KR20200077037A - Method and apparatus of extracting multiple cuts from a comic book - Google Patents

Abstract

The present invention relates to a method and a device for making a comic book into multiple cuts. The method comprises the following steps of: recognizing the outline of an image on a page of an inputted comic book; generating boxes by detecting all the objects on the page; removing boxes smaller than the size of a predetermined box from the page; regenerating one box by combining overlapping boxes among boxes remaining on the page; and separating each of the regenerated boxes into one cut, wherein the objects are closed figures. Therefore, a new e-book or webtoon can be produced by easily converting each cut of an existing paper comic book into an image file.

Description

Method and apparatus of extracting multiple cuts from a comic book}

The present invention relates to a method of making a comic book into a plurality of cuts, and more specifically, to convert each cut of an existing paper comic book into an image file, thereby making a comic book into a plurality of cuts that can produce a new e-book or webtoon. And devices.

In the production of cartoon contents for mobile (smartphone), in accordance with the trend of the expression media of cartoons from paper to computer through smartphones, newly published cartoons can be produced for mobile from the beginning, In the case of published cartoon contents for computers, it is natural that it is economical to use the existing cartoon contents for computers for mobile rather than creating new cartoon contents for mobile.

However, in using the computer cartoon content for mobile, the computer cartoon image suitable for output on a large screen of several tens of centimeters due to the difference in size and resolution of the display (eg, LCD) of a mobile device represented by a computer and a smartphone. If the file is output as it is on a mobile screen of several centimeters, it may not be sufficient to recognize the content expressed in the image, so the cartoon content for the computer needs to be converted (edited) for the mobile.

Currently, the task of converting an image file of a cartoon content for a computer into an image file for a mobile is done by, for example, using a general-purpose image program such as Photoshop and manually modifying the image file to fit the mobile environment. In the case of fingerprints, the operator erases the traces of the existing fingerprints and writes new fingerprints, which results in a very slow working speed, not only creates a typo in the newly created fingerprints, but also improves the quality uniformity due to differences in the functions of the workers. It is difficult, and as a result, the production cost and supply price are rising, and the reality is that production of mobile cartoon contents does not exceed the breakeven point, except for some popular works, which is a barrier to entry of cartoon's mobile supply. In addition, in creating an image file for mobile from the beginning from a newly created cartoon original, the image file obtained by scanning the cartoon original is substantially the same as the screen size of the mobile, in the same way as converting a computer image file to a mobile image file. As it is inevitable that the operation to be converted to fit must be carried out, it cannot be freed from the problems of increasing the operation speed and cost as described above.

Prior art 1 (KR10-2016-0138930) discloses a method of dividing and converting an original image into a background area and a cartoon area.

Therefore, the first problem to be solved by the present invention is to provide a method of making a comic book into a plurality of cuts so that each cut of an existing paper comic book can be easily converted into an image file to produce a new e-book or webtoon.

The second problem to be solved by the present invention is to provide an apparatus for making a comic book into a plurality of cuts so that it can be utilized when separating each cut from a comic book by learning various forms of each cut of the existing paper comic book.

It is also to provide a recording medium readable by a computer recording a program for executing the above-described method on a computer.

The present invention, in order to achieve the first task, recognizing the outline of the image on the page of the input comic book; Generating a box by detecting all objects on the page; Removing a box smaller than a preset box size from the page; Combining the overlapping boxes among the boxes remaining on the page and regenerating into one box; And separating the regenerated boxes into one cut each, and the object provides a method of making a comic book into a plurality of cuts, wherein the object is a closed figure.

According to one embodiment of the present invention, the box is preferably the smallest polygon that the line segment of the object and the box contact.

In addition, displaying the regenerated box of the page together with the page; And moving each point of the regenerated box to match the boundary of each cut of the page. Learning a shape of a box applied to the comic book by moving each point of the regenerated box with a plurality of pages of the comic book to match the boundary of each cut of the page; And reproducing the shape of the learned box to the step of regenerating into the one box, and combining the overlapping boxes among the boxes remaining on the page to regenerate into one box. It is preferable to regenerate into one box using the shape of the box.

According to another embodiment of the present invention, receiving at least one or more of the number of cuts displayed on the screen, the sound effect corresponding to each cut, and the color applied to each cut, and corresponding to the separated cuts It may further include.

The present invention, in order to achieve the second problem, the outline recognition unit for recognizing the outline of the image on the page of the input comic book; A box generator that detects all objects on the page and creates a box; A box regenerating unit that removes a box smaller than the size of a preset box from the page, and combines the overlapping boxes among the boxes remaining on the page to regenerate them into one box; And a separating part for separating the regenerated boxes into a single cut, and the object is a closed figure.

