, text, image, etc.) data included in the tables 1100 , 1200 , and 1300 may be stored.

Next, the OCR (Optical Character Reader, 130) unit is a means for recognizing the content included in the image 1000, and may recognize the content included in the image 1000 through at least one of various content recognition algorithms. The OCR unit 130 may recognize the content by using an algorithm based on artificial intelligence. Here, the content may include text (characters).

The OCR unit may extract text included in the image and location information of the text. Here, the location information of the text may include information on where the text is located in the image 1000 input through the receiver 110 or where the text is located in the tables 1100 , 1200 , and 1300 . have.

Next, the controller 140 may be configured to control the overall operation of the table recognition system 100 related to the present invention. The controller 140 may include a processor (or artificial intelligence processor) capable of processing an artificial intelligence algorithm. The controller 140 may recognize the tables 1100 , 1200 , and 1300 from the image 1000 based on a deep learning algorithm, and extract a plurality of lines constituting the tables 1100 , 1200 , and 1300 . Then, the controller 140 calculates the degree of rotation of the plurality of lines based on the reference line, and based on the calculated degree of rotation, the image 1000 so that the rotation of the tables 1100 , 1200 , and 1300 is performed. can be corrected.

That is, the controller 140 controls the tables 1100 , 1200 , and 1300 as the image 1000 is rotated as shown in FIG. 1 so that accurate information can be extracted from the tables 1100 , 1200 , and 1300 . ) is rotated, correction can be performed to correct it.

As shown, when the image 1000 is rotated to the left by a predetermined angle, the control unit 140 performs a correction to rotate the image 1000 by the predetermined angle to the right, so that the table ( 1100, 1200, and 1300) may be positioned in the forward direction.

At this time, the controller 140 determines in which direction and by what angle the image 1000 should be rotated based on the plurality of lines constituting the tables 1100 , 1200 , and 1300 included in the image 1000 . can judge

Meanwhile, prior to the description of the present invention, a “table” referred to in the present specification may also be referred to as a table, and may mean that certain contents are displayed in a certain format or order. Referring to the components of the table with reference to FIG. 2 , the table 200 may include at least one cell 210 . That is, the table 200 includes at least one cell or space (or region), and the table is configured to include information in the cell 210 .

The cell 210 included in the table may be defined by at least four line segments. The table 200 and the shape of the cells included in the table 200 are rectangular.

That is, a cell included in the table may be formed of a rectangle surrounded by four line segments. In the present invention, a “line segment” may be understood as a line or at least a part of a line. Line segments constituting each cell may be extended to form a line or a line.

As described above, in order to form one square-shaped cell, at least four line segments (a, b, c, d) are required, which are two horizontal lines constituting the table, or a line in the row direction. , or an edge, 201, 202, and two vertical lines, or a line in a column direction, or an edge, 203, 204 may be formed by at least a portion. More specifically, the rectangular-shaped cell may be formed by at least a portion of the two horizontal lines 201 and 202 and the two vertical lines 203 and 204 crossing each other.

Accordingly, the table 200 includes at least four lines, and the number of horizontal lines, the number of vertical lines, the spacing between lines, the number of cells, the size of the cells, the position of the cells, etc. are defined according to the arrangement position of the lines. can be

On the other hand, the plurality of lines constituting the table 200 may be divided into any one type among different first and second types, which may be specified according to which direction the corresponding line extends.

As illustrated, the lines 201 and 202 extending in the horizontal direction may be defined as the first type, and the lines 203 and 204 extending in the vertical direction may be defined as the second type.

The first type of line may be variously named as a horizontal line, a horizontal line, a row line, a horizontal line, etc. as terms having the same or similar meaning.

Furthermore, the second type of line may be variously named as a term having the same or similar meaning as a vertical line, a vertical line, a column line, a vertical line, and the like.

On the other hand, the number of cells included in the table 200, the size of the cells, the positions of cells, etc. may be variously modified based on the number of horizontal and vertical lines, arrangement relationships, etc., and may be variously modified by merging cells can be

Furthermore, the components constituting the table described in the present invention are a corner (or vertex, reference numeral) formed while the lines constituting the table (horizontal lines and vertical lines) and the lines constituting the table intersect. “e, f, g, h”).

