KR20180011524A - Platform for establishing a database for translation through cartoon based on a crowd sourcing - Google Patents

Abstract

The present invention provides an image display apparatus comprising: an image display unit displaying a cartoon image; A character input window capable of inputting a translation character corresponding to a character group in a cartoon image displayed on the image display unit; A translation display unit for displaying translations already translated by a plurality of translators corresponding to a character group in a cartoon image displayed on the image display unit; A recommendation input unit capable of giving a recommendation value or a non-recommendation value to each of the translations in the pre-translation display unit; And a status value input unit for inputting a status value of a cartoon image displayed on the image display unit.

Description

{PLATFORM FOR ESTABLISHING DATABASE FOR TRANSLATION THROUGH CARTOON BASED ON A CROWD SOURCING}

The present invention relates to a method for securing a translation database through cartoons based on crowd sourcing.

Various methods are being developed for the translation of foreign languages under the global age. However, even in the development of foreign language translation technology, there is a limit to the translation suitable for the situation, and there is a limit to the living translation in the actual used language. For example, if a sentence in Korean is to be translated into a sentence under anger, and if it is a sentence under a situation to joke, the translation must be different. The present translation technique is a mechanical translation There was a dissatisfaction with the translation, but there is no current technology to resolve such complaints. In addition, with the development of the era, there are newly emerging languages other than the pre-existing languages. Currently, the technology is limited to the translation into the new language according to these times.

However, in order to solve such a problem, it is conceivable to prepare a database in which context expressions of sentences to be translated are arranged and translate them using this database. However, there is no database in which such situation expressions are arranged, There are temporal method limitations.

The present invention recognizes the limitations of the present translation technology and develops a new method for easily and accurately securing a database in which contextual expressions are arranged.

In this way, we have developed accurate automatic machine translation methods for cartoons such as Webtoon.

The present invention provides an image display apparatus comprising: an image display unit displaying a cartoon image; A character input window capable of inputting a translation character corresponding to a character group in a cartoon image displayed on the image display unit; A translation display unit for displaying translations already translated by a plurality of translators corresponding to a character group in a cartoon image displayed on the image display unit; A recommendation input unit capable of giving a recommendation value or a non-recommendation value to each of the translations in the pre-translation display unit; And a situation value input unit for inputting a situation value of a cartoon image displayed on the image display unit.

The crowd sourcing of the present invention is a combination of the English word "crowd", which means the public, and "outsourcing," which means utilization of external resources, so that an unspecified number of people can participate to complete the content, It is to solve. Mainly through online space. Jeff Howe, editor of magazine Wired, used the term for the first time in 2006.

The translation database of the present invention refers to a database in which translations of a second language for a first language are arranged for use in machine translation. A translation of a second language suitable for the first language to be translated for machine translation can be obtained from the data of the database. The database of the present invention refers to a database composed of second translations for a first language, recommendation values of the respective translations, and context values of the first language. In machine translation, context values can be used as filter entries, and recommendation values can be used to set the selection order.

The platform of the present invention may be a program or application running on an Internet computer or mobile terminal.

The image display unit refers to a portion where an image in a display means (for example, a computer monitor or a screen of a mobile device) is displayed.

The character input window means a portion where a character (language or symbol) input by an input means (for example, a computer keyboard or a screen touch of the mobile device) is displayed on a part of the screen of the display means.

The translation display unit refers to a portion where translations previously translated by a plurality of translators are displayed in the display means.

The recommendation input unit is a configuration for receiving a recommendation value to the translations. The recommendation value may be a number assigned to the translation. You can provide a recommendation value of 1 at a time. The recommendation values of several persons are gathered and stored as accumulated recommendation values of the corresponding translations.

The situation value input unit is configured to receive the status value of the corresponding translation. May be provided to illustrate a particular situation topic (e.g., a situation topic of speaker emotion), to provide and select a view list on the illustrated contextual topic, or may be configured to directly fill in any situation value. The subject of the situation may include a speaker's emotion, a speaker's gender, a speaker's age, a genre of a cartoon, a celadon emotion, a celadon gender, a celadon age, a speaker and a celadon relationship. Such a situation topic can be a contextual subject of the present invention if it is a context topic that can affect translation. This situation theme can also be configured to directly set.

