KR101674310B1 - System and method for matching advertisement for providing advertisement associated with video contents - Google Patents

System and method for matching advertisement for providing advertisement associated with video contents Download PDF

Info

Publication number
KR101674310B1
KR101674310B1 KR1020150108831A KR20150108831A KR101674310B1 KR 101674310 B1 KR101674310 B1 KR 101674310B1 KR 1020150108831 A KR1020150108831 A KR 1020150108831A KR 20150108831 A KR20150108831 A KR 20150108831A KR 101674310 B1 KR101674310 B1 KR 101674310B1
Authority
KR
South Korea
Prior art keywords
content
matching
image
index data
advertisement
Prior art date
Application number
KR1020150108831A
Other languages
Korean (ko)
Inventor
이성용
한정석
Original Assignee
이성용
한정석
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 이성용, 한정석 filed Critical 이성용
Priority to KR1020150108831A priority Critical patent/KR101674310B1/en
Priority to PCT/KR2015/008604 priority patent/WO2017022880A1/en
Application granted granted Critical
Publication of KR101674310B1 publication Critical patent/KR101674310B1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0242—Determining effectiveness of advertisements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0255—Targeted advertisements based on user history
    • G06Q30/0256—User search
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0277—Online advertisement
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

The present invention relates to an advertisement matching system and method for providing advertisement contents which are closely related to image contents by extracting index data representative of contents from each of image contents and advertisement contents, . In the present invention, the viewer viewing the image content can recognize the advertisement content as if it is a part of the image content, and can also display the advertisement related to the image content to the viewers who know the content of the image content to some extent So that it is possible to increase the concentration of the viewer on the advertisement viewing.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to an advertisement matching system and a method for providing an advertisement having relevance to image contents,

The present invention relates to an advertisement matching system and method for providing advertisements having relevance to video contents including movies, real-time broadcast programs, VOD (Video on Demand) contents, Internet videos, and the like.

In general, advertising is a type of advertisement that is used to promote consumers' demands and purchasing behavior by presenting the existence, characteristics, and convenience of goods and services to an unspecified number of consumers, or to increase the general attitude and reliability of advertisers themselves. It can be called massive mass communication.

These advertisements are typically exposed before a movie is shown in a theater or before a broadcast program or VOD content is played. At this time, advertisements may be organized in front of a popular movie or broadcast program or VOD content to enhance the advertisement effect, And a method of repeatedly reproducing an advertisement.

However, the advertisement exposed prior to playing the video content is provided randomly regardless of the viewers or viewers watching the video content or the video content, which causes the viewer to feel a sense of rejection, which reduces the effectiveness of the advertisement and hinders the rights of viewers and viewers .

Related Prior Art Documents 10-2001-0047094 (entitled " System and method for providing moving picture movie advertisement over the Internet, published on June 15, 2001) are available.

In order to solve the above problems, the present invention extracts index data representative of content from each of image content and advertisement content, and matches each extracted index data with data having a high degree of association, It is an object of the present invention to provide an advertisement matching system and method for providing contents.

It is another object of the present invention to provide a method and apparatus for viewing a video content by allowing a viewer who views the video content to perceive the ad content as if it is a part of the video content and minimizing the rejection of the ad content, So as to minimize the interference of viewers' rights of viewers, thereby enhancing the advertising effect.

According to another aspect of the present invention, there is provided an advertisement matching system for providing an advertisement having relevance to a video content, the system including an index item to associate an image content with an advertisement content provided by an advertisement provider, A matching condition setting unit for setting a matching condition including at least one of a type and a category; An index data extracting unit for selecting the image content and the advertisement content matching the matching condition set by the matching condition setting unit and extracting the index data for the selected image content and the advertisement content, respectively; And a matching unit for matching between the index data of the image content extracted through the index data extracting unit and the index data of the advertisement content, wherein the index data includes at least two of keyword data, sound data, Wherein the index data extracting unit includes a keyword extracting unit for extracting keyword data that can represent each content with respect to the image content and the advertisement content, a keyword extracting unit that can represent each content with respect to the image content and the advertisement content, A sound extracting unit for extracting sound data, and a color extracting unit for extracting the image content and an image color representative of each content with respect to the advertisement content, wherein the matching unit extracts, through the index data extracting unit The two or more index data of the extracted image content Complex also matched more than one index data of the advertising content if matching or non-identical items among the same items to recommend a matching high-frequency video content and advertising content.

delete

delete

The matching unit may include two or more index data among keyword data, sound data, or image color of the image content extracted through the index data extracting unit, and two or more index data among keyword data, sound data, or image color of the advertisement content And weighting can be given according to the items of the matched index data.

delete

Wherein the matching unit stores matching keywords having high relevance to the image content through a blog analysis on the image content in the database when the matching is based on the keyword data, And matches the extracted keyword with the keyword extracted from the advertisement contents.

delete

The sound extracting unit classifies and classifies the sound data into natural sounds, human sounds, artificial sounds, and the like, and a sound classifying unit for classifying and classifying emotional expressions of the main character from the sound data in the scene where the main character appears .

In this case, when the sound data is matched based on the sound data, the matching unit matches the sound data extracted from the image content and the sound data extracted from the advertisement content with the same category.

Wherein the color extracting unit comprises: a background image setting unit that sets an image to be a background from the image content or the advertisement content; And a color analyzer for analyzing a color distribution in which a background image set through the background image setting unit is displayed and extracting an average image color value.

