KR102592989B1 - Method and device for detecting and tracking video piracy - Google Patents

Abstract

A method and device for detecting and tracking video duplication are provided. A video copy detection and tracking method according to an embodiment of the present invention includes the steps of a server searching for and extracting an identification image including a plurality of dots from a video to be detected, and the server extracting identification information included in the extracted identification image. It may include a step of obtaining, wherein each of the plurality of dots is configured to have a size including a plurality of pixels, and the identification image may include different identification information according to the pattern of the plurality of dots.

Description

Video copy detection and tracking method and device {METHOD AND DEVICE FOR DETECTING AND TRACKING VIDEO PIRACY}

The present invention relates to a method and device for detecting and tracking video duplication.

As information and communication technologies and networks develop, various contents such as education, publishing, music, movies, and games that were provided offline are being produced in the form of digital data and distributed and provided online.

Unlike offline provision methods, content produced in the form of digital data is much easier to copy, transform, process, modify, and disseminate by obtaining the original, so there are frequent cases where the original creator of the content suffers damage due to illegal leakage, etc. It's happening a lot.

Therefore, from the original producer's perspective, in order to preserve the value of the produced content and generate legitimate profits, the content should be prevented from being copied and distributed without permission by unauthorized third parties, and if leaked, the leaker should be tracked down and appropriate compensation and punishment provided. A means to make a claim is essential.

In particular, video content among digital content is one of the most active areas for illegal leakage. In the case of video content, the URL where the video is located can be easily obtained, and illegal downloading is possible using various software and programs based on the obtained URL.

Meanwhile, a widely used method to prevent copying of video content is to use digital rights management (DRM) by encrypting the video itself. DRM refers to a solution that encrypts digital content so that only customers who have purchased it can use it. However, there was a problem that there were many ways to neutralize and avoid DRM, such as programs that could capture or record the entire video.

In addition, in the case of inserting identification information consisting of a string such as a user ID within a video frame, if the video with the identification information inserted goes through processing such as lossy compression or resolution change, part of the video is lost and part of the identification information is also lost. There was a problem that the leaker could not be traced.

Public Patent Publication No. 10-1439475, 2014.09.02

The present invention, in order to solve the above-mentioned problems, seeks to provide a video copy detection and tracking method and device that can generate a video that can trace the leaker of the unauthorized video copy.

In addition, the present invention seeks to provide a video copy detection and tracking method and device that can detect unauthorized copying of distributed videos and track the leaker.

In addition, the present invention seeks to provide a video copy detection and tracking method and device that inserts an identification image, rather than the identification information itself of a string, into a video and extracts identification information even if part of the identification image is lost during video processing.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned can be clearly understood by those skilled in the art from the description below.

The method of generating a corrected video according to an embodiment of the present invention to solve the above-described problem includes the steps of a server inserting an identification image including a plurality of dots into an original video, and the server inserting the identification image. and generating a corrected video, wherein each of the plurality of dots has a size including a plurality of pixels, and the identification image may include different identification information according to the pattern of the plurality of dots. .

Additionally, each of the plurality of dots may be configured at a size or interval that can maintain the dot pattern of the identification image even when the generated corrected video is processed.

Additionally, each of the plurality of dots may represent one bit value, and the identification image may represent a specific bit string according to the pattern of the plurality of dots.

Additionally, before inserting the identification image, the server may further include matching specific identification information to a specific bit string.

In addition, the step of inserting the identification image includes extracting a candidate area composed of colors within a reference value difference from a candidate frame of the original video, and setting the identification image input area within the candidate area to the average color of the candidate area. and inserting the identification image into the identification image input area with a color that differs from the average color by a minimum value.

In addition, the dots include first dots and second dots each representing different bit values, and at least one of the first dot and the second dot is a color that differs from the average color by a minimum value, and the second dot is One dot and the second dot may be composed of different colors.

A corrected video generating device according to another embodiment of the present invention includes an identification image insertion unit that inserts an identification image including a plurality of dots into an original video, and a corrected video generator that generates a corrected video with the identification image inserted. Each of the plurality of dots has a size including a plurality of pixels, and the identification image may include different identification information depending on the pattern of the plurality of dots.

A video copy detection and tracking method according to another embodiment of the present invention includes the steps of a server searching and extracting an identification image including a plurality of dots from a video to be detected, and the server extracting identification information included in the extracted identification image. It includes the step of obtaining, wherein each of the plurality of dots is configured in a size including a plurality of pixels, and the identification image may include different identification information according to the pattern of the plurality of dots.

In addition, each of the plurality of dots represents one bit value, the identification image represents a specific bit string according to the pattern of the plurality of dots, and the step of obtaining the identification information includes: It may include obtaining a bit string corresponding to the image and obtaining identification information matched to the bit string acquired by the server.

In addition, the step of searching and extracting the identification image includes the steps of the server extracting one or more candidate frames from the video to be detected, and the step of the server searching and extracting the identification image for a candidate area within the candidate frame. Including, the candidate area may be an area composed of a color within the difference between the reference value.