All tables have at least 4 corners, which may be corners included in the outermost edge of the table. Furthermore, the number of corners included in the table depends on the number of cells included in the table.

Hereinafter, based on the configuration of the table recognition system according to the present invention as described above, a method for generating a table will be described in more detail.

Referring to FIG. 3 , in the table recognition method according to the present invention, a process of receiving an image 1000 (refer to FIG. 1 ) including tables 1100 , 1200 , and 1300 may be performed ( S310 ).

As described above, the image 1000 including the tables 1100 , 1200 , and 1300 may be received through various routes. For example, the image 1000 may be transmitted through a communication unit, scanned through a scan unit, or received through an input unit.

When the image 1000 is received, next, in the present invention, a process of extracting a plurality of lines constituting the tables 1100 , 1200 , and 1300 may be performed ( S320 ).

As described above, the tables 1100 , 1200 , and 1300 may include horizontal lines and vertical lines.

The control unit 140 recognizes an area in which horizontal lines and vertical lines constituting the tables 1100, 1200, and 1300 are located from the image 1000 based on the machine learning (ML) method, and from the recognized area, the table 1100, 1200 and 1300 ) and horizontal lines and vertical lines constituting them may be extracted.

As described above, the plurality of lines constituting the tables 1100 , 1200 , and 1300 may be divided into any one type among different first and second types, which depends on which direction the corresponding line extends. can be specified accordingly.

For example, as shown in (b) of FIG. 4 , a line extending in the horizontal direction is defined as the first type of line, and as shown in FIG. 4 ( b ), a line extending in the vertical direction A line may be defined as a second type of line.

The first type of line may be variously named as a horizontal line, a horizontal line, a row line, a horizontal line, etc. as terms having the same or similar meaning.

Furthermore, the second type of line may be variously named as a term having the same or similar meaning as a vertical line, a vertical line, a column line, a vertical line, and the like.

The controller 140 may extract a plurality of lines from the image 1000 by using various image processing methods (eg, Hough transform, etc.).

Meanwhile, as shown in FIGS. 4A and 5A , the image 1000 may include a plurality of tables 1100 , 1200 , and 1300 . In this case, the controller 140 may extract a plurality of lines from the first table 1100 , the second table 1200 , and the third table 1300 included in the image 1000 , respectively.

More specifically, as shown in (a) of Figure 4, the control unit 140 from the first table 1100, the second table 1200, and the third table 1300, in Figure 4 (b) As illustrated, the first type of line, that is, the horizontal lines 1100a , 1200a , and 1300c may be extracted, respectively.

Furthermore, as shown in (a) of FIG. 5 , the control unit 140 receives data from the first table 1100 , the second table 1200 , and the third table 1300 as shown in FIG. 5 ( b ). As such, the second type of line, that is, the vertical lines 1100b, 1200b, and 1300b, may be extracted, respectively.

Information on the horizontal lines 1100a, 1200a, and 1300c and the vertical lines 1100b, 1200b, and 1300b extracted in this way may be stored in the storage unit 120 .

In this way, when a plurality of lines are extracted from the tables 1100 , 1200 , and 1300 , in the present invention, a process of extracting the degree of rotation of the plurality of lines may be performed ( S330 ).

More specifically, the controller 140 may calculate the degree of rotation of the plurality of lines based on the reference line.

As shown in (a) and (b) of FIG. 6A , the reference line 600 is a line (or line) serving as a reference, and may also be referred to as a “reference line”.

The reference line 600 is a concept serving as a reference for determining the degree of rotation of a plurality of lines, and may be expressed in various terms. For example, the reference line 600 may be replaced with various terms, such as a reference, a specific reference, a reference point, an origin, a zero point, a zero-degree line, and the like.

The reference line 600 may be a reference line for calculating the degree of rotation of the plurality of lines 1100a, 1200a, 1300a, 1100b, 1200b, and 1300b extracted from the tables 1100 , 1200 , and 1300 .

When each of the plurality of lines intersects the reference line 600 , the controller 140 may calculate the degree of rotation based on an angle formed by each of the plurality of lines with the reference line 600 .

