KR20220059892A - Online print production system using keyword analysis of text data input from users - Google Patents

Abstract

An online printed matter production system using keyword analysis of text input from a user comprises: a server generating a sample image; and a user terminal connected to the server via a communication network, transmitting text data containing print information and text information to the server and receiving the sample image from the server, wherein the print information includes size information of print paper. The server is configured to: extract main keywords in the text data through natural language processing; generate a front with a character size interpolated from a preset master font by using the text data; set a character color with an average value of color codes of a predetermined number of searched images resulted from search by the main keywords; and determine a background color by using an average value of colors of all pixels in a character block which is a rectangular block with a minimum size that can include a character with a character color, wherein the main keywords are extracted on the basis of a preset mathematical equation through related word analysis with respect to the text data. The present invention can easily produce printed matter.

Description

ONLINE PRINT PRODUCTION SYSTEM USING KEYWORD ANALYSIS OF TEXT DATA INPUT FROM USERS

The present invention relates to an online-based print production system, and more particularly, to an online print production system using keyword analysis of text input from a user.

Recently, various services such as e-commerce and electronic advertisement have been provided based on the Internet. In the past, all commerce transactions were physically performed offline, but an infrastructure was built to easily receive services through the Internet.

In the case of an online printing service that wants to produce printed materials such as calendars, catalogs, posters, pamphlets (leaflets), flyers, manuals (textbooks), newsletters (monthly, quarterly, etc.), business cards, envelopes, special prints, etc. online, the customer When the contents to be entered are provided to the service provider, the service provider prepares a sample proposal based on the contents to be entered in the printed matter, and the customer finally confirms the prepared sample proposal is common.

Recently, in order to sufficiently guarantee the customer's choice, online printing services are emerging that allow the customer to select the format or design of the print as well as the content in the print through the user interface.

However, if the customer chooses both the format and design of the printed matter, it will put a significant burden on the customer, which will impair customer satisfaction and make it difficult to provide quality service. When a design is decided internally and provided to customers, there is a problem that competitiveness is lowered due to an increase in cost due to an increase in the input labor force.

Therefore, a method to increase competitiveness by reducing the burden of design selection by customers and minimizing costs by providing customers with sample images that can satisfy customers with little labor while giving enough authority to choose the format or design of printed matter This is a necessary situation.

An object of the present invention is to provide an online print production system using keyword analysis of text input by a user.

The problems of the present disclosure are not limited to the problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below.

An online print production system using keyword analysis of text input by a user according to an embodiment of the present invention for solving the above problems is connected to a server for generating a sample image, and the server and a communication network, and to the server and a user terminal that transmits print information and text data containing information on texts, and receives a sample image from the server, wherein the print information includes size information of a print paper, and the server, Extracting main keywords in text data, using the text data to generate fonts having interpolated letter sizes from the master font, and using the average value of color codes of a predetermined number of images searched for by the main keywords to color text , and the background color is determined using the average value of the colors of all pixels in the letter block, which is the smallest rectangular block that can contain the letters having the letter color.

The interpolated character size generates a glyph with an enlarged width and height in the character block of the master font, and a weight value designated as the q square of the number of characters compared to the paper area in the glyph (where q is the weight value exceeds 0) scaled to have a real value less than 2).

The server may generate a virtual block having a predetermined interval from the boundary between the texts, and set the size of the blank space of the printing paper based on the virtual block.

The predetermined interval may be determined by multiplying the enlarged glyph by an absolute value of the weight value.

The color of the text may be determined using a color indicated by a value obtained by averaging a factor value representing the degree of red, a factor representing the degree of green, and a factor value representing the degree of blue of the predetermined number of images. there is.

The main keyword may search for related words with respect to a plurality of keywords, and it may be determined that the sum of the respective keywords and the number of related words corresponding to the respective keywords is the largest among the plurality of keywords.

According to an embodiment of the present invention, since a user can receive a sample image even with basic print information and print content, it is possible to easily produce a print, and as a service provider, a high-quality sample image can be created with minimal labor burden. This can increase competitiveness and reduce costs.

Effects according to the embodiments are not limited by the contents exemplified above, and more various effects are included in the present specification.