Additionally, the step of searching for and extracting the identification image for the candidate area may be searching for and extracting the identification image from an area in which the colors of the candidate area differ by a minimum value.

A video copy detection and tracking device according to another embodiment of the present invention includes an identification image extraction unit that searches for and extracts an identification image including a plurality of dots from a video to be detected, and identification information included in the extracted identification image. and an identification information acquisition unit, wherein each of the plurality of dots has a size including a plurality of pixels, and the identification image may include different identification information according to the pattern of the plurality of dots.

Other specific details of the invention are included in the detailed description and drawings.

According to the present invention as described above, a video author can create a corrected video that can track the leaker in preparation for the original video being copied and leaked without permission. Therefore, video authors can prepare in advance for damages arising from video leaks, and there is an effect of preventing video playback requesters from copying and distributing the video without permission.

In addition, according to the present invention, it is possible to determine whether a distributed video has been copied without permission, and if so, to detect and track the person who leaked the video. Therefore, when damage occurs due to video leakage, the video author can track the leaker and demand compensation from the leaker or request that the leaker be punished.

In addition, according to the present invention, the identification image inserted in the corrected video is inserted in a way that is difficult for the viewer to recognize with the naked eye, so there is an effect of preventing unauthorized copying of the video without causing inconvenience to the viewer watching the video. .

In addition, according to the present invention, even if a video containing an identification image is processed and part of the identification image is lost, it is possible to extract identification information from it and track the leaker.

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

1 is a schematic diagram of a video copy detection and tracking system according to an embodiment of the present invention.
Figure 2 is a flowchart of a method for generating a corrected video according to an embodiment of the present invention.
Figure 3 is an example diagram of an original video and a corrected video according to an embodiment of the present invention.
Figure 4 is an exemplary diagram of an identification image according to an embodiment of the present invention.
Figure 5 is a flowchart of a method for generating a corrected video further including an identification information matching step according to an embodiment of the present invention.
Figure 6 is an example diagram illustrating matching of identification information to an identification image according to an embodiment of the present invention.
Figure 7 is a flowchart of a method for inserting an identification image into a candidate area of a candidate frame according to an embodiment of the present invention.
Figure 8 is an example diagram for explaining the candidate frame extraction step according to an embodiment of the present invention.
Figure 9 is a flowchart of a method for inserting an identification image into an identification image input area according to an embodiment of the present invention.
Figure 10 is an example diagram of an original video and a corrected video according to an embodiment of the present invention.
Figures 11 and 12 are exemplary diagrams to explain how an identification image is inserted into an identification image input area according to an embodiment of the present invention.
Figure 13 is a flowchart of a method for inserting an identification image into an average color area according to an embodiment of the present invention.
Figure 14 is an example diagram explaining a candidate area and an average color area within a candidate frame according to an embodiment of the present invention.
Figure 15 is a flowchart of a method of dividing an identification image into a plurality of parts and inserting each into a plurality of average color areas according to an embodiment of the present invention.
Figure 16 is an exemplary diagram showing a case where there are a plurality of average color regions according to an embodiment of the present invention.
Figures 17 and 18 are flowcharts of a video copy detection and tracking method according to another embodiment of the present invention.

The advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. The present embodiments are merely provided to ensure that the disclosure of the present invention is complete and to provide a general understanding of the technical field to which the present invention pertains. It is provided to fully inform the skilled person of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terminology used herein is for describing embodiments and is not intended to limit the invention. As used herein, singular forms also include plural forms, unless specifically stated otherwise in the context. As used in the specification, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other elements in addition to the mentioned elements. Like reference numerals refer to like elements throughout the specification, and “and/or” includes each and every combination of one or more of the referenced elements. Although “first”, “second”, etc. are used to describe various components, these components are of course not limited by these terms. These terms are merely used to distinguish one component from another. Therefore, it goes without saying that the first component mentioned below may also be a second component within the technical spirit of the present invention.

As used in the specification, the term “unit” or “module” refers to a hardware component such as software, FPGA, or ASIC, and the “unit” or “module” performs certain roles. However, “part” or “module” is not limited to software or hardware. A “unit” or “module” may be configured to reside on an addressable storage medium and may be configured to run on one or more processors. Thus, as an example, a “part” or “module” refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, Includes procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functionality provided within components and “parts” or “modules” can be combined into smaller components and “parts” or “modules” or into additional components and “parts” or “modules”. Could be further separated.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless clearly specifically defined.

In this specification, 'original video' is the final version for distribution of a video produced by the original author of the video, and refers to the original video without any processing, modification, alteration, insertion of identification information, or reproduction.

In this specification, 'corrected video' refers to a video that is created by inserting a user's identification image in a specific way into the original video, so that the leaker can be traced.

In this specification, ‘frame’ refers to a cut of an image used in a video.

In this specification, 'Key Frame' refers to the most central frame, such as the start frame and end frame of a single action.

In this specification, ‘identification information’ refers to the identification information of the user who requested to play the video. For example, it may include, but is not limited to, the user's identification ID, playback device identification number, network IP address (Internet Protocol Address), MAC address (Media Access Control Address), playback request time, etc. A variety of information may be included.

Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

1 is a schematic diagram of a video copy detection and tracking system according to an embodiment of the present invention.

Referring to FIG. 1, a system 1000 for detecting and tracking video duplication according to an embodiment of the present invention may include a corrected video creation server 100 and a video duplication detection and tracking server 200.

The corrected video creation server 100 and the video copy detection and tracking server 200 are not necessarily separate servers and may be managed as one server.

The corrected video creation server 100 receives the original video 10 and inserts an identification image 60 corresponding to the identification information of the user who requested playback into the original video 10 to create a corrected video ( 20) plays a role in generating.

To this end, the corrected video creation server 100 may insert the identification image 60 into the original video 10.

The video copy detection and tracking server 200 determines whether the distributed video is a copied video and, if it is an unauthorized video, tracks the leaker.

To this end, the video copy detection and tracking server 200 can extract the identification image 60 from the video to be detected. Subsequently, the video copy detection and tracking server 200 may extract identification information 80 included in the extracted identification image 60.

Hereinafter, a method by which the corrected video generating server 100 generates the corrected video 20 will be described in detail.

Figure 2 is a flowchart of a method for generating a corrected video according to an embodiment of the present invention. Figure 3 is an example diagram of an original video and a corrected video according to an embodiment of the present invention.

Referring to Figures 2 and 3, the method of generating a corrected video according to an embodiment of the present invention includes the step of the corrected video generating server inserting an identification image into the original video (S200), and the corrected video generating server inserting the identification image into the original video. It includes a step of generating an inserted correction video (S300). Below, a detailed description of each step is provided.

The corrected video creation server 100 inserts the identification image 60 into the original video 10 (S200).

The identification image 60 is an image corresponding to identification information and refers to an image inserted into the original video 10 in place of identification information in the form of a string.

Figure 4 is an exemplary diagram of an identification image according to an embodiment of the present invention.

Referring to FIG. 4, the identification image 60 according to an embodiment of the present invention may be an image including a plurality of dots. As a specific example, the identification image 60 may be an image in which a plurality of dots are arranged as shown in FIG. 4. Figure 4 shows an example of an identification image 60 in which a plurality of dots are sequentially arranged, but the identification image 60 of the present invention is not limited to this, and each dot is arranged spaced apart from each other or forms a specific shape. It may be arranged to do so.

The identification image 60 may include different identification information depending on the pattern in which the plurality of dots are arranged, which will be described in detail later.

In one embodiment, each dot constituting the identification image 60 can store 1 bit of data. That is, each dot included in the identification image can represent one bit value (0 or 1).

As a specific example, the plurality of dots may include a first dot 61 and a second dot 62 each representing different bit values. For example, referring to FIG. 4(a), the first dot 61 may mean '0' and the second dot 62 may mean '1'.

In the above-described embodiment, as each dot represents one bit value, the identification image 60 including a plurality of dots may represent a specific bit string. For example, the identification image 60 in FIG. 4(a) may represent a bit string of '10011010'.

Additionally, in one embodiment, the number of dots constituting the identification image 60 may be determined in various ways by the correction video creation server 100. In other words, the larger the number of dots that make up one identification image 60, the larger the size of data that can be stored, so a large number of identification information 80 can be stored, but the identification image 60 is not included in the original video 10. The area required for insertion may be expanded.

Specifically, when the identification image 60 includes 8 dots as shown in FIG. 4(a), the storable data is 8 bits, and a total of 2 ⁸ =256 pieces of identification information can be stored. The number of dots included in the identification image 60 can be determined in various ways, such as 8, 16, 32, or 64, but for convenience of explanation, eight examples are used below.

Additionally, in one embodiment, each of the plurality of dots may be configured at a size or interval that can maintain the dot pattern (i.e., bit string) of the identification image 60 even when the generated corrected video 20 is processed. there is.

As a specific example, each dot may be configured to include a plurality of pixels. When the corrected video 20 in which the identification image 60 is inserted undergoes processing such as lossy compression or resolution change, some pixels of the inserted identification image 60 may be lost. At this time, in order to enable restoration of the identification image 60 even if some pixels of the identification image 60 are lost (i.e., to enable acquisition of a bit string), each dot constituting the identification image 60 is divided into multiple dots. It can be configured in a size containing pixels.

The number of pixels constituting each dot may be determined in various ways by the corrected video creation server 100. Specifically, referring to FIG. 4(b), each dot 61 and 62 is composed of a size including 25 pixels (5x5), but this is only an example and is not limited thereto. As the number of pixels constituting each dot increases, the possibility of restoring the identification image 60 during video processing may increase, but the area required for the identification image 60 to be inserted into the original video 10 may increase.

In step S200, the identification image 60 may be inserted into a partial area of the original video 10 so that it is difficult for the viewer to visually detect the inserted identification image when watching the corrected video 20.

When the identification image 60 is inserted into the original video 10, the corrected video server 100 generates the corrected video 20, which is a video in which the identification image 60 is inserted (S200).