Here, the degree of rotation may be related to at least one of a rotation angle and a rotation direction formed by each of the plurality of lines with respect to the reference line 600 .

As shown in (a) and (b) of FIG. 6B , the control unit 140 extracts the rotation direction of the lines 610 and 620 extracted from the table based on the reference line 600, and the reference line ( 600) and the angles θ1 and θ2 formed by the lines 610 and 620 extracted from the table may be calculated.

The controller 140 may perform the above process for all lines extracted from the tables 1100 , 1200 , and 1300 .

Alternatively, the controller 140 may perform the above process with respect to at least some of all lines extracted from the tables 1100 , 1200 , and 1300 .

On the other hand, based on the reference line 600, the rotation direction of the lines 610 and 620 extracted from the table is counterclockwise as shown in (a) of FIG. 6b and shown in (b) of FIG. 6b. As such, it may be either clockwise.

Furthermore, the controller 140 considers the angle (or the angle between the lines, θ1, θ2) between the reference line 600 and the lines 610 and 620 extracted from the table within a preset range (-90 degrees to +90 degrees). can That is, within a preset range (or angle range), the controller 140 may extract an angle (interval angle) between the reference line 600 and the lines 610 and 620 extracted from the table.

The reference line 600 and the line 610 extracted from the table form a rotation angle within a range of up to 90 degrees counterclockwise, or within a range of up to 90 degrees clockwise (corresponding to -90 degrees). It is possible to achieve an angle between

For example, as shown in (a) of FIG. 6B, when the rotation angle θ1 of the extracted line 610 is +30 degrees (when the angle between them is +30 degrees), this is the extracted line 610 This may mean that it is rotated 30 degrees counterclockwise with respect to the reference line 600 .

Furthermore, as shown in (b) of FIG. 6B , when the rotation angle θ2 of the extracted line 620 is -30 degrees (when the angle between the lines is -30 degrees), this means that the extracted line 620 is the reference It may mean rotated 30 degrees clockwise with respect to line 600 .

In this way, the control unit 140 determines which direction (for example, It is possible to extract (or calculate) how many degrees rotated counterclockwise or clockwise), which can be expressed within the range of -90 degrees to +90 degrees.

These, -90 degrees to +90 degrees may be a range of the angle between the angles. That is, the angle between the angles may represent the rotated angle and the rotation direction together.

That is, the sign of + or - means whether the extracted lines 1100a, 1200a, 1300a, 1100b, 1200b, 1300b are rotated counterclockwise or clockwise with respect to the reference line 600. have.

In this way, the control unit 140 calculates the degree of rotation of each of the plurality of lines with respect to the reference line through the process of calculating the rotation angle and the rotation direction of each of the plurality of lines with respect to the reference line (or can be extracted).

Meanwhile, in the present invention, the reference line may correspond to the first type of line, that is, a horizontal line. In this case, the controller 140 may calculate the degree of rotation by performing different calculations according to which type of the first type and the second type each of the plurality of lines extracted from the table corresponds to.

First, with respect to the first-type lines 1100a , 1200a , and 1300a having the same type as the reference line among the plurality of lines, the controller 140 performs the same method as described above, and the degree of rotation of the first-type lines can be extracted.

That is, the controller 140 determines the angle of intersections within a preset range (or preset angle range) formed by each of the reference line 600 and the first type of lines 1100a, 1200a, 1300a. The degree of rotation of the first type of lines 1100a , 1200a , and 1300a may be extracted through the calculation.

Next, the controller 140 extracts the degree of rotation of the first type of lines 1100b, 1200b, and 1300b of the second type of lines 1100b, 1200b, and 1300b having a different type from the reference line 600 among the plurality of lines. In the operation to be performed, additional operations may be performed.

The control unit 140 performs a first operation for calculating angles of intersections within a preset range formed by each of the reference line 600 and the second type lines 1100b, 1200b, and 1300b, and the first A second operation may be performed by adding a preset angle to an angle corresponding to the result of the operation. Here, the preset angle may correspond to +90 degrees.