The color analysis unit may extract an image color value for the advertisement content in a color symbolized by the advertisement brand.

The matching unit matches the image color value represented by the image content and the image color value indicated by the advertisement content when the image color value belongs to a certain threshold range.

Meanwhile, an advertisement matching method for providing an advertisement having relevance to an image content according to an embodiment of the present invention is an advertisement matching method in a system for providing an advertisement having relevance to an image content, Setting a matching condition including at least one of an index item, a content type, and a category to be associated with the image content and the advertisement content provided by the advertisement provider; The index data extracting unit of the system selecting the image content and the advertisement content matching the matching condition set in the matching condition setting unit and extracting the index data for the selected image content and the advertisement content respectively; And a matching unit in the system, matching the index data of the image content extracted through the index data extracting unit and the index data of the advertisement content, wherein the index data includes keyword data, sound data, Wherein the step of extracting the index data comprises extracting keyword data in which the index data extracting unit of the system can represent each content for the image content and the advertisement content, Extracting sound data that can be represented by the extracting unit with respect to the image content and the advertisement content, or extracting image data that can represent each content with respect to the image content and the advertisement content, Two or more of the extraction steps Wherein the matching step is a step in which the matching unit of the system matches two or more index data of the image content extracted through the index data extracting unit and two or more index data of the advertisement content with each other or the same item And recommends image contents and advertisement contents with high matching frequency.

delete

delete

In this case, in a case where the matching unit of the system matches the index data of the image content and the index data of the advertisement content, when matching is performed based on the keyword data, the matching unit performs a blog analysis And a keyword extracted from the image content and a keyword extracted from the advertisement content are matched with each other based on the database.

delete

At this time, when the matching unit of the system matches the index data of the image content and the index data of the advertisement content, when the matching is performed based on the sound data, The data and the sound data extracted from the advertisement contents are matched with each other in which the categories classified are the same.

Wherein the step of extracting the image color comprises the steps of: setting an image for a background or an object from the image content or the advertisement content by a color extracting unit of the index data extracting unit; And extracting an average image color value by analyzing a color distribution in which the image set through the color extracting unit appears.

At this time, in the step of matching the matching part of the system, the matching part matches when the image color value represented by the image content and the image color value indicated by the advertisement content belong to a certain critical range.

INDUSTRIAL APPLICABILITY According to the present invention, index data that can represent contents can be extracted from each of image contents and advertisement contents, and matching between index data having a high degree of association can be made so that even if the viewer sees the advertisement contents, Therefore, it is effective to raise the concentration of viewers and increase the advertisement effect.

Especially, maximizing the relevance by linking the concept of image contents and advertisement with keyword, sound, image color, etc., is effective in inducing a positive image of consumers for advertisement.

In addition, by excluding the viewing of the advertisement contents that are not related to the image contents, it is possible to minimize the rejection of the advertisement and minimize the interference of the viewer's right of view. In addition, since related advertisements can be provided to a viewer who knows the content of the video contents to some extent, it has an effect of causing not only the advertising effect but also the viewer's desire to purchase.

1 is a block diagram illustrating a network connection of an advertisement matching system for providing advertisements having relevance to image contents according to an exemplary embodiment of the present invention.
FIG. 2 is a block diagram of an advertisement matching system according to an embodiment of the present invention. Referring to FIG.
3 is a block diagram of a keyword extracting unit applied to an advertisement matching system according to an embodiment of the present invention.
4 is a block diagram of a sound extracting unit applied to an advertisement matching system according to an embodiment of the present invention.
5 is a diagram illustrating a configuration of a color extracting unit applied to an advertisement matching system according to an embodiment of the present invention.
FIG. 6 is a flowchart illustrating an advertisement matching method for providing advertisements having relevance to image contents according to an exemplary embodiment of the present invention.
7 is a flowchart illustrating a method for extracting and matching keyword data in the method of FIG.
8 is a flowchart illustrating a method for extracting and matching sound data in the method of FIG.
Figure 9 is a flow chart illustrating a method for extracting and matching image colors in the method of Figure 6;

It is noted that the technical terms used herein are used only to describe specific embodiments and are not intended to limit the invention. Also, the technical terms used herein should be interpreted in a sense that is generally understood by those skilled in the art to which the present invention belongs, unless otherwise defined in this specification, and it should be understood that an overly comprehensive It should not be construed as a meaning or an overly reduced meaning.

Also, the singular forms "as used herein include plural referents unless the context clearly dictates otherwise. In the present application, the term "comprising" or "comprising" or the like should not be construed as necessarily including the various elements or steps described in the specification, Or may be further comprised of additional components or steps.

Hereinafter, the present embodiment will be described in detail with reference to the accompanying drawings.

FIG. 1 is a diagram illustrating a network connection of an advertisement matching system for providing advertisements having relevance to image contents according to an exemplary embodiment of the present invention. FIG. 2 is a block diagram illustrating the configuration of an advertisement matching system according to an exemplary embodiment of the present invention. FIG.

Referring first to FIG. 1, an advertisement matching system 300 according to an embodiment of the present invention is associated with a video provider server 100 and an advertisement provider server 200.