Figure 5 is a flowchart of a method for generating a corrected video further including an identification information matching step according to an embodiment of the present invention. Figure 6 is an example diagram illustrating matching of identification information to an identification image according to an embodiment of the present invention.

Referring to Figures 5 and 6, the method of generating a corrected video according to an embodiment of the present invention may further include a step (S100) in which the corrected video generating server matches identification information to an identification image before step S200. there is.

The corrected video creation server 100 matches specific identification information 80 to a specific identification image 60 (S100).

In one embodiment, each identification image 60 may be matched with different identification information 80 on a one-to-one basis. That is, the identification image 60 may include different identification information depending on the pattern in which the plurality of dots are arranged.

As a specific example, step S100 may involve the corrected video creation server 100 matching specific identification information 80 to a specific bit string.

For example, referring to Figure 6, 'Identification information A' matches the bit string '00000001', 'Identification information B' matches the bit string '00000010', and 'Identification information C' matches the bit string '00000011'. It can be. Accordingly, identification images A, B, and C representing each bit string may include identification information A, B, and C, respectively. For example, identification information A, B, and C may be the identification IDs of users A, B, and C.

Additionally, the corrected video creation server 100 may update the identification information 80 matched to the identification image 60. For example, the corrected video creation server 100 may update the identification information 80 matched to the bit string at predetermined intervals. Alternatively, when new identification information is registered (eg, new user ID registration), the new identification information can be matched to a previously unmatched bit string.

According to the present invention, each identification image 60 representing a specific bit string is matched one-to-one with specific identification information 80, and the identification image is generated so that the dot, which is a unit constituting the identification image, includes a plurality of pixels, Even if part of the identification image is lost, the identification information 60 can be obtained.

Figure 7 is a flowchart of a method for inserting an identification image into a candidate area of a candidate frame according to an embodiment of the present invention.

Referring to FIG. 7, in one embodiment, step S200 includes extracting a candidate frame into which an identification image can be inserted from the original video (S220), extracting a candidate area from the candidate frame (S240), and It may include inserting an identification image (S260). Below, a detailed description of each step is provided.

The corrected video creation server 100 extracts a candidate frame 30 into which the identification image 60 can be inserted from the original video 10 (S220).

Figure 8 is an example diagram for explaining a step (S220) of extracting a candidate frame from a plurality of frames constituting a video according to an embodiment of the present invention.

A video is a video that appears to be moving to the human eye by continuously collecting and displaying images called frames. Therefore, one video consists of a plurality of frames, and the higher the frame, the smoother and clearer the video becomes. Meanwhile, the number of frames that make up one second of video is expressed in terms of x frames per second, fps, etc.

Step S220 is a candidate frame including a candidate area 40 that satisfies the requirements for inserting the user's identification image 60 among the plurality of frames constituting the original video 10. This is the stage of selecting and extracting 30).

The candidate area 40 is a space larger than a specific range in which the identification image 60 within the frame can be inserted, and may mean an area composed of colors within a difference in a reference value.

In one embodiment, the reference value may be set differently depending on the type of the original video 10, and may mean that no difference in color is felt when the frame is viewed with the naked eye or that it is a difference value at a very subtle level.

Referring to FIG. 8, a portion of the frames constituting the video is shown as an example. Among the plurality of frames shown, the frames in the first half include a candidate area 40, which is an area composed of colors within a difference in a reference value at the same or similar position. The corrected video creation server 100 may extract frames sharing the same characteristics as candidate frames 30.

In one embodiment, the candidate frame 30 may form a plurality of candidate groups, and each candidate group may be composed of candidate frames 30 into which identification information of the same color can be inserted. The candidate group may be selected based on a key frame, but is not limited to this, and various methods may be applied to extract frames that share a color less than or equal to the reference value.

Next, the corrected video generation server 100 extracts the candidate area 40 from the candidate frame 30 (S240).

As described above, the candidate area 40 may refer to an area composed of a color within the difference between the reference values shared by the candidate frames 30. That is, if the color difference between adjacent areas of a specific area on the screen constituting the frame is within the reference value, the area may become the candidate area 40. Accordingly, the candidate frames 30 sharing the same candidate area 40 can insert the identification image 60 at the same location with the same color and size.

When extracting the candidate area 40 and inserting the identification image 60, unlike when the identification image is inserted at a random position, the identification image is targeted to all areas of all frames in the duplicate video detection and tracking step described later. It is possible to select and detect the candidate frame 30 and candidate area 40 where the identification image 60 may be inserted without the need to detect. Therefore, the present invention has the effect of enabling efficient and rapid detection of duplicate videos.

Next, the corrected video generation server 100 inserts the identification image 60 into the candidate area 40 within the candidate frame 30 (S260).

In step S200, the identification image 60 may be inserted into the candidate area 40 within the candidate frame 30 so that it is difficult for the viewer to visually detect the inserted identification image when watching the corrected video 20.

Figure 9 is a flowchart of a method for inserting an identification image into an identification image input area according to an embodiment of the present invention.

Referring to FIG. 9, in one embodiment, step S260 may include setting the identification image input area to the average color of the candidate area (S261) and inserting the identification image into the identification image input area (S262). there is.