In another aspect, the present invention provides a method of extracting a character value of a character layer from a layered image file format including an image layer and a character layer; And selecting a translation corresponding to the extracted character value from the translation data secured by the platform.

The selection of the translation corresponding to the character value extracted from the translation data is a selection of the translation of the most recommended value among the translation of the character value in the database.

And the database is filtered with the original situation value when a translation corresponding to the character value extracted from the translation data is selected.

The translation database securing platform of the present invention allows a plurality of users to accurately understand the intrinsic situation of characters in a comic by using a comic, and thereby, the situation values can be collected accurately, and the translations can also be translated appropriately Effect. In addition, the collective intelligence of various users can be gathered by the cloud Internet environment, accurate situation values can be gathered, and it is possible to recommend the translations suitable for the situation value, thereby improving the accuracy of the translation database.

It is also possible to improve the accuracy of comic translation by using new comic translation.

1 is an illustration of a screen displayed by a platform of the present invention through a display means of a computer or a mobile terminal,

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings. Wherein like reference numerals refer to like elements throughout.

The platform of the present invention may be a web browser running on a computer or a mobile terminal (e.g., smart phone or tablet, etc.) and may be an application or a program running on a computer or mobile terminal or the like.

FIG. 1 is an illustration of a screen displayed by a platform of the present invention through a display means of a computer or a mobile terminal, including an image display unit 110 displaying a cartoon image; A character input window 120 for inputting a translation character corresponding to a character group in a cartoon image displayed on the image display unit; A translation display unit (130) for displaying already translated translations by a plurality of translators corresponding to a character group in a cartoon image displayed on the image display unit; A recommendation input unit (140) which can give a recommendation value or a non-recommendation value to each of the translations in the pre-translation display unit; And a status value input unit 150 for inputting a status value of a cartoon image displayed on the image display unit.

The image display unit displays an image of a cartoon. In the case of a one-digit cartoon, one cartoon image can be displayed. In the case of a cartoon including multiple cuts, a subsequent cut image can be continuously displayed in accordance with a screen scroll in one direction, and a cartoon image can be displayed in a manner of changing one cut. Cartoon images can contain characters. The character may be a character within a speech balloon, or may be a character describing a scene or the like within the image. Characters can be grouped by any criterion. For example, they can be grouped into speech balloons, and they can be divided into sentence units. The grouped characters can be selected by a platform input means (for example, a mouse click or a screen touch), and the selected group can be displayed by a box or the like as indicated by a reference numeral 160, The translations are displayed on the translation display section 130. Also, it is possible to input the translation corresponding to the selected group into the character input window 120, and similarly, a situation value corresponding to the selected group can be input through the situation value input unit.

The translation display unit 130 is a part for displaying the translation previously translated by other translators. For example, in an Internet cloud environment, when a user selects a character group of a comic image by the platform of the present invention and completes the input of the translation through the translation character input window, the corresponding translation is stored in the server of the cloud environment, Can be displayed on the translation display section.

The recommendation input unit 140 is a means for inputting a recommendation value or a non-recommendation value for each of the translations of the translation display unit. In FIG. 1, a recommendation input unit capable of inputting a recommendation value is illustrated. The platform user gives an input signal to the means by an input means (for example, a mouse click or a screen touch) of the platform, and a recommendation value is accumulated in the translation server to the platform server. Although FIG. 1 exemplifies inputting only a recommended value, the platform may have a deprecated value input means for inputting a deprecated value.

The recommendation input section may display a number of currently accumulated recommendation values or deprecated values.

The character input window 120 is a means for inputting a translation corresponding to the selected group. The user can input the translation of the selected group considered by the current platform user through the input means (for example, the keyboard). The entered translation is stored in the cloud server.