1 is a schematic diagram of the overall configuration of a print production system according to an embodiment of the present invention.
2 is a flowchart exemplarily illustrating a method of generating a sample image by a print production system according to an embodiment of the present invention.
3 and 4 are diagrams illustrating print information and text received from a user in a method for generating a sample image according to an embodiment of the present invention.
5 is a view for explaining a method of identifying text content through natural language processing in a method of generating a sample image according to an embodiment of the present invention.
6 and 7 are diagrams for explaining a method of adjusting a font in a method of generating a sample image according to an embodiment of the present invention.
8 is a first diagram for explaining a method of adjusting a blank space in a method of generating a sample image according to an embodiment of the present invention.
FIG. 9 is an enlarged view of area A of FIG. 8 .
10 is a second diagram for explaining a method of adjusting a blank space in a method of generating a sample image according to an embodiment of the present invention.
11 and 12 are diagrams for explaining a step of determining a color of an area in which text is not arranged in a sample image according to an embodiment of the present invention.
13 is a diagram for explaining another embodiment of determining a background of an area where text is not arranged in a method of generating a sample image according to an embodiment of the present invention.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but will be implemented in a variety of different forms, and only these embodiments allow the disclosure of the present invention to be complete, and common knowledge in the technical field to which the present invention belongs It is provided to fully inform the possessor of the scope of the invention, and the present invention is only defined by the scope of the claims.

Although the first, second, etc. are used to describe various elements, these elements are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first component mentioned below may be the second component within the spirit of the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. The same or similar reference numerals are used for the same elements in the drawings.

1 is a schematic diagram of the overall configuration of a print production system according to an embodiment of the present invention.

The print production system can provide a print production service to the user anytime, anywhere through the user terminal. A user can easily and conveniently produce a print using the print production service provided. For example, the user can check the expected result for the printed matter requested through the provided print production service, such as an image, so that more efficient work is possible.

For example, printed materials are available in various forms such as booklets, business cards, gift certificates, stickers, banners, calendars, catalogs, posters, pamphlets (leaflets), flyers, manuals (textbooks), newsletters (monthly, quarterly, etc.), envelopes, special printed materials, etc. can be implemented.

Referring to FIG. 1 , the print production system receives information input by a user (or orderer) from one or more user terminals 10 and user terminals 10 and uses a communication network 40 to produce a sample image to be used in print production. It may include a server 20 that generates the , and a printing device 30 that outputs a sample image.

The server 20 may generate a sample image through a plurality of algorithms in response to the received information input by the user (eg, print information and text data), and may provide the sample image to the user terminal 10 . The server 20 is a computing device capable of communicating with other devices through a wired or wireless network, and uses any one of a central processing unit (CPU), a graphic processing unit (GPU), a digital signal processor (DSP), and the like. It may be a computing device capable of performing a calculation operation, but is not limited thereto.

Here, the plurality of algorithms may be algorithms for generating a sample image for the final printed matter based on the print information and text data provided from the user terminal 10 . Such an algorithm may reflect general user preference types. For example, the preference type may be a preference type for cleanliness, warmth, harmony, etc. for fonts, margins, and backgrounds, depending on the content of the print.

The preference types for a plurality of users may be stored or updated in the database DB of the server 20 periodically or aperiodically (eg, in real time). In FIG. 1 , the database DB is illustrated as being included in the server 20 , but it is not limited thereto, and according to an embodiment, the database DB exists separately outside the server 20 to be connected to the communication network 40 . may be For example, the database DB may be implemented using a disk drive, magnetic disk, optical disk or other suitable storage device.

Also, when the sample image is selected as the final image from the user terminal, the server 20 may transmit the final image data to the printing device 30 through the communication network 40 .

The printing device 30 may output the final image by using the final image data received from the server 20 . For example, the printing device 30 may be implemented in various forms, such as an inkjet printer, a laser printer, a multifunction printer, or a 3D printer.