The corrected video generation server 100 sets the identification image input area 50 to the average color of the candidate area 40 (S261).

The identification image input area 50 may refer to an area where the identification image 60 is actually located.

As the correction video creation server 100 sets the identification image input area 50 to the average color of the candidate area 40, it is difficult for viewers of the correction video 20 to discover the identification image input area 50 with the naked eye. Thus, you can avoid any inconvenience in viewing.

Next, the corrected video generation server 100 inserts the identification image 60 into the identification image input area 50 (S262). That is, the identification image 60 can be inserted to be located in the identification image input area 50 set to the average color of the candidate area 40.

In one embodiment, the identification image 60 may be composed of a color that differs from the average color of the candidate area 40 by a minimum value.

Specifically, when color is expressed in the RGB (Red, Green, Blue) system, the color intensity can be expressed as a number from 0 to 255. The closer it is to 255, the higher the color density of that series is, and the closer it is to 0, the lower the color density of that series is. At this time, the minimum value may mean a difference of 1 in the RGB system.

As a specific example, the color of the identification image 60 may be 1 higher or lower than the average color of the candidate area 40 (i.e., the color of the identification image input area 50).

For example, if the average color of the candidate area 40 is a green color among the three primary colors and the green value is 180, the color of the identification image 60 is the average color and the values of red and blue. The value of (Blue) is the same, but the value of green can be set to 181 or 179, which is different from 180 by 1.

Since this corresponds to a color difference that is difficult to distinguish with the naked eye, it does not cause viewing discomfort to those requesting video playback. Meanwhile, in the detection stage, the color difference as much as the minimum value plays a very important role, and a detailed description of this will be provided later.

In addition, in one embodiment, at least one of the first dot 61 and the second dot 62 in the identification image 60 is composed of a color that differs from the average color by a minimum value, and the first dot and the second dot Dots may be composed of different colors. This will be described in detail later.

Figure 10 is an example diagram of an original video and a corrected video according to an embodiment of the present invention.

According to the example shown in FIG. 10, the identification image (bit string '10011010') of FIG. 4 may be inserted into the corrected video 20. As described above, specific identification information 80 may be matched to a specific identification image 60. For example, the identification information 80 may include, but is not limited to, the identification number of the playback device, network IP address (Internet Protocol Address), MAC address (Media Access Control Address), and playback request time in addition to the user's identification ID. . Meanwhile, the inserted identification image 60 is not actually visible to the naked eye to the playback requester, so the viewer cannot recognize it during video playback.

Figures 11 and 12 are exemplary diagrams to explain how an identification image is inserted into an identification image input area according to an embodiment of the present invention.

Referring to FIG. 11, the identification image input area 50 may include a space large enough to insert the identification image 60.

In one embodiment, the plurality of dots 61 and 62 included in the identification image 60 may be arranged in a row as shown in FIGS. 11(a) and 11(b).

In another embodiment, the plurality of dots 61 and 62 included in the identification image 60 may be arranged according to a predetermined rule to include a plurality of rows or columns depending on the size of the identification image input area 50. .

For example, a plurality of dots 61 and 62 may be arranged sequentially from the upper left to the lower right as shown in FIG. 11(c). In this case, when detecting and tracking video duplication, a bit string of '10011010' can be obtained by sequentially restoring a plurality of dots 61 and 62 from the upper left to the lower right of the identification image 60.

In addition, the identification image 60 of FIG. 11 shows an example in which a plurality of dots 61 and 62 are arranged adjacently, but as described above, the dots in the identification image may be arranged to be spaced apart from each other.

Additionally, in one embodiment, at least one of the first dot 61 and the second dot 62 in the identification image 60 is the color of the identification image input area 50 (i.e., the average of the candidate area 40). Color) may be composed of colors that differ by a minimum value, and the first dot 61 and the second dot 62 may be composed of different colors.

As a specific example, as shown in FIG. 12(a), the color of the first dot 61 is 1 lower than the color of the identification image input area 50, and the color of the second dot 62 is 1 lower than that of the identification image input area 50. It can be 1 higher than the color.

As another specific example, as shown in FIG. 12(b), the color of the first dot 61 is the same as the color of the identification image input area 50, and the color of the second dot 62 is the color of the identification image input area 50. It can be 1 higher than the color.

As another specific example, as shown in FIG. 12(c), the color of the first dot 61 is 1 higher than the color of the identification image input area 50, and the color of the second dot 62 is 1 higher than the color of the identification image input area 50. ) may be higher by 2 (that is, the color of the second dot 62 may be higher than the color of the first dot 61 by 1).

The color difference of the identification image 60 inserted into the identification image input area 50 has been described above with reference to FIG. 12, but this is only an example and is not limited thereto. In addition, the example diagram shows the identification information input area 50 and the identification image 60 to be clearly distinguishable for explanation purposes, but in the actual playback screen, they cannot be distinguished with the naked eye and the inserted identification information cannot be recognized.

Figure 13 is a flowchart of a method for inserting an identification image into an average color area according to an embodiment of the present invention.