According to one embodiment of the present invention, the box is preferably the smallest polygon that the line segment of the object and the box contact.

In addition, when the regenerated box of the page is displayed together with the page, each point of the regenerated box is moved to match the boundary of each cut of the page, and the regenerated box of the regenerated box for a plurality of pages of the comic book. By moving each point to match the boundary of each cut of the page, the machine learning unit for learning the shape of the box applied to the comic book and feeding the learned shape of the box to the box regenerating unit is further included, The box regenerating unit may regenerate into a single box by using the shape of the feedback box.

According to another embodiment of the present invention, at least one or more of the number of cuts displayed on the screen, the sound effect corresponding to each cut, and the color applied to each cut is input, and corresponds to each cut separated from the separation unit It may further include an editing unit.

In order to solve the above other technical problems, the present invention provides a computer-readable recording medium recording a program for executing a method for making a comic book into a plurality of cuts on a computer.

According to the present invention, it is possible to easily convert each cut of an existing paper comic book into an image file to produce a new e-book or webtoon.

In addition, according to the present invention, various forms of each cut of the existing paper comic book can be learned and utilized when separating each cut from the comic book.

1 is a block diagram of an apparatus for making a comic book into a plurality of cuts according to an exemplary embodiment of the present invention.
2 is a flowchart of a method for making a comic book into a plurality of cuts according to an exemplary embodiment of the present invention.
Figure 3 shows the result of detecting the contour using the Canny filter.
Figure 4 shows the result of generating a box by detecting all the objects in the comic book page in step 210.
FIG. 5 shows the result of removing a box of an object smaller than the size of a preset box in step 220.
FIG. 6 shows the result of regenerating into one box by combining overlapping boxes in step 230.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. Advantages and features of the present invention, and methods for achieving them will be clarified with reference to embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only the embodiments allow the disclosure of the present invention to be complete, and are conventional in the art to which the present invention pertains. It is provided to fully inform the knowledgeable person of the scope of the present invention, and the present invention is only defined by the scope of the claims.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used as meanings commonly understood by those skilled in the art to which the present invention pertains. In addition, terms defined in the commonly used dictionary are not ideally or excessively interpreted unless explicitly defined.

1 is a block diagram of an apparatus for making a comic book into a plurality of cuts according to an exemplary embodiment of the present invention.

Referring to FIG. 1, an apparatus for making a comic book according to the present embodiment into a plurality of cuts includes a contour recognition unit 110, a box generation unit 120, a box regeneration unit 130, a machine learning learning unit 140, and separation It is composed of a unit 150 and an editing unit 160.

The outline recognition unit 110 recognizes the outline of the image on the input comic book page.

It is possible to detect the contours of an image using edge detection methods using primary derivatives such as Sobel, Prewitt and Roberts, and edge detection methods using secondary derivatives such as Laplacian and Laplacial of Gaussian (LoG). In addition, the Canny edge detection method can be used separately.

The box generating unit 120 detects all objects on the page and creates a box.

Preferably, the object is a closed figure. A closed figure has the same starting point and ending point. On the other hand, since the starting point and ending point are different and are not surrounded by sides, the figures without distinction between inside and outside are open figures.

As an embodiment, the box may be the smallest rectangle in contact with the object and four line segments. As another embodiment, the box may be the smallest polygon that the object and the line segment of the box contact.

In the edge detection method using the second derivative, since the detection is performed twice during the edge detection, the detected contour forms a closed curve, so it is not necessary to distinguish the closed figure from the open figure in step 610.

The box regenerating unit 130 removes a box smaller than the size of a preset box from the page, and combines overlapping boxes among the boxes remaining on the page to regenerate a single box.

Since the boxes generated by the box generator 120 are the result of detecting all objects on the page, a lot of boxes are generated. To organize this, the box regeneration unit 130 sets a minimum size box in advance to remove a box of an object smaller than the minimum size box.

As an embodiment, when the length of the perimeter of the minimum size box is greater than the perimeter of the box generated in the box generator 120, the generated box may be removed.

When the regenerated box of the page is displayed together with the page, the machine learning unit 140 moves each point of the regenerated box to match the boundary of each cut of the page, and the plurality of comic books By moving each point of the regenerated box with respect to the page to match the boundary of each cut of the page, the shape of the box applied to the comic book is learned.

The machine learning learning unit 140 feeds the shape of the learned box to the box regeneration unit 130.