The user may input print information and contents (eg, text) of the printed matter into the user terminal 10 . In addition, the user may be provided with a sample image to be used in print production through the user terminal 10 . Here, the user terminal 10 may be an appropriate terminal device having a user interface capable of receiving download request information for a sample image from a user. For example, the user terminal 10 may include, but is not limited to, general PCs such as desktops and laptops, and mobile terminals such as smart phones, tablet PCs, personal digital assistants (PDAs), and mobile communication terminals, It can be broadly interpreted as an electronic device capable of communicating with the server 20 . Also, the input device of the user terminal 10 may include one or more of a touch display, a keyboard, a mouse, a touch pen or stylus, a microphone, a motion recognition sensor, and the like, but is not limited thereto.

The communication network 40 described above may be configured such that the server 20 can communicate with the user terminal 10 . Communication network 40, depending on the installation environment, for example, Ethernet (Ethernet), wired home network (Power Line Communication), a telephone line communication device and a wired network such as RS-serial communication or WLAN (Wireless LAN), Bluetooth and A wireless network such as ZigBee may be variously selected and configured.

2 is a flowchart exemplarily illustrating a method of generating a sample image by a print production system according to an embodiment of the present invention.

Referring to FIG. 2 , the method for the print production system to generate a sample image includes the steps of collecting print information and text from the user terminal 10 ( S110 ), and identifying main contents of the text through natural language processing ( S120 ). , a font adjustment step (S130), a blank space adjustment step (S140), and a background adjustment step (S150) may be included. In the present specification, it is described that each step is performed in turn according to the flowchart, but unless the spirit of the invention is changed, some steps shown to be performed in succession are performed simultaneously, the order of each step is changed, or some steps are performed It is obvious that , may be omitted, or other steps may be further included between each step.

Each step will be described in detail with reference to the drawings below.

3 and 4 are diagrams illustrating print information and text received from a user in a method for generating a sample image according to an embodiment of the present invention.

3 illustrates an example of a print information input screen according to an embodiment, and FIG. 4 illustrates an example of collected text. 4 to 5, the text is shown as an example of some contents of the traditional fairy tale "Rabbit Jeon".

2 to 4 , in the step of collecting print information and text ( S110 ), the server 20 collects data about print information 101 and data about text 102 from the user terminal 10 . corresponds to the step The user terminal 10 may collect print information 101 and text 102 from the user. For example, the user can print information 101 including drawings, image resolution information, color information, font information, line information, printing paper size information, crop line information, etc. inserted into the printed matter and text ( 102) may be input to the user terminal 10 . The user terminal 10 may transmit the print information 101 and data related to the text 102 to the server 20 . Each content of the print information 101 and the text 102 shown in FIGS. 3 and 4 is merely an example, and the present embodiment is not limited thereto.

5 is a view for explaining a method of identifying text content through natural language processing in a method of generating a sample image according to an embodiment of the present invention.

2 and 5 , the server 20 may perform natural language processing on the collected text 102_1 to identify main contents of the text 102_1 ( S120 ). That is, the server 20 extracts keywords (refer to the highlighted portion of 102_1) from the received text 102_1, and determines what content is being dealt with by performing natural language processing (NLP) on the extracted keywords. can do. For example, natural language processing is a framework or algorithm that can implement natural language processing, Python's NTLK (Python NLTK), Sanford's CoreNLP (Sanford CoreNLP), Apache's OpenNLP (Apache OpenNLP), Microsoft It can be implemented using Turing, BERT, GPT-2, or the like.

In an exemplary embodiment, the server 20 may determine the main content of the text 102 by performing context analysis and/or morphological analysis on the received text 102_1 . For this determination method, deep learning technology may be used. The server 20 may use a pre-built data dictionary as a means for determining a target subject. The server 20 may determine a related word from the text 102_1 through the data dictionary and extract a main keyword.

The server 20 may extract the main keyword (K _main ) through the following [Equation 1] through the analysis of the related word from the text 102_1.

Here, i is a natural number greater than 1 and means the number of keywords included in the text 102_1. K _i is the total number of related words of the i-th keyword and the i-th keyword in the text.