Referring to FIG. 13, in one embodiment, step S260 may include extracting an average color area from the candidate area (S265) and inserting an identification image into the average color area (S266). Hereinafter, the description will focus on the differences from FIG. 9.

The corrected video generation server 100 extracts the average color area 70 from the candidate area 40 (S265).

The average color area 70 is a specific area within the candidate area 40, is composed of the average color of the candidate area 40, and means an area containing a space into which the identification image 60 can be inserted.

Next, the corrected video generation server 100 inserts the identification image 60 into the average color area 70. That is, according to this embodiment, instead of extracting a separate identification image input area 50 and setting it to the average color of the candidate area 40, the average color area 70 composed of the average color of the candidate area is extracted and identified. An image 60 can be inserted. Accordingly, it becomes more difficult to distinguish with the naked eye than when the identification image input area 50 is located in another space within the candidate area 40, which has the effect of making the viewing of the video playback requester more comfortable.

Figure 14 is an example diagram explaining the candidate area and average color area within the candidate frame according to an embodiment of the present invention.

Within the candidate frame 30, there is a candidate area 40 with a color difference less than the reference value. In the candidate area 40, there is an area composed of a color that does not exactly match the average color of the candidate region 40 and an average color region 70 composed of the average color of the candidate region 40.

In the example diagram, the candidate area 40 and the average color area 70 are shown to be distinct for explanation purposes, but they cannot be distinguished with the naked eye in the actual playback screen.

According to various embodiments of the present invention, the identification image 60 can be inserted into the candidate area 40 within the candidate frame 30 of the video without a separate identification image input area 50. In other words, the identification image input area 50 corresponds to a means used to facilitate searching for the user's identification information 60 by creating an area where the colors differ by the minimum value in the duplicate video detection and tracking stage.

In one embodiment, the identification image 60 may not be inserted into a random area within the candidate area 40 of the candidate frame 30, but may be inserted into the average color area 70. At this time, the color of the identification image 60 may be composed of a color that differs from the average color by a minimum value. Therefore, as in the case of the embodiment in which the identification image 60 is inserted into the identification image input area 50 set to the average color described above, the area in which the color differs by the minimum value can be detected in the detection step.

Next, a method for generating a corrected video in an embodiment with a plurality of average color areas 70 will be described.

Figure 15 is a flowchart of a method of dividing an identification image into a plurality of parts and inserting each into a plurality of average color areas according to an embodiment of the present invention. Figure 16 is an exemplary diagram showing a case where there are a plurality of average color regions according to an embodiment of the present invention.

Referring to Figures 15 and 16, when a plurality of average color regions 70 exist, step S260 is a step of extracting an average color region from the candidate region (S265), and extracting the first average color region and the second average color region from the average color region. 2 Step of extracting the average color area (S267), separating the identification image into a first part and a second part (S268), dividing the first and second parts of the identification image into the first average color area and the second average It may include a step (S269) of inserting each color into the color area.

Following the above-described step S265, the corrected video generation server 100 extracts the first average color area 71 and the second average color area 72 from the plurality of average color areas 70 (S267).

The plurality of average generated areas 70 are all composed of the average color of the candidate area 40, and are identical in that they include a space into which the identification image 60 can be inserted. Meanwhile, the area and location of each average color area 70 may be different. Accordingly, the first average color area 71 and the second average color area 72 can be extracted according to a specific standard. For example, based on area, the widest average color area can be selected as the first average color area 71, and the next wide average color area can be selected as the second average color area 72. However, it is not limited to this, and various criteria other than area may be applied.

Referring to the example shown in FIG. 16, there are a plurality of average color areas 70 in the candidate area 40, which are composed of the average color of the candidate areas and into which the user's identification information 60 can be inserted.

As shown in FIG. 16, when there are multiple average color areas 70, each area can be set as the first average color area 71 and the second average color area 72. The first average color area 71 and the second average color area 72 may correspond to the two widest areas among the plurality of average color areas 70. Meanwhile, Figure 16 shows an example of extracting two average color areas, but if necessary, up to the nth average color area can be extracted.

Next, the corrected video generation server 100 separates the identification image 60 into a first part 65 and a second part 66 (S268). The identification image 60 can be separated in various ways depending on the area and number of extracted average color regions 71 and 72.

Next, the corrected video generation server 100 inserts the first part 65 and the second part 66 of the identification image into the first average color area 71 and the second average color area 72, respectively (S269) ).

For example, referring to FIG. 16, the identification image 60 representing the bit string '10011010' is divided into a first part 65 of '1001' and a second part 66 of '1010' and a first average is obtained. It can be inserted into the color area 71 and the second average color area 72, respectively.

Through this, if there is not enough space in any one of the plurality of average color areas 70 to insert the identification image 60 by itself, the identification image 60 is inserted separately, so that it can still be used according to the embodiment of the present invention. Implementation is possible.

The corrected video generating server 100 according to another embodiment of the present invention includes an identification image insertion unit and a corrected video generating unit, and can perform the above-described corrected video generating method.

Hereinafter, a method for detecting and tracking video duplication by the video duplication detection and tracking server 200 will be described.