The video provider server 100 and the advertisement provider server 200 may also include terminals of a non-server computer, a PC, and the like. The smartphone may connect to the advertisement matching system 300 to transmit image contents or advertisement contents to be matched, and may transmit a setting input of a user necessary for matching, and the advertisement matching system 300 ) According to the matching result.

The video provider server 100 includes a movie company, a cable broadcasting company, a terrestrial / terrestrial broadcaster, a VOD provider, an IPTV provider, a mobile TV broadcasting provider, a mobile Internet provider, a streaming broadcasting provider and a video content provider Stores and manages video content provided by a vendor. Therefore, the image provider server 100 can be classified according to each company. In this case, the advertisement matching system 300 can access each image provider server 100 and acquire corresponding image contents. The video contents include various kinds of movie contents, cable broadcasting programs, terrestrial / terrestrial broadcasting programs, Internet videos, VOD contents (including both downloading method and streaming method), and IPTV contents.

The advertisement provider server 200 includes a plurality of advertisement brands (advertisement brands A and B), and stores and manages advertisement contents provided by the respective advertisement brands. Each of the advertisement brands can request a service by subscribing to a service provided by the advertisement matching system 300 and inputting a condition for matching the advertisement. Conditions for matching the advertisement include a content type, a category, or an index data item, which will be described later in detail in FIG.

The advertisement matching system 300 receives image contents and advertisement contents from the image provider server 100 and the advertisement provider server 200, respectively, and can symbolically represent the contents with respect to the provided image contents and advertisement contents Extracts index data such as a keyword, a sound, and an image color. Then, the advertisement matching system 300 matches the image content having high relevance between the extracted index data and the advertisement content.

In addition, the advertisement matching system 300 can recommend a content having a high matching frequency.

Accordingly, when the advertisement content is inserted and provided before, after, or during the reproduction of the image content, the advertisement matching system 300 selectively provides the advertisement content highly related to the image content, so that even if the viewer views the advertisement content, It is possible to recognize as if it is a part of the content, and to provide the advertisement contents closely related to the image contents, above all, it is possible to increase the concentration of the viewer on the advertisement viewing and minimize the interference of the viewer's viewing right.

Specifically, the advertisement matching system 300 for implementing this is as shown in FIG.

The advertisement matching system 300 includes a matching condition setting unit 310, an advertisement provider setting information DB 320, an image content classification unit 330, an image content DB 320, an index data extracting unit 350, (360), and an index data DB (370).

The matching condition setting unit 310 receives and sets matching conditions such as a type of content, a category, and an index data item to be associated with the image content from the advertisement provider server 200 or the advertisement provider terminal. The content type and category include types of video contents to be matched, that is, video contents, VOD contents, IPTV contents, more specifically, drama, movies, documentary, music, education, war, love and the like. The index data item includes an item as to which index data of the keyword, sound, or image color is to be matched. This allows advertisers to reflect the opinions of their ad provider by setting a minimum matching condition to match the concept of their brand.

The advertisement provider setting information DB 320 stores the subscription information of the advertisement provider subscribed to the service of the present invention and stores and manages the setting information inputted through the matching condition setting unit 310 for each advertisement provider.

In addition, the database of the advertisement matching system 300 may store subscription information for a video provider subscribing to the service of the present invention and providing image contents.

In addition, the image content classification unit 330 may collect the image content from the image provider server 100 subscribed to the service, classify the image content into a form suitable for the service of the present invention, and store the image content in the image content DB 340.

For example, the image content classifying unit 330 may classify the category based on the attribute of the content such as a commonly used drama, movie, document, music, or education, 350 based on the index data.

The index data extracting unit 350 selects a video content list matching the matching condition set by the matching condition setting unit 310 when a matching request is received from the advertisement provider server 200 or the terminal of the advertisement provider, And extracts index data for each image content and advertisement contents provided by the advertisement provider in the image content list. The extracted index data and the corresponding contents are stored in the index data DB 370.

The index data means data that can represent each content symbolically, and is not limited to the form of text or image. In the present embodiment, at least one of keyword data, sound data, and image color is included.

The index data extracting unit 350 includes a keyword extracting unit 352, a sound extracting unit 354, and a color extracting unit 356.

The matching unit 360 matches the index data of the image content extracted through the index data extracting unit 350 with the index data of the advertisement content. According to the matching result, the matching unit 360 can recommend the image content having a high matching frequency with respect to the corresponding advertisement content.

At the time of matching, the matching unit 360 matches only the same item (kind) between the index data of the image contents and the index data of the advertisement contents. For example, the keyword data of the advertisement content matches the keyword data of the advertisement content, the sound data of the advertisement content matches only the sound data of the image content, and the image color of the advertisement content matches only the image color of the image content. By matching data of the same kind in this manner, the correlation between the index data can be made closer to anything, and the accuracy of matching can be increased. In addition, the matching unit 360 may combine one or more index data having the same item for each content. For example, the keyword data and the sound data of the image content extracted through the index data extracting unit 350 may be matched with each other between the keyword data of the advertisement content and the sound data. In this case, the matching unit 360 can assign weights according to the items of the matched index data. For example, a weight value can be given to a keyword data having high matching accuracy and a weight value can be given relatively low for an image color. Alternatively, the weight of the sound data may be given to the highest level according to the demand of the advertisement provider.