Machine learning is a method of giving enough data to a computer and finding general patterns there, as humans learn knowledge through various experiences and trial and error, and machine learning's representative algorithm is deep learning.

When the box regenerating unit 130 combines the overlapping boxes among the boxes remaining on the page and regenerates them into one box, the feedback box type is used as the regenerated box type.

The rectangular box shape can be easily separated with one cut, but the irregular shape box is not easy to separate.

To this end, when the regenerated box of the page is displayed together with the page, the user modifies the regenerated box according to the shape of the actual box, and repeats multiple times for a plurality of pages to change the shape of the box through machine learning. You will learn.

When the box shape of the page has an outline, a box is formed along the outline, but when there is no outline, the box may be difficult to form.

At this time, if the boxes generated on the page are displayed together with the page, the user may create a box including an image for images without an outline, thereby forming a box for an image without an outline.

The separation unit 150 separates the boxes regenerated by the box regeneration unit 130 into one cut.

When the separation unit 150 separates the boxes into a single cut, they are separated in a predetermined order. For example, you can set the order in the order of left->right, top->bottom.

The editing unit 160 receives at least one of the number of cuts displayed on the screen, the sound effect corresponding to each cut, and the color applied to each cut and corresponds to the separated cuts.

2 is a flowchart of a method for making a comic book into a plurality of cuts according to an exemplary embodiment of the present invention.

Referring to FIG. 2, a method of making a comic book according to the present embodiment into a plurality of cuts includes steps processed in time series in an apparatus for making the comic book shown in FIG. 1 into a plurality of cuts. Therefore, even if omitted below, the description of the above with respect to an apparatus for making a comic book shown in FIG. 1 into a plurality of cuts is also applied to a method for making a comic book according to the present embodiment into a plurality of cuts.

In step 200, the apparatus for making a comic book into a plurality of cuts recognizes the outline of an image on a page of the comic book.

It is possible to detect the contours of an image using edge detection using primary derivatives such as Sobel, Prewitt and Roberts, and edge detection using secondary derivatives such as Laplacian and Laplacial of Gaussian (LoG). In addition, the Canny edge detection method can be used separately.

Figure 3 shows the result of detecting the contour using the Canny filter.

Referring to FIG. 3, the outline is detected without distinguishing pictures, letters, and spaces on any one page of the comic book.

In step 210, the apparatus for making a comic book into a plurality of cuts detects all objects on the page and creates a box.

Preferably, the object is a closed figure. A closed figure has the same starting point and ending point. On the other hand, since the starting point and ending point are different and are not surrounded by sides, the figure without distinction between inside and outside is an open figure.

As an embodiment, the box may be the smallest rectangle in contact with the object and four line segments. As another embodiment, the box may be the smallest polygon that the object and the line segment of the box contact.

In the case of using the edge detection method using the second derivative when recognizing the contour of the image in step 200, step 210 may be omitted.

In the edge detection method using the second derivative, since the differential is performed twice in the process of detecting the edge, the detected contour forms a closed curve, so there is no need to distinguish the closed figure from the open figure in step 210.

Figure 4 shows the result of generating a box by detecting all the objects in the comic book page in step 210.

Referring to FIG. 4, a lot of boxes are generated because there are many eyes, noses, and faces of a person as a closed figure object in a comic book page.

On the other hand, since objects less than a certain size are difficult to form one cut, it is also possible to not create a box from the beginning in step 210.

In step 220, the apparatus for making a comic book into a plurality of cuts removes a box of an object smaller than a preset box size.

Since the boxes generated in step 210 are the result of detecting all objects on the page, a lot of boxes are generated. To organize this, a minimum size box is set in advance to remove a box of objects smaller than the minimum size box.

As an embodiment, when the length of the perimeter of the minimum size box is greater than the perimeter of the box generated in step 210, the generated box may be removed.

FIG. 5 shows the result of removing a box of an object smaller than the size of a preset box in step 220.

Referring to FIG. 5, the size of the preset box may be input by a user or may be set by receiving feedback through machine learning.

Even if you remove the box of small objects, you can see that there are still several rectangles in one cut.

In step 230, the apparatus for making a comic book into a plurality of cuts combines overlapping boxes among the boxes remaining on the page and regenerates them into one box.

As a result of the regeneration, each cut of the comic book can be clearly separated.

FIG. 6 shows the result of regenerating into one box by combining overlapping boxes in step 230.

Referring to FIG. 6, it can be seen that, by combining overlapping boxes and regenerating them into one box, it becomes a box corresponding to one cut.