Figures 17 and 18 are flowcharts of a video copy detection and tracking method according to another embodiment of the present invention.

Referring to Figure 17, the video copy detection and tracking method according to another embodiment of the present invention includes the steps of searching and extracting an identification image from the video to be detected (S1100), and obtaining identification information included in the extracted identification image. Includes step S1200.

The video copy detection and tracking server 200 searches for and extracts an identification image from the video to be detected (S1100). The detection target video may be the corrected video 20 generated by the corrected video generating method of the present invention described above.

Next, the video copy detection and tracking server 200 obtains the identification information 80 included in the extracted identification image 60 (S1200).

As in the above-described embodiment, when the identification image 60 inserted into the video to be detected represents a specific bit string, and different identification information is matched for each bit string, step S1200 obtains the bit string corresponding to the extracted identification image. It may include a step (S1210) and a step (S1220) of obtaining identification information matched to the obtained bit string.

That is, the video copy detection and tracking server 200 may obtain a specific bit string based on the dot pattern of the identification image 60 extracted in step S1100 (S1210).

Subsequently, the video copy detection and tracking server 200 may obtain specific identification information (eg, user identification ID) matched to the acquired bit string (S1220).

Also, referring to FIG. 18, in one embodiment, step S1100 includes extracting a candidate frame from a video to be detected (S1110) and searching and extracting an identification image for a candidate area within the candidate frame (S1120). can do.

According to the present invention, for video copy detection and tracking, it is necessary to extract a specific candidate frame 30 and search only a specific candidate area 40, without having to search the entire area of all the plurality of frames constituting the detection target video. . Therefore, it has the effect of enabling more efficient and economical detection of duplicate videos and tracking of leakers.

Since step S1110 is substantially the same as step S100 described in the previously described correction video generation method, detailed description is omitted.

Next, the video copy detection and tracking server 200 searches for and extracts the identification image 60 for the candidate area 40 within the candidate frame 30 (S1120).

According to one embodiment of the present invention, the identification image input area 50 within the candidate area 40 is set to the average color of the candidate area. Meanwhile, the inserted identification image 60 consists of a color that differs from the average color by the minimum value. Specifically, at least one of the first dot 61 and the second dot 62 in the identification image 60 has the minimum value in the color of the identification image input area 50 (i.e., the average color of the candidate area 40). It can be made up of different colors.

The video copy detection and tracking server 200 uses the characteristic that the color value of the identification image input area 50 and the color value of the identification image 60 differ by the minimum value, and detects the identification image inserted in the candidate area 40 ( 60) can be explored. That is, step S1120 may include detecting an area where colors differ by a minimum value and extracting an identification image 60 from the detected area.

As described above, when a color is expressed in the RGB system, the minimum color value is 1 in the color values displayed from 0 to 255. Accordingly, the video copy detection and tracking server 200 searches for an area in the candidate area 40 to be detected where the difference in color value is the minimum value of 1. When an area in which the difference in color value differs by the minimum value of 1 is detected, the video copy detection and tracking server 200 can extract and obtain an identification image 60 from the detected area.

For example, the color value of the identification image input area 50 is the average color of the candidate area 40, and the case where the green color value of the average color is 180 will be described. At this time, the red and blue color values of the color of the identification image 60 are the same as the color value of the identification image input area 50, and the green color value may be 181 (or 179). there is). The video copy detection and tracking server 200 searches for an area where the difference in green color values differs by the minimum value of 1. The green color values of the identification image input area 50 and the identification image 60 are 180 and 181, respectively, which differ by the minimum value of 1, so they are detected by the video copy detection and tracking server 200.

As described above, when an area in which the color value difference differs by the minimum value of 1 is detected, the video copy detection and tracking server 200 extracts and obtains an identification image 60 from the detected area.

On the other hand, if an area where the color value difference differs by the minimum value of 1 is not found, a new search operation is performed for other candidate frames, or until an area where the color value difference differs by the minimum value of 1 is found. The search operation may not stop.

According to another embodiment of the present invention, after the step of extracting the candidate frame 30 from the video to be detected (S1110), the step of extracting the average color area 70 may be further included. This is a case in which the identification image 60 is inserted into the average color area 70 composed of the average color of the candidate area 40 existing within the candidate area 40 during the process of generating the corrected video. In this case, the video copy detection and tracking server 200 does not need to search the entire candidate area 40 to be detected, but only searches the extracted average color area 70. Accordingly, the search target area becomes narrower, enabling faster and more efficient search.

Next, a case where multiple areas with minimum color value differences are found in the video copy detection and tracking stage will be described.

Referring again to FIG. 16, the identification image representing the bit string '10011010' is divided into '1001' as the first part 65 and '1010' as the second part 66, and is divided into the first average color area. The first part 65 is inserted at 71 and the second part 66 is inserted into the second average color area 72. The video copy detection and tracking server 200 obtains parts of the identification image representing bit strings '1001' and '1010' from the first average color area 71 and the second average color area 72, respectively, and combines them. Thus, the entire bit string of '10011010' can be obtained.