In addition, the matching unit 360 can also perform complex matching for cases in which the index data is not the same item. For example, the keyword data of the image content and the sound data of the advertisement content can be matched. It is also possible to match the sound data of the video content with the image color of the advertisement content.

On the other hand, in the specific matching method, the matching unit 360 may apply the matching method differently according to the item of the index data.

In the case of matching based on keyword data, the matching unit 360 performs a blog analysis on the image content in advance. By analyzing the posts and comments posted on the blog, the keywords associated with the image contents are stored in a database, thereby reflecting the keywords that are exposed on the Internet, thereby providing objective relevance for keyword matching. At this time, the rankings of blogs are managed according to categories, and keywords can be extracted from the power blogs in particular, so that weights can be given according to the ranking of each blog. Therefore, in the keyword matching, it is also possible to assign a weight to the association degree and to calculate the matching frequency according to the weight.

Here, blog analysis is just one example, and the analysis result of social network service (SNS) can be used. Therefore, the blog analysis also includes a social network service analysis.

When matching is performed on the basis of the sound data, the matching unit 360 can match the sound data extracted from the image content and the sound data extracted from the advertisement content to each other with the same category. For example, in the case of a movie in which the video content is a pilot, the main sound data extracted from the video content will be a flight related sound, and the flight related sound will be recorded in the index data extracting unit 350, Category, and matches the sound data of the same advertisement content as the classified category.

When matching is performed based on the image color, the matching unit 360 compares the image color value of the image content extracted by the index data extracting unit 350 with the image color value of the advertisement content, If they belong to the range, they can be matched with each other so that they are related to each other.

3 is a block diagram of a keyword extracting unit applied to an advertisement matching system according to an embodiment of the present invention.

The keyword extracting unit 352 according to the embodiment of the present invention includes a metabolism extracting unit 3521, a climax extracting unit 3522, a scene extracting unit 3523, and a keyword analyzing unit 3524.

The metabolism extracting unit 3521 extracts metabolism information from the image contents or the advertisement contents, respectively. The metabolism information may be text information or caption data included in the image content, and if there is no caption data, the metabolism information may be obtained through speech recognition included in the image.

delete

The scene extracting unit 3523 extracts a scene from the image content or the advertisement content. The scene extraction may be performed on the whole of the video content or the advertisement content, or may be limited to the climax section, or may be performed by limiting the duration of the specific reproduction time and the main character. Specifically, the scene extraction method can be applied to a scene extraction algorithm that is conventionally applied in the technical field. For example, an I-frame can be extracted from each content.

The keyword analyzing unit 3524 can extract a keyword having a high appearance frequency from the image content or the advertisement content based on the metabolism information extracted through the metabolism extracting unit 3521. [ The keyword analyzing unit 3524 analyzes the scene extracted by the scene extracting unit 3523 and extracts a keyword included in the scene in the scene based on the scene in which the main character appears or the radius in which the main character acts have. The extracted keywords are stored in the index data DB (370 in FIG. 2) for each image content or advertisement content, and may include cumulative data based on the frequency of occurrence, number of times, and the like.

4 is a block diagram of a sound extracting unit applied to an advertisement matching system according to an embodiment of the present invention.

The sound extracting unit 354 according to the embodiment of the present invention includes a sound separating unit 3541, a climax extracting unit 3542, a scene extracting unit 3543, a sound analyzing unit 3544, and a sound classifying unit 3545 do.

The sound separation unit 3541 separates only the sound data from the movie content or the advertisement content. The sound data may include human voice, or may include natural sounds, artificial sounds that occur artificially, and the like. If necessary, the sound separation unit 3541 can separate and apply human voice, natural sound, artificial sound, and the like.

The scene extracting unit 3543 extracts scenes from the image contents or the advertisement contents.

delete

Accordingly, when analyzing in the sound analysis unit 3544, the sound data separated through the sound separation unit 3541 is analyzed to extract the effective sound data, or the scene extracted through the scene extraction unit 3543 is analyzed, And the sound data in the scene based on the radius in which the main character is acting can be extracted.

The sound classification unit 3545 extracts, from the sound data extracted through the sound analysis unit 3544, natural sounds, human sounds, artificial sounds, airplanes, vehicle driving sounds, wind sounds, animal sounds, Such as laughter, crying, surprise, based on the baby's crying, or the expression of a person's emotions.

5 is a diagram illustrating a configuration of a color extracting unit applied to an advertisement matching system according to an embodiment of the present invention.

The color extraction unit 356 according to the embodiment of the present invention includes a background image setting unit 3561 and a color analysis unit 3562 as shown in FIG.

delete

The background image setting unit 3561 may extract an image of a specific object of the advertisement brand as well as a background image. Here, an object is defined as not only a thing in a dictionary meaning but also a person, an animal, a plant, a thumbnail mark, and the like. For example, it is possible to extract an image related to a mark representing a car or an automobile brand among objects appearing in the image content or the advertisement content, and to match the color of the extracted image.

The color analyzer 3562 extracts an average image color by analyzing a color (color) distribution in which the set (background) image appears in the background image setting unit 3561. The color distribution can use the histogram algorithm method or the average color can be determined by extracting representative feature vectors using a VQ (Vector Quantization) algorithm. In particular, if there is a representative mark in the advertisement brand, image color can be extracted based on the mark, or if there is a symbolic color in the advertisement brand such as airplane in blue color and spicy food in red color, can do.