That is, the control unit 140 adds 90 degrees (+90 degrees) counterclockwise to the angle between the reference line 600 and the second type lines 1100b, 1200b, and 1300b within a preset range, respectively. , to obtain the corrected angle of inclination. Such a corrected angle between the angles may also be expressed as “a degree of rotation of the second type of lines”.

For example, when the interstitial angle corresponding to the result of the first operation of the specific second type of line is -70 degrees, the corrected interstitial angle may be +20 degrees. In this case, it may be understood that the rotation degree of the specific second type of line is rotated by 20 degrees counterclockwise with respect to the reference line.

This second operation may be understood as an operation for converting the lines of the second type into the same type as the line type (first type) of the reference line, which is a plurality of This is to consider the degree of rotation of the lines together.

On the other hand, the controller 140 determines the degree of rotation of each of the first-type lines 1100a, 1200a, and 1300a with respect to the reference line 600 calculated in the manner described above and the second-type lines 1100b, 1200b, and 1300b. ) in consideration of the respective rotational degrees, the median value (or median) can be calculated.

The process of calculating the intermediate value includes the degree of rotation (or the angle between the angles) of each of the first type of lines 1100a, 1200a, and 1300a and the degree of rotation of the second type of lines 1100b, 1200b, and 1300b ( Alternatively, it may be understood as a process of calculating an intermediate value for the corrected angle of indentation).

The median value is the rotation degree (or angle between each of the first types of lines 1100a, 1200a, and 1300a) and the degree of rotation (or the corrected angle between the lines) of the second type of lines 1100b, 1200b, and 1300b, respectively. It may mean a value located in the center of the corresponding value. The control unit 140 controls the intermediate values (hereinafter, " Intermediate angle”) can be calculated.

For example, when there are 8 lines 1100a, 1200a, and 1300a of the first type, 8 angles between them may exist. Similarly, when there are 8 lines 1100b, 1200b, and 1300b of the second type, 8 corrected angles may exist.

In this case, the 8 angles and the 8 corrected angles are +20, +17, +18, +15, +20, +20, +15, +15, +15, +25, +15, +15, respectively. Assume +16, +15, +13, +20.

In this case, the intermediate angle between the eight angles and the eight corrected angles may be calculated as +15.5.

In this way, when the degree of rotation of the plurality of lines is calculated, in the present invention, based on the degree of rotation, a process of correcting the image so that the rotation of the table is performed may proceed (S340).

In more detail, the control unit 140 provides an intermediate angle between each of the first-type lines 1100a, 1200a, and 1300a and the corrected angles between each of the second-type lines 1100b, 1200b, and 1300b. median value) can be used to correct the image.

The controller 140 may correct the image 1000 including the tables 1100 , 1200 , and 1300 so that rotation of the tables 1100 , 1200 , and 1300 is performed based on the intermediate angle.

The controller 140 may perform a correction in which the image 1000 is rotated in one direction by a specific angle corresponding to the degree of rotation with respect to the reference line corresponding to the intermediate angle. Here, the one direction may be a rotation direction opposite to the rotation direction corresponding to the intermediate angle. For example, when the rotation direction corresponding to the intermediate angle is a counterclockwise direction, one direction may be a clockwise direction.

More specifically, the controller 140 controls the image 1000 by an angle corresponding to the intermediate angle in the rotation direction (counterclockwise or clockwise) opposite to the rotation direction (clockwise or counterclockwise) corresponding to the intermediate angle. can be rotated. For example, when the intermediate angle between the plurality of lines of the tables 1100, 1200, and 1300 included in the image 1000 shown in FIGS. 4A and 4B is +20 degrees, this is the table ( 1100 , 1200 , and 1300 ) may mean that the reference line is rotated counterclockwise by about 20 degrees.

In this case, the controller 140 rotates the image by 20 degrees in the clockwise direction as shown in FIG. The image 1000' may be corrected. That is, the controller 130 may rotate the image by -20 degrees.

On the other hand, rotating the image 1000 by an angle corresponding to the intermediate angle means the tables 1100 , 1200 , 1300 included in the image 1000 , that is, the first type of lines 1100a and 1200a included in the image. , 1300a) and the second type of lines 1100b, 1200b, and 1300b may have the same meaning as rotating.