For example, as shown in the figure, the server 20 includes the terms 'Namhae', 'Yonggung', 'Seafood', and 'Yongwang' (first related words) through the data dictionary among the contents of the text 102_1. It can be determined that it is a related word for 'sea' (the first keyword). The server 20 may determine that a total of eight related words for 'sea' and 'sea' are disposed in the text 102_1. In addition, the server 20, through the data dictionary among the contents of the text 102_1, the terms 'alcohol', 'drinking', 'medicine', 'eat' (second related words) are 'food' (the first 2 keyword) can be judged as a related word. The server 20 may determine that a total of five related words for 'food' and 'food' are arranged in the text 102_1.

The server 20 may determine that 'sea' is the main keyword (K _main ) in the text 102_1 through [Equation 1], and may determine that the content related to 'sea' is the main content.

6 and 7 are diagrams for explaining a method of adjusting a font in a method of generating a sample image according to an embodiment of the present invention.

6 shows a part of the step of adjusting the font size and font, and FIG. 7 shows a part of the step of determining the character color.

Referring to FIGS. 2, 6 and 7 , the font adjustment step S130 corresponds to the server 20 adjusting the font of the text 102 . For example, the font may include a font size, a font color, and the like.

The server 20 uses the text data to change the thickness or width, design features, etc., and interpolates the text 102_2 from the master font to include a font having a new font size (eg, different from the font size of the master font). can create The server 20 may designate the input text data as a master font (designate as a small value), and automatically generate three glyphs using the small value (or master font). A glyph may mean a basic unit for the shape of a single character. The three glyphs include Goal, Wide, and Tall, where Wide is a horizontally enlarged form of Small, Tall is a vertical enlarged form of Small, Goal is a horizontally and vertically enlarged form of Small (refer to (A of FIG. 6)) or It is a reduced type (refer to FIG. 6(B)) glyph. The server 20 may generate glyphs such as Tall, Wide, and Goal through Small values using [Equation 2] below.

이고, q는 0 초과 2 미만의 실수 값으로 미리 설정될 수 있다. 수학식 2에서 BoundingBox는 Small 값을 포함할 수 있는 최소 크기의 사각 블록인 글자 블록에 해당하며, BoundingBoxWidth와 BoundingBoxHeight는 글자블록의 너비와 높이이고, Small_left_Top_Point는 글자블록의 좌측최상단과 대응하는 좌표일 수 있다. x, y는 각각 글자의 너비 및 높이에 대한 변수로서, 예를 들어, Small_x는 글자블록에 포함된 글자의 너비이고, Small_y는 글자블록에 포함된 글자의 높이이며, Wide_x는 Wide에 따른 글자의 너비이고, Tall_y는 Tall에 따른 글자의 높이이다.Here, α is a weight value determined according to the number of texts 102_1 and the paper size to be output, for example,

, and q may be preset to a real value greater than 0 and less than 2. In Equation 2, BoundingBox corresponds to a letter block, which is a rectangular block with the smallest size that can contain a Small value, BoundingBoxWidth and BoundingBoxHeight are the width and height of the letter block, and Small_left_Top_Point is a coordinate corresponding to the upper left corner of the letter block. there is. x and y are variables for the width and height of the character, respectively. For example, Small _x is the width of the character included in the character block, Small _y is the height of the character included in the character block, and Wide _x is the width of the character block. is the width of the character according to Tall, and Tall _y is the height of the character according to Tall.

The server 20 may adjust the character color using text data. For example, the server 20 acquires a search image by performing an image search in various public portal servers (google, naver, etc.) using a main keyword, and a color code of the search image through computer vision technology can be analyzed to determine the text color. For example, the server 20 may determine the text color through HTML code analysis when the search image is viewed on the browser. For example, the server 20 searches for a predetermined number of images (eg, n, n is a natural number greater than or equal to 1) related to the main keyword 'sea', and determines the color through [Equation 3] below. can Here, the HTML code is a code (#abcdef) consisting of # and 6 digits (eg, #FFFF00 is the HTML code for yellow color).