The situation value input unit 150 is a means for inputting a situation value of a cartoon image displayed on the image display unit. The context value can be, for example, without limitation, speaker's emotion, speaker's gender, age of speaker, genre of comic, celadon emotion, audience list, audience history, speaker and listener relationship. As illustrated in Fig. 1, the situation value can be prepared in advance and an arbitrary situation value can be given by the current platform user. Cartoons are well suited for platform users to determine their own implicit situations. The platform user, that is, the cartoon reader can easily judge the information of the speaker's feelings in the cartoon, the age, the state in the cartoon, and the gender, and the platform of the present invention provides a means by which information judged by the user can be inputted. The translations of various users (translators) and these situation values are gathered. In case of situation value, the highest number of situation values among several gathered situation values will be the exact situation value which was originally given by the power of collective intelligence, A translation with a recommendation value will also be the basis for judging the most appropriate translation.

The platform of the present invention operates on an Internet cloud basis to translate characters in comic images from multiple users into translations; A recommendation value input signal of a previously translated translation; And status values. These real-time input values are stored in a specific server under the Internet cloud. In other words, it will accumulate and accumulate as many cumulative translations of various users, recommendation accumulation value and situation value of the translations for the character value of the comic (in FIG. 1, "I do not know whether or not you know, , And the database thus stored can be used as a database for machine translation later.

For example, if a translation of the phrase "You do not have money in the club" is input for the machine translation, it is possible to provide a machine translation with the most recommended translations through the database obtained by the present invention. Also, if the sentence "There is no money in the club" is a baseball related sentence, then in the stored database, the value of the situation may provide the highest number of translations among the translations corresponding to baseball. Furthermore, if the expression "no money in the club" is entered at the time of machine translation into a sentence by an angry speaker, it is possible to provide the highest number of translations among the translations corresponding to the angry situation value in the stored database.

In addition, the database of the present invention can be utilized for a method for machine translation of cartoons such as Webtoons. The webtoon image can be made in a layered image file format including an image layer and a character layer at the time of manufacturing the webtoon, and the situation value can be applied to each webtoon image in advance. It is possible to give a situation value to the translation target webtoon in various ways such as giving the status value in advance by the author and giving it by the translator of the webtoon. It is possible to extract the character value of the character layer from the layered image file format and to select the translation from among the translations corresponding to the extracted character value in the secured translation data of the present invention, . At this time, the database may be filtered with the predetermined situation value (original situation value), so that the translation can be selected in the desired situation value.

While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined by the appended claims. Accordingly, the true scope of the present invention should be determined by the following claims.

Claims (5)

An image display unit for displaying a cartoon image;
A character input window capable of inputting a translation character corresponding to a character group in a cartoon image displayed on the image display unit;
A translation display unit for displaying translations already translated by a plurality of translators corresponding to a character group in a cartoon image displayed on the image display unit;
A recommendation input unit capable of giving a recommendation value or a non-recommendation value to each of the translations in the pre-translation display unit; And
And a situation value input unit for inputting a situation value of a cartoon image displayed on the image display unit.
Platform for obtaining translation database through crowdsourcing based comics.
The method according to claim 1,
Wherein the situation value includes a speaker emotion, a speaker sex, a speaker age, a genre of a comic, a celadon emotion, a celadon gender, a celadon age, a speaker and a celadon relationship.
Platform for obtaining translation database through crowdsourcing based comics.
Extracting a character value of the character layer from a layered image file format including an image layer and a character layer;
Selecting a translation corresponding to the extracted character value from the translation data obtained according to any one of claims 1 to 3,
How to translate cartoons through a platform to secure a translation database through crowdsourcing based comics.
The method of claim 3,
Wherein selecting a translation corresponding to a character value extracted from the translation data is a selection of translation of the most recommended value among the translations of the character value in the database,
How to translate cartoons through a platform to secure a translation database through crowdsourcing based comics.
The method of claim 3,
Determining an original situation value for the comic image,
Wherein when the translation corresponding to the character value extracted from the translation data is selected, the database is filtered with the original context value.
How to translate cartoons through a platform to secure a translation database through crowdsourcing based comics.

Images