Hereinafter, an advertisement matching method for providing an advertisement having relevance to image contents according to an exemplary embodiment of the present invention based on the system configuration described above will be described with reference to FIG.

First, in step S100, the matching condition setting unit of the advertisement matching system according to the embodiment of the present invention receives and sets matching conditions that are basically necessary to associate with the image contents from the advertisement provider server or the advertisement provider terminal. The matching condition includes a type of content, a category, an index item, and the like. The type of content and category indicate the content type and category for the image content to be matched, and the index item is the type of index data such as keyword, sound, and image color.

In step S110, the matching condition setting unit of the advertisement matching system stores and sets the inputted matching condition in the advertisement provider setting information DB. Before setting up, it is possible to request the ad provider to subscribe to the service and input and save the subscription information.

In step S120, when an index data extracting unit of the advertisement matching system receives a matching request from a terminal of the advertisement provider server or the advertisement provider, the image content matching the matching condition set in the matching condition setting unit can be searched and selected. The inquiry can be performed by accessing each video provider server or based on the data constructed in the video content DB in the advertisement matching system. At this time, the advertisement matching system can collect image contents from each image provider server, classify the image contents into a form suitable for the service of the present invention, and build the image contents DB. The entire contents of the image contents can be stored at the time of construction, or only the metadata of the image contents can be stored and the index data can be extracted, and when the image contents are needed, they can be obtained by connecting to the server.

In step S130, the index data extracting unit of the advertisement matching system extracts each image content from the selected image content list and index data representative of the advertisement content provided by the advertisement provider, respectively.

In step S140, the matching unit of the advertisement matching system matches the index data of the image contents extracted through the index data extracting unit and the index data of the advertisement contents. Various matching methods can be applied according to the items of the index data at the time of matching.

For example, when matching is performed based on keyword data, the matching unit performs blog analysis on the image content in advance to convert the fitness keywords having high relevance to the image content into a database, and extracts the fitness keywords extracted from the image content based on the database- The keyword and the keyword extracted from the advertisement contents can be objectively matched. At this time, the matching frequency can be calculated by assigning weights according to the ranking of the blog.

As another example, when matching is performed based on sound data, it is possible to match the sound data extracted from the matching image content and the sound data extracted from the advertisement content having the same category.

As another example, when matching based on color, the matching unit compares the image color value of the image content with the image color value of the advertisement content. If the difference between the two image color values belongs to a certain threshold range, .

Finally, in step S150, the matching unit of the advertisement matching system can recommend image contents having high relevance to arbitrary advertisement contents. Recommendation is made to recommend a playback segment associated with the video content, and the advertisement content may be inserted into a playback segment closely related to the video content.

Next, a method for matching to extract the index data will be described with reference to FIGS. 7 to 9. FIG.

7 is a flowchart for explaining a method for extracting and matching keyword data. For reference, the method of FIG. 7 is performed in the keyword extracting unit of the index data extracting unit in the advertisement matching system.

In step S200, the keyword extraction unit extracts the metabolism information from the image content or the advertisement content, respectively.

delete

In step S220, the keyword extracting unit extracts a scene from the video content or the advertisement content.

In the following steps S230 and S240, the keyword extracting unit can extract keywords having high validity by analyzing based on the metabolism information, climax section information, and scene information extracted in steps S200 to S220. In other words, by analyzing the metabolism information basically, it is possible to extract a keyword having a high appearance frequency. Alternatively, the keyword extracting unit can extract a keyword included in the metabolism in a scene in which the main character appears through scene analysis.

This process is possible only when the scene information is synchronized with the metabolic information.

8 is a flowchart for explaining a method for extracting and matching sound data. The method of FIG. 8 is performed in the sound extracting unit of the index data extracting unit in the advertisement matching system.

In step S300, the sound extracting unit separates only the sound data from the video content or the advertisement content, respectively.

delete

In step S320, the sound extracting unit extracts a scene from the image content or the advertisement content.

delete

In steps S330 and S340, the sound extracting unit analyzes the sound data corresponding to the climax section in the sound data separated from the image content and the advertisement content through S310 or S320, extracts the valid sound data, or analyzes the extracted scene And extracts sound data in a scene where the main character appears.

Through this process, the matching unit matches the sound data of the image contents with the sound data of the advertisement contents. Alternatively, the keyword data of the image content and the sound data of the advertisement content may be matched.

Next, a method of extracting and matching the image colors of FIG. 9 will be described.

In step S400, the color extracting unit of the index data extracting unit may extract an image of an object appearing in the image content or the advertisement content. Alternatively, the playback time of the content may be limited, such that a portion of the content of the climax section may be set as the background image.

In the next step S410, the color extraction section analyzes the color distribution with respect to the background image in the set range.

In step S420, an average color value may be extracted as a result of the color extraction unit analysis. In the case of an advertisement brand in which an advertisement content symbolically has a specific color at the time of extraction, it can be set to an image color without setting a background image.

The color values extracted through this process can be used for matching based on the degree of color similarity in the matching unit.

The foregoing description is merely illustrative of the present invention, and various modifications may be made by those skilled in the art without departing from the spirit of the present invention. Accordingly, the embodiments disclosed in the specification of the present invention are not intended to limit the present invention. The scope of the present invention should be construed according to the following claims, and all the techniques within the scope of the claims should be construed as being included in the scope of the present invention.