The control unit 140, as shown in (a), (b) and (c) of FIG. 7 , the first type of lines 1100a' corresponding to the first table 1100, the second table ( The first-type lines 1200a' corresponding to 1200 and the first-type lines 1300a' corresponding to the third table 1300 may be rotated by an angle corresponding to the intermediate angle. More specifically, the controller 140 may rotate the lines in a direction opposite to the rotation direction corresponding to the intermediate angle. As a result, the first type of lines 1100a ′ corresponding to the first table 1100 , the first type of lines 1200a ′ corresponding to the second table 1200 , and the third table 1300 ) The first type of lines 1300a ′ corresponding to may be disposed parallel to or similar to the reference line 620 .

Similarly, as shown in (a), (b) and (c) of FIG. 8 , the control unit 140 controls the second type of lines 1100b' corresponding to the first table 1100, the second The second type of lines 1200b ′ corresponding to the table 1200 and the second type of lines 1300b ′ corresponding to the third table 1300 may be rotated by an angle corresponding to an intermediate angle between them. . More specifically, the controller 140 may rotate the lines in a direction opposite to the rotation direction corresponding to the intermediate angle. As a result, the first type of lines 1100b' corresponding to the first table 1100, the first type of lines 1200b' corresponding to the second table 1200, and the third table 1300 The corresponding first type of lines 1300b may be disposed perpendicular to or similar to the reference line 620 .

On the other hand, when the image 1000' is rotationally corrected based on the intermediate angle as shown in (a) of FIG. , 1300') may extract at least one of information characteristics and structure characteristics.

That is, the controller 140 may process at least one of the structure of the table and information included in the table from the rotated table.

Here, the extraction of the information properties included in the tables 1100', 1200', 1300' is the text (or characters) included in the tables 1100', 1200', 1300', or the rotated table and the text It may include extracting the location information of

Furthermore, in the extraction of the structural characteristics included in the tables 1100', 1200', 1300', at least one cell included in the tables 1100', 1200', 1300', or a rotated table is recognized and recognized It may include extracting the association between the cells.

Through the extraction of information characteristics and structure characteristics, various information can be extracted and dataized in the tables 1100', 1200', and 1300'.

On the other hand, information corresponding to these information characteristics and structure characteristics may be stored in the low wooop (120). Furthermore, information according to information characteristics may be extracted through the above-described OCR unit 130 .

On the other hand, the control unit 130, based on the completion of the extraction of the information characteristics and the structural characteristics from the rotated tables (1100', 1200', 1300'), the rotated tables (1100', 1200', 1300') can be re-rotated again.

The control unit 130 rotates the tables 1100', 1200', and 1300' shown in FIG. 9(a) in a direction (eg, clockwise) corresponding to the angle between the middle angles in FIG. 9(b) As shown in , to be rotated again, the image (or rotated A correction can be performed to rotate the image, 1000').

This is for the user's psychological reliability of the data extracted from the image 1000, and in the present invention, the psychological stability and reliability that data is extracted from the first scanned image 1000 to the user through the above process can give

As described above, in the image correction method and system including a table according to the present invention, when the table included in the image is rotated, the recognition performance of the table can be improved by correcting the rotation.

More specifically, the image correction method and system including the table according to the present invention can extract a plurality of lines constituting the table, and calculate the degree of rotation thereof. Furthermore, the table recognition method and system according to the present invention determine how much the table is rotated relative to the reference line based on this degree of rotation, and based on the determined result, the table rotation is corrected for the image. Calibration may be performed.

Such image correction may correspond to a pre-processing process for increasing the recognition rate of information constituting the table, and as a result, the image correction method and system including the table according to the present invention has the recognition performance of the table included in the image. can increase Furthermore, when the recognition performance for a table is improved by the image correction method and system including the table according to the present invention, it is possible to create a table having the same structure as the table included in the image.

Meanwhile, the present invention described above may be implemented as a program storable in a computer-readable medium (or recording medium), which is executed by one or more processes in a computer.

Furthermore, the present invention as seen above can be implemented as computer-readable codes or instructions on a medium in which a program is recorded. That is, the present invention may be provided in the form of a program.