In step 240, the apparatus for making the comic book into a plurality of cuts feeds back the learned box shape to the step 230 for a plurality of pages of the comic book.

Looking in more detail, when the regenerated box is displayed along with the page in step 230, each point of the regenerated box is moved to match the boundary of each cut of the page.

Then, by moving each point of the regenerated box with respect to a plurality of pages of the comic book to match the boundary of each cut of the page, the shape of the box applied to the comic book is learned, and the shape of the learned box is recalled. Feedback can be made to the box regeneration unit.

In step 230, overlapping boxes are combined using the shape of the box fed back in step 240 to regenerate into one box.

In step 250, the apparatus for making a comic book into a plurality of cuts separates the regenerated boxes into one cut.

In step 260, the apparatus for making a comic book into a plurality of cuts receives at least one of the number of cuts displayed on the screen, the sound effect corresponding to each cut, and the color applied to each cut, and corresponds to the separated cuts.

Embodiments of the present invention are implemented in the form of program instructions that can be executed through various computer means and can be recorded on computer readable media. The computer-readable medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and usable by those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, and magnetic media such as floptical disks. Includes hardware devices specifically configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language code that can be executed by a computer using an interpreter, etc., as well as machine language codes produced by a compiler. The hardware device described above may be configured to operate as one or more software modules to perform the operation of the present invention, and vice versa.

As described above, in the present invention, specific matters such as specific components and the like have been described by limited embodiments and drawings, but they are provided only to help a more comprehensive understanding of the present invention, and the present invention is not limited to the above embodiments , Anyone who has ordinary knowledge in the field to which the present invention pertains can make various modifications and variations from these descriptions.

Therefore, the spirit of the present invention is not limited to the described embodiments, and should not be determined, and all claims that are equivalent to or equivalent to the scope of the claims, as well as the claims described below, will be included in the scope of the spirit of the invention. .

Claims (9)

Recognizing an outline of an image on the input comic book page;
Detecting all objects on the page and creating a box;
Removing a box smaller than a preset box size from the page;
Combining the overlapping boxes among the boxes remaining on the page and regenerating into one box; And
And separating the regenerated boxes into one cut each,
The method of making a comic book into a plurality of cuts, characterized in that the object is a closed figure.
According to claim 1,
The box is a method of making a comic book into a plurality of cuts, characterized in that the object is the smallest polygon that the line segment of the box touches.
According to claim 1,
Displaying the regenerated box of the page together with the page; And
Moving each point of the regenerated box to match the boundary of each cut of the page;
Learning a shape of a box applied to the comic book by moving each point of the regenerated box with a plurality of pages of the comic book to match the boundary of each cut of the page; And
Further comprising the step of feedback to the step of regenerating the shape of the learned box to the one box,
The step of combining the overlapping boxes among the boxes remaining on the page and regenerating them into one box,
A method of making a comic book into a plurality of cuts, characterized by regenerating into a single box using the feedback box shape.
According to claim 1,
The comic book as a plurality of cuts further comprises the step of receiving at least one of the number of cuts displayed on the screen, the sound effect corresponding to each cut, and the color applied to each cut, and corresponding to the separated cuts. how to make.
A contour recognition unit that recognizes the contour of the image on the input comic book page;
A box generator that detects all objects on the page and creates a box;
A box regeneration unit that removes a box smaller than the size of a preset box from the page, and combines the overlapping boxes among the boxes remaining on the page to regenerate them into one box; And
It includes a separation unit for separating each of the regenerated boxes into a single cut,
The object is a device for making a comic book into a plurality of cuts, characterized in that the closed figure.
The method of claim 5,
The box is a device for making a comic book into a plurality of cuts, characterized in that the object is the smallest polygon that the line segment of the box touches.
The method of claim 5,
When the regenerated box of the page is displayed together with the page, each point of the regenerated box is moved to match the boundary of each cut of the page,
Learning the shape of the box applied to the comic book by moving each point of the regenerated box with respect to a plurality of pages of the comic book to match the boundary of each cut of the page,
Further comprising a machine learning learning unit for feeding back the shape of the learned box to the box regeneration unit,
The box regenerating unit is a device for making a comic book into a plurality of cuts, characterized in that using the form of the feedback box to regenerate into a single box.
The method of claim 5,
A plurality of comic books further comprising an editing unit that receives at least one of the number of cuts displayed on the screen, the sound effect corresponding to each cut, and the color applied to each cut, and corresponds to each cut separated from the separation unit. Device made of cuts of dogs.
A computer-readable recording medium recording a program for executing the method of any one of claims 1 to 4 in a computer.

Images