100: Video provider server 200: Ad provider server
300: matching system 310: matching condition setting unit
320: Ad provider setting information DB 330:
340: image content DB 350: index data extracting unit
360: matching unit 370: index data DB
356: color extracting unit 3521: metabolism extracting unit
3522: climax extracting unit 3523: scene extracting unit
3524: Keyword analysis section 3541: Sound separation section
3542: climax extracting unit 3543: scene extracting unit
3544: sound analysis section 3545: sound classification section
3561: Background image setting unit 3562: Color analysis unit

Claims (20)

delete delete delete A matching condition setting unit that sets a matching condition including at least one of an index item, a type of content, and a category to be associated with the image content and the advertisement content provided by the advertisement provider;
An index data extracting unit for selecting the image content and the advertisement content matching the matching condition set by the matching condition setting unit and extracting the index data for the selected image content and the advertisement content, respectively; And
And a matching unit for matching between the index data of the image content extracted through the index data extracting unit and the index data of the advertisement content,
Wherein the index data includes at least two of keyword data, sound data, and image color,
The index data extracting unit may include a keyword extracting unit for extracting keyword data that can represent each content with respect to the image content and the advertisement content, a sound data representative of each content with respect to the image content and the advertisement content, And a color extraction unit for extracting an image color representative of each content with respect to the image content and the advertisement content,
The matching unit may match two or more index data of the image content extracted through the index data extracting unit and two or more index data of the advertisement content with each other in the same item or in a case where the same item is not the same item, Recommend content and ad content,
When matching the same items,
The matching unit may match two or more index data among keyword data, sound data, or image color of the image content extracted through the index data extracting unit and two or more index data among keyword data, sound data, or image color of the advertisement content And assigning weights according to items of matched index data. The advertisement matching system according to claim 1,
delete A matching condition setting unit that sets a matching condition including at least one of an index item, a type of content, and a category to be associated with the image content and the advertisement content provided by the advertisement provider;
An index data extracting unit for selecting the image content and the advertisement content matching the matching condition set by the matching condition setting unit and extracting the index data for the selected image content and the advertisement content, respectively; And
And a matching unit for matching between the index data of the image content extracted through the index data extracting unit and the index data of the advertisement content,
Wherein the index data includes at least two of keyword data, sound data, and image color,
The index data extracting unit may include a keyword extracting unit for extracting keyword data that can represent each content with respect to the image content and the advertisement content, a sound data representative of each content with respect to the image content and the advertisement content, And a color extraction unit for extracting an image color representative of each content with respect to the image content and the advertisement content,
The matching unit may match two or more index data of the image content extracted through the index data extracting unit and two or more index data of the advertisement content with each other in the same item or in a case where the same item is not the same item, Recommend content and ad content,
In case of matching based on the keyword data,
Wherein the matching unit is configured to convert the adaptation keywords having high relevance to the image content into a database through blog analysis on the image content, and generate a keyword extracted from the image content based on the adaptation keyword, Wherein the keyword is matched with the advertisement content.
delete A matching condition setting unit that sets a matching condition including at least one of an index item, a type of content, and a category to be associated with the image content and the advertisement content provided by the advertisement provider;
An index data extracting unit for selecting the image content and the advertisement content matching the matching condition set by the matching condition setting unit and extracting the index data for the selected image content and the advertisement content, respectively; And
And a matching unit for matching between the index data of the image content extracted through the index data extracting unit and the index data of the advertisement content,
Wherein the index data includes at least two of keyword data, sound data, and image color,
The index data extracting unit may include a keyword extracting unit for extracting keyword data that can represent each content with respect to the image content and the advertisement content, a sound data representative of each content with respect to the image content and the advertisement content, And a color extraction unit for extracting an image color representative of each content with respect to the image content and the advertisement content,
The matching unit may match two or more index data of the image content extracted through the index data extracting unit and two or more index data of the advertisement content with each other in the same item or in a case where the same item is not the same item, We recommend content and ad content,
The sound extracting unit further includes a sound classifying unit for classifying and classifying the sound data into natural sounds, human sounds, artificial sounds, and the like, and classifying and classifying emotional expressions of the main character from sound data in a scene in which the main character appears And the advertisement content is related to the video content.
A matching condition setting unit that sets a matching condition including at least one of an index item, a type of content, and a category to be associated with the image content and the advertisement content provided by the advertisement provider;
An index data extracting unit for selecting the image content and the advertisement content matching the matching condition set by the matching condition setting unit and extracting the index data for the selected image content and the advertisement content, respectively; And
And a matching unit for matching between the index data of the image content extracted through the index data extracting unit and the index data of the advertisement content,
Wherein the index data includes at least two of keyword data, sound data, and image color,
The index data extracting unit may include a keyword extracting unit for extracting keyword data that can represent each content with respect to the image content and the advertisement content, a sound data representative of each content with respect to the image content and the advertisement content, And a color extraction unit for extracting an image color representative of each content with respect to the image content and the advertisement content,
The matching unit may match two or more index data of the image content extracted through the index data extracting unit and two or more index data of the advertisement content with each other in the same item or in a case where the same item is not the same item, Recommend content and ad content,
In case of matching based on the sound data,
Wherein the matching unit matches the sound data extracted from the image content and the sound data extracted from the advertisement content with each other in which the classified categories are the same.
A matching condition setting unit that sets a matching condition including at least one of an index item, a type of content, and a category to be associated with the image content and the advertisement content provided by the advertisement provider;
An index data extracting unit for selecting the image content and the advertisement content matching the matching condition set by the matching condition setting unit and extracting the index data for the selected image content and the advertisement content, respectively; And
And a matching unit for matching between the index data of the image content extracted through the index data extracting unit and the index data of the advertisement content,