Meanwhile, the computer-readable medium includes all types of recording devices in which data readable by a computer system is stored. Examples of computer-readable media include Hard Disk Drive (HDD), Solid State Disk (SSD), Silicon Disk Drive (SDD), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. There is this.

Furthermore, the computer-readable medium may be a server or a cloud storage that includes a storage and that an electronic device can access through communication. In this case, the computer may download the program according to the present invention from a server or cloud storage through wired or wireless communication.

Furthermore, in the present invention, the computer described above is an electronic device equipped with a processor, that is, a CPU (Central Processing Unit, Central Processing Unit), and there is no particular limitation on the type thereof.

On the other hand, the above detailed description should not be construed as limiting in all respects, but should be considered as exemplary. The scope of the present invention should be determined by a reasonable interpretation of the appended claims, and all modifications within the equivalent scope of the present invention are included in the scope of the present invention.

Claims (15)

receiving an image comprising a table;
extracting a plurality of lines constituting the table from the image;
calculating a degree of rotation of the plurality of lines based on a reference line; and
and correcting the image so that rotation of the table is performed based on the degree of rotation. According to claim 1,
In the step of calculating the degree of rotation,
The image correction method, characterized in that for calculating the rotation angle and rotation direction of each of the plurality of lines with respect to the reference line. 3. The method of claim 2,
Each of the plurality of lines constituting the table,
It is divided into any one type of a first type extending in a horizontal direction and a second type extending in a vertical direction,
The degree of rotation of each of the plurality of lines with respect to the reference line is,
The image correction method according to claim 1, wherein each of the plurality of lines is calculated through different calculations depending on which type of the first type and the second type correspond to. 4. The method of claim 3,
When the reference line is a horizontal line extending in a horizontal direction,
The degree of rotation of the first type of lines among the plurality of lines is,
The image correction method, characterized in that it is calculated through an operation to obtain an angle of intersections within a preset range between the reference line and each of the first type of lines. 5. The method of claim 4,
The degree of rotation of the second type of lines among the plurality of lines is,
The image correction method, characterized in that the calculation is calculated by adding a preset angle to the angle of intersections within a preset range between the reference line and each of the second type of lines. 6. The method of claim 5,
The preset angle is
and the angle is such that the second type of lines become the first type of lines. 6. The method of claim 5,
In the step of calculating the degree of rotation,
The method further includes calculating a median for the rotational degree of the first type of lines and the rotational degree of the second type of lines,
In the step of correcting the image,
and correcting the image so that rotation of the table is performed based on the degree of rotation corresponding to the intermediate value. According to claim 1,
When a plurality of tables are included in the image, horizontal lines and vertical lines constituting each of the plurality of tables are extracted. According to claim 1,
In the step of correcting the image,
An image correction method, characterized in that performing correction by rotating the image in one direction by a specific angle corresponding to the degree of rotation. 10. The method of claim 9,
and extracting, from the rotated table, at least one of an information characteristic and a structure characteristic included in the rotated table. 11. The method of claim 10,
The extraction of the information characteristic is
An image correction method comprising extracting text included in the rotated table and location information of the text. 12. The method of claim 11,
Extraction of the structural characteristics,
At least one cell included in the rotated table is recognized, and a correlation between the recognized cells is extracted. 11. The method of claim 10,
Based on the completion of the extraction of the information characteristic and the structural characteristic from the rotated table, performing correction of rotating the image by the specific angle in a direction opposite to the one direction how to correct an image. storage;
a receiver for receiving an image including a table; and
A control unit for extracting a plurality of lines constituting the table from the image,
The control unit is
The image correction system according to claim 1, wherein the rotation degree of the plurality of lines is calculated based on a reference line, and the image is corrected so that rotation of the table is performed based on the rotation degree. A program that is executed by one or more processes in an electronic device and can be stored in a computer-readable recording medium,
The program is
receiving an image comprising a table;
extracting a plurality of lines constituting the table from the image;
calculating a degree of rotation of the plurality of lines based on a reference line; and
A program storable in a computer-readable recording medium comprising instructions for performing a step of correcting the image so that rotation of the table is performed based on the degree of rotation.

Images