In Equation 3, #abcdef is an HTML code corresponding to a specific color, and among #abcdef, ab, cd, and ef may be hexadecimal numbers between 00 and FF, respectively. For example, ab, cd, and ef may each represent an integer between 0 and 255 in the decimal system. In the HTML code expressed by #abcdef, ab is the factor value representing the degree of red, cd is the factor value representing the degree of green, and ef is the factor value representing the degree of blue. That is, the server 20 separately calculates a factor value representing the degree of red, a factor value representing the degree of green, and a factor value representing the degree of blue in a predetermined number of search images, and a factor representing the degree of red The average of the values, the average of the print values indicating the green level, and the average of the print values indicating the blue color may be calculated.

The server 20 converts the color indicated by the calculated average of the factor values representing the degree of red, the mean of factor values representing the degree of green, and the mean of factor values representing the degree of blue to color inversion (for example, By inverting complementary colors (like inverting black to white), you can determine the color of the font.

8 is a first diagram for explaining a method of adjusting a blank space in a method of generating a sample image according to an embodiment of the present invention. FIG. 9 is an enlarged view of area A of FIG. 8 . 10 is a second diagram for explaining a method of adjusting a blank space in a method of generating a sample image according to an embodiment of the present invention.

8 and 10 respectively show a part of the blank adjustment step ( S140 ).

2 and 8 to 10 , the step of adjusting the blank space ( S140 ) corresponds to the step of the server 20 adjusting the size of the blank of the paper 200 . In one embodiment, adjusting the blank space ( S140 ) may include blocking the text 102_2 and setting a blank space based on the blocked block 110 .

The step of turning the text 102_2 into a block 110 corresponds to a step of generating a virtual block 110 having a predetermined interval from the boundary of the text 102_2. If the margin is set to be minute without the step of blocking (that is, when there is no predetermined interval), printing may not be performed on the edge of the paper 200 depending on the printing device 30 .

The block 110 may be set through [Equation 4] below.

Here, P ₁ (X) is the horizontal spacing from the boundary of the texts 102_2, Q ₁ (Y) corresponds to the vertical spacing from the boundary of the texts 102_2, Wide _x and Tall _y are through the master font. Among the three generated glyphs, it may be the width of the character according to Wide and the height of the character according to Tall.

The generated block 110 may be arranged in the paper 200 through [Equation 5] below.

Here, P ₂ (X) is the margin from the left side of the paper 200 to the boundary of the block 110 , and P ₃ (X) is the margin from the right side of the paper 200 to the boundary of the block 110 . , Q ₂ (Y) is the margin from the upper side of the paper 200 to the boundary of the block 110 , and Q ₃ (Y) is the margin from the lower side of the paper 200 to the boundary of the block 110 . am. A and B are the horizontal and vertical sizes of the paper 200, respectively. β and γ are parameter values assigned according to the text size of the paper 200, respectively. m is any number, for example any rational number from 1.3 to 1.4 may be selected.

11 and 12 are diagrams for explaining a step of determining a color of an area in which text is not arranged in a sample image according to an embodiment of the present invention.

2, 11, and 12, the background adjustment step (S150) is the server 20, the text 102_2 within the paper 200 in the non-arranged area (the inside of the block 110 and the background 120) )) corresponds to the step of determining the color of

As an embodiment, the server 20 may adjust the color of the area where the text 102_2 is not arranged based on the adjusted character color. The color of the area where the text 102_2 is not placed is to be determined based on the letter block 130 and the letter color, which is a rectangular block of the minimum size that can include letters (interpolated new letters formed in the font adjustment step S130). can For example, the color of the region where the text 102_2 is not arranged may be determined by using an average value of the colors of all pixels in the text block 130 . Here, a pixel in which characters are not arranged is defined as having a white code (#000000). That is, the server 20 may determine the color of the area where the text 102_2 is not arranged through [Equation 6] below.

Here, L and M represent the width and height of the letter block 130, respectively, and when the letter block 130 includes i × j (i, j is a natural number equal to or greater than 1) pixels horizontally and vertically, ab ( i,j) is a factor value indicating the degree of red of the i-th horizontally and vertically j-th pixel, cd(i,j) is a factor value indicating the degree of green of the i-th horizontally and vertically j-th pixel, ef(i,j) is a factor value indicating the degree of blue of the i-th horizontally and vertically j-th pixel, gh is a factor value indicating the degree of redness in the area where the text 102_2 is not arranged, and ij is the text (102_2) may be a factor value representing the degree of green in the non-arranged area, and kl may be a factor value representing the degree of blue in the area where the text 102_2 is not disposed.