Wherein the index data includes at least two of keyword data, sound data, and image color,
The index data extracting unit may include a keyword extracting unit for extracting keyword data that can represent each content with respect to the image content and the advertisement content, a sound data representative of each content with respect to the image content and the advertisement content, And a color extraction unit for extracting an image color representative of each content with respect to the image content and the advertisement content,
The matching unit may match two or more index data of the image content extracted through the index data extracting unit and two or more index data of the advertisement content with each other in the same item or in a case where the same item is not the same item, We recommend content and ad content,
Wherein the color extracting unit comprises: a background image setting unit for extracting an image of an object including an animal, an object, and a thumbnail mark appearing in the image content or the advertisement content; And a color analyzer for analyzing a color distribution in which the image extracted through the background image setting unit is analyzed and extracting an average image color value.
11. The method of claim 10,
The color analyzer
And the image color of the advertisement content can be extracted in a color symbolized by the advertisement brand.
11. The method of claim 10,
The matching unit
Wherein the matching is performed when an image color value represented by the image content and an image color value represented by the advertisement content belong to a certain threshold range.
delete delete delete A method for matching advertisements in a system for providing advertisements having relevance to video content,
Setting a matching condition including at least one of an index item, a content type, and a category to be associated with the image content and the advertisement content provided by the advertisement provider;
The index data extracting unit of the system selecting the image content and the advertisement content matching the matching condition set in the matching condition setting unit and extracting the index data for the selected image content and the advertisement content respectively; And
Wherein the matching unit of the system includes matching between the index data of the image content extracted through the index data extracting unit and the index data of the advertisement content,
Wherein the index data includes at least two of keyword data, sound data, and image color,
Wherein the step of extracting the index data comprises the steps of: extracting an index data extracting unit of the system from the keyword data that can represent each content with respect to the image content and the advertisement content; Extracting sound data that can represent each piece of content with respect to the content or extracting an image color in which the index data extracting unit can represent each piece of the content with respect to the image content and the advertisement content and,
The matching step may be a step in which the matching unit of the system matches two or more index data of the image content extracted through the index data extracting unit and two or more index data of the advertisement content with each other, Matching and recommending video contents and advertisement contents with high matching frequency,
Wherein when the matching unit matches the keyword data based on the keyword data in the matching step, the matching unit makes a database of relevance keywords having high relevance to the image content through a blog analysis on the image content, And matching the keyword extracted from the image content with the keyword extracted from the advertisement content.
delete A method for matching advertisements in a system for providing advertisements having relevance to video content,
Setting a matching condition including at least one of an index item, a content type, and a category to be associated with the image content and the advertisement content provided by the advertisement provider;
The index data extracting unit of the system selecting the image content and the advertisement content matching the matching condition set in the matching condition setting unit and extracting the index data for the selected image content and the advertisement content respectively; And
Wherein the matching unit of the system includes matching between the index data of the image content extracted through the index data extracting unit and the index data of the advertisement content,
Wherein the index data includes at least two of keyword data, sound data, and image color,
Wherein the step of extracting the index data comprises the steps of: extracting an index data extracting unit of the system from the keyword data that can represent each content with respect to the image content and the advertisement content; Extracting sound data that can represent each piece of content with respect to the content or extracting an image color in which the index data extracting unit can represent each piece of the content with respect to the image content and the advertisement content and,
The matching step may be a step in which the matching unit of the system matches two or more index data of the image content extracted through the index data extracting unit and two or more index data of the advertisement content with each other, Matching and recommending video contents and advertisement contents with high matching frequency,
Wherein the matching unit matches the sound data extracted from the image content and the sound data extracted from the advertisement content with the same category in which the sound data is classified based on the sound data in the matching step, The method comprising the steps of:
A method for matching advertisements in a system for providing advertisements having relevance to video content,
Setting a matching condition including at least one of an index item, a content type, and a category to be associated with the image content and the advertisement content provided by the advertisement provider;
The index data extracting unit of the system selecting the image content and the advertisement content matching the matching condition set in the matching condition setting unit and extracting the index data for the selected image content and the advertisement content respectively; And
Wherein the matching unit of the system includes matching between the index data of the image content extracted through the index data extracting unit and the index data of the advertisement content,
Wherein the index data includes at least two of keyword data, sound data, and image color,
Wherein the step of extracting the index data comprises the steps of: extracting an index data extracting unit of the system from the keyword data that can represent each content with respect to the image content and the advertisement content; Extracting sound data that can represent each piece of content with respect to the content or extracting an image color in which the index data extracting unit can represent each piece of the content with respect to the image content and the advertisement content and,
The matching step may be a step in which the matching unit of the system matches two or more index data of the image content extracted through the index data extracting unit and two or more index data of the advertisement content with each other, Matching and recommending video contents and advertisement contents with high matching frequency,
Wherein the extracting of the image color comprises:
Extracting an image of an object including an animal, a plant, and a thumbnail mark appearing in the image content or the advertisement content by a color extracting unit of the index data extracting unit; And extracting an average image color value by analyzing a color distribution in which the image extracted through the color extracting unit is displayed.
20. The method of claim 19,
In the matching step of the matching part of the system,
Wherein the matching unit matches the image color value represented by the image content and the image color value indicated by the advertisement content when the matching unit falls within a certain threshold range.
KR1020150108831A 2015-07-31 2015-07-31 System and method for matching advertisement for providing advertisement associated with video contents KR101674310B1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
KR1020150108831A KR101674310B1 (en) 2015-07-31 2015-07-31 System and method for matching advertisement for providing advertisement associated with video contents
PCT/KR2015/008604 WO2017022880A1 (en) 2015-07-31 2015-08-18 Advertisement matching system and method for providing advertisements having relevance to video contents