In one embodiment, the order of combining parts of the extracted identification image (or bit string) may be applied in the same manner as the criteria for selecting the first average color region 71 and the second average color region 72. As in the above-described embodiment, the widest average color area based on area among the plurality of average raw color areas 70 is referred to as the first average color area 71, and the next wide average color area is referred to as the second average color area ( 72), the order of combination follows accordingly. That is, the identification image detected in the first average color area 71 is set as the first part 65, and the identification image detected in the second average color area 72 is set as the second part 66, and the first part ( 65) The entire identification image can be obtained by combining the second part 66.

The video copy detection and server 200 according to another embodiment of the present invention includes an identification image extraction unit and an identification information acquisition unit, and can perform the video copy detection and tracking method described above.

In addition, the corrected video generation method or video copy detection and tracking method of the present invention described above may be implemented as a program (or application) and stored in a medium in order to be executed in conjunction with a computer, which is hardware.

The steps of the method or algorithm described in connection with embodiments of the present invention may be implemented directly in hardware, implemented in a software model executed by hardware, or a combination thereof. The software model can be RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains.

Above, embodiments of the present invention have been described with reference to the attached drawings, but those skilled in the art will understand that the present invention can be implemented in other specific forms without changing its technical idea or essential features. You will be able to understand it. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive.

10: Original video
20: Correction video
30: Candidate frame
40: Candidate area
50: Identification image input area
60: Identification image
70: Average color gamut
80: User identification information

Claims (10)

A server inserting an identification image including a plurality of dots into the original video; and
A step of the server generating a corrected video into which the identification image is inserted,
Each of the plurality of dots is composed of a size including a plurality of pixels and represents one bit value,
The identification image is a string of identification information expressed as a specific bit string according to the pattern of the plurality of dots,
In each of a plurality of candidate frames sharing the same candidate area among all frames of the original video, the bit string is inserted identically at the same position with the same color and size,
The bit string is distributed to include a plurality of rows and columns in the order from upper left to lower right according to the size of the identification image input area in the candidate area. According to claim 1,
Each of the plurality of dots is configured with a size or interval that allows the dot pattern of the identification image to be maintained even when the generated correction video is processed. delete According to claim 1,
Before the insertion step,
A method of generating a corrected video further comprising the step of the server matching one piece of identification information to one bit string. According to claim 1,
The insertion step is,
extracting the candidate area composed of a color within a difference between a reference value from the plurality of candidate frames;
setting the identification image input area within the candidate area to an average color of the candidate area; and
Inserting the identification image into the identification image input area with a color that differs from the average color by a minimum value,
The dots include first dots and second dots, each representing a different bit value,
At least one of the first dot and the second dot is a color that differs from the average color by a minimum value,
A method of generating a corrected video, characterized in that the first dot and the second dot are composed of different colors. An identification image insertion unit that inserts an identification image including a plurality of dots into the original video; and
Comprising a correction video generator that generates a correction video into which the identification image is inserted,
Each of the plurality of dots is composed of a size including a plurality of pixels and represents one bit value,
The identification image is a string of identification information expressed as a specific bit string according to the pattern of the plurality of dots,
In each of a plurality of candidate frames sharing the same candidate area among all frames of the original video, the bit string is inserted identically at the same position with the same color and size,
The bit string is distributed to include a plurality of rows and columns in the order from upper left to lower right according to the size of the identification image input area in the candidate area. A server searching for and extracting an identification image including a plurality of dots from a video to be detected; and
Comprising the step of the server obtaining identification information included in the extracted identification image,
Each of the plurality of dots is composed of a size including a plurality of pixels and represents one bit value,
The identification image is one in which the identification information of a string is expressed as a specific bit string according to the pattern of the plurality of dots,
In each of the plurality of candidate frames sharing the same candidate area among all frames of the detection target video, the bit string is equally inserted at the same position with the same color and size,
The bit string is distributed to include a plurality of rows and columns in the order from upper left to lower right according to the size of the identification image input area in the candidate area. According to clause 7,
The acquisition step is,
Obtaining, by the server, the bit string corresponding to the extracted identification image; and
A video copy tracking method comprising the step of obtaining the identification information matched to the bit string acquired by the server. According to clause 7,
The step of searching and extracting the identification image is,
The server extracting the plurality of candidate frames from the detection target video; and
A step of the server searching and extracting the identification image from an area in which colors differ by a minimum value among the candidate areas within the plurality of candidate frames,
A video copy tracking method, characterized in that the candidate area is an area composed of colors within a difference between reference values. An identification image extraction unit that searches for and extracts an identification image including a plurality of dots from a video to be detected; and
It includes an identification information acquisition unit that acquires identification information included in the extracted identification image,
Each of the plurality of dots is composed of a size including a plurality of pixels and represents one bit value,
The identification image is one in which the identification information of a string is expressed as a specific bit string according to the pattern of the plurality of dots,
In each of the plurality of candidate frames sharing the same candidate area among all frames of the detection target video, the bit string is equally inserted at the same position with the same color and size,
The bit string is distributed to include a plurality of rows and columns in the order from upper left to lower right according to the size of the identification image input area in the candidate area.

Images