TWI718355B - Device and method for triggering commercial of digital television program - Google Patents

Abstract

The invention provides a new type of method for triggering commercial of television, which is adapted to time-shifted IPTV. The method comprises: Analyzing program content. Capturing still images. Generating program part object. Downloading commercial information and generating commercial object. Correlating the program part object and the commercial object by Pearson product-moment correlation coefficient. Determining whether the correlation between the program and the commercial reaches a threshold, and if yes, put the corresponded commercial on the program without covering the program while the program part object is appearing. When the program part object disappears, stopping broadcasting commercial and playing the program normally. The invention enables the commercial to be broadcasted with the correlated program instantly so as to make products correspond to the person or object in the program, therefore, the commercial effectiveness can be improved.

Description

Device and method for triggering advertisement of digital television program

The present invention relates to a method for playing advertisements, and more particularly to a device and method for triggering advertisements of digital television programs.

With the advancement of science and technology and the improvement of multimedia application technology, the related technology of multimedia advertising is constantly developing. Nowadays, multimedia advertisements are presented in very diversified ways, such as TV advertisements, web advertisements, or outdoor electronic billboards, etc., which are common advertising means. Among them, TV advertisements are the bulk of multimedia advertising.

In the conventional technology, the design and production of programs and advertisements are different fields, and the two are developed separately and lack an overall consideration. TV commercials are played randomly, so they have low relevance to the program and it is difficult to gather users' attention to stimulate purchases