13 is a diagram for explaining another embodiment of determining a background of an area where text is not arranged in a method of generating a sample image according to an embodiment of the present invention.

The server 20 according to an embodiment of the present invention may fill an area (background 120) where the text 102_2 is not arranged in the paper 200 using a search image.

For example, the server 20 extracts keywords in which the total number of keywords and related words corresponding to the keywords is greater than or equal to a preset number from among the keywords included in the text 102_2, and from among the extracted keywords, an object ) and corresponding keywords can be selected.

The server 20 detects an object corresponding to the selected keywords from a plurality of search images searched using the main keyword selected according to Equation 1 above, and uses the search image with the largest number of detected objects. The background 120 may be filled. In this case, as an algorithm for detecting an object in the search images, an object detection algorithm using a convolutional neural network (CNN), which is advantageous for image recognition among artificial neural networks, may be used. Examples include SSDs (W. Liu, D. Anguelov, D. Erhan, C. Szegedy, and S. E. Reed. SSD: single shot multibox detector. CoRR, abs/1512.02325, 2015.).

For example, as shown in FIG. 13 , from among the keywords included in the text 102_2 , keywords in which the total number of keywords and related words corresponding to the keyword is greater than or equal to a preset number are extracted, and from among the extracted keywords, a thing or an object Selecting corresponding keywords (for example, dragon king, new god, rabbit, etc.), and using the search image with the most selected keywords and corresponding objects (yong king, new god, rabbit), the background 120 is shown in FIG. 13 can be filled together.

On the other hand, by combining the embodiment according to FIG. 13 and the embodiment according to FIG. 11 , the server 20 sets the background color inside the block 110 in which the text 102_2 is blocked using the embodiment according to FIG. 11 . It may be determined, and the external background 120 of the text 102_2 may be filled using the embodiment according to FIG. 13 .

When the background adjustment is completed, the server 20 may generate a sample image in which the font, blank space, background, etc. are adjusted, and transmit the generated sample image to the user terminal 10 .

As such, the server 20 according to an embodiment of the present invention dynamically determines a font size and color, a background color, and a background image according to the print information and text information provided through the user terminal 10 to generate a sample image. And, since the generated sample image is provided to the user terminal 10, the user has less hassle of setting all the details from the text to the background, and receives the sample image most optimized for the text content and performs some modifications to remotely control the final print. can be printed with

In the above, embodiments according to the technical idea of the present invention have been described with reference to the accompanying drawings, but those of ordinary skill in the art to which the present invention pertains will have other specific forms without changing the technical spirit or essential features of the present invention. It will be understood that it can be implemented as It should be understood that the embodiments described above are illustrative in all respects and not restrictive.

10: user terminal 20: server
30: printing device 40: communication network
101: print information 102: text
110: block 120: background
130: block of letters 200: paper

Claims (1)

An online print production system using keyword analysis of text input by a user,
a server that generates sample images; and
a user terminal connected to the server through a communication network, transmitting text data containing print information and text information to the server, and receiving a sample image from the server;
The print information includes size information of the print paper,
The server extracts the main keywords in the text data through natural language processing, generates a font having a font size interpolated from a preset master font using the text data, and searches a predetermined number of searched for the main keywords. The text color is set using the average value of the color codes of the images, and the background color is determined using the average value of the colors of all pixels in the letter block, which is the smallest square block that can contain the text with the text color. do,
The interpolated font size is
A glyph with an enlarged width and height is generated from the letter block of the master font, and a weight value assigned to the glyph as the q square of the number of characters compared to the paper area (where q is a real value in which the weight value is greater than 0 and less than 2) It is determined by multiplying by a preset constant to have
The main keyword (K _main ) is extracted based on the following equation through a related word analysis for the text data,

In the above equation, i is a natural number greater than 1 and is the number of keywords included in the text data, and K _i is the total number of related words between the i-th keyword and the i-th keyword in the text data.

Images