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020150108831A KR101674310B1 (en) 2015-07-31 2015-07-31 System and method for matching advertisement for providing advertisement associated with video contents

Publications (1)

Publication Number Publication Date
KR101674310B1 true KR101674310B1 (en) 2016-11-22

Family

ID=57540313

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020150108831A KR101674310B1 (en) 2015-07-31 2015-07-31 System and method for matching advertisement for providing advertisement associated with video contents

Country Status (2)

Country Link
KR (1) KR101674310B1 (en)
WO (1) WO2017022880A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20210026643A (en) * 2019-08-30 2021-03-10 (주) 달라라네트워크 Method for providing advertisement based on behavior patterns of character in interactive contents and the service server using the same
KR20220157033A (en) * 2021-05-20 2022-11-29 주식회사 한글과컴퓨터 Video streaming service server for displaying advertisement information related to video and operating method thereof

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20110016251A (en) * 2009-08-11 2011-02-17 김황철 Internet advertising method by context matching image-text link
JP5211066B2 (en) * 2006-12-11 2013-06-12 ソー カン，ミン Method and system for providing advertisement content-linked advertising content being played

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100957007B1 (en) * 2008-03-21 2010-05-13 주식회사 다음커뮤니케이션 System and method for providing customized advertisement in real-time personal broadcasting
KR20150068531A (en) * 2013-12-11 2015-06-22 주식회사 쿠노소프트 Method and apparatus for providing advertisement using linkage live broadcast
JP5726359B1 (en) * 2014-09-05 2015-05-27 株式会社電通 Television program-related content providing system and providing method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5211066B2 (en) * 2006-12-11 2013-06-12 ソー カン，ミン Method and system for providing advertisement content-linked advertising content being played
KR20110016251A (en) * 2009-08-11 2011-02-17 김황철 Internet advertising method by context matching image-text link

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20210026643A (en) * 2019-08-30 2021-03-10 (주) 달라라네트워크 Method for providing advertisement based on behavior patterns of character in interactive contents and the service server using the same
KR102239991B1 (en) * 2019-08-30 2021-04-14 (주) 달라라네트워크 Method for providing advertisement based on behavior patterns of character in interactive contents and the service server using the same
KR20220157033A (en) * 2021-05-20 2022-11-29 주식회사 한글과컴퓨터 Video streaming service server for displaying advertisement information related to video and operating method thereof
KR102500735B1 (en) * 2021-05-20 2023-02-16 주식회사 한글과컴퓨터 Video streaming service server for displaying advertisement information related to video and operating method thereof

Also Published As

Publication number Publication date
WO2017022880A1 (en) 2017-02-09

Similar Documents

Publication Publication Date Title
US11301505B2 (en) Topic and time based media affinity estimation
CN107171932B (en) Picture style conversion method, device and system
US9471936B2 (en) Web identity to social media identity correlation
US8605958B2 (en) Method and apparatus for generating meta data of content
US9100701B2 (en) Enhanced video systems and methods
US8126763B2 (en) Automatic generation of trailers containing product placements
US20100217671A1 (en) Method and apparatus for extracting advertisement keywords in association with situations of video scenes
GB2503878A (en) Generating interstitial scripts for video content, based on metadata related to the video content
CA2858992A1 (en) Video recommendation based on video co-occurrence statistics
US20170041648A1 (en) System and method for supplemental content selection and delivery
US20170041649A1 (en) Supplemental content playback system
US20130311287A1 (en) Context-aware video platform systems and methods
KR20100116412A (en) Apparatus and method for providing advertisement information based on video scene
US20170041644A1 (en) Metadata delivery system for rendering supplementary content
CN112507163A (en) Duration prediction model training method, recommendation method, device, equipment and medium
KR101927965B1 (en) System and method for producing video including advertisement pictures
WO2016032399A1 (en) Selecting adaptive secondary content based on a profile of primary content
KR101674310B1 (en) System and method for matching advertisement for providing advertisement associated with video contents
WO2016206035A1 (en) Information recommendation method and user terminal
EP3270600A1 (en) System and method for supplemental content selection and delivery
US20150227970A1 (en) System and method for providing movie file embedded with advertisement movie
WO2017149447A1 (en) A system and method for providing real time media recommendations based on audio-visual analytics
De Michele et al. Viewer-tailored advertising for video on demand platforms
Yuksel et al. DeepCens: A deep learning‐based system for real‐time image and video censorship
US20240187667A1 (en) Affinity profile system and method

Legal Events

Date Code Title Description
GRNT Written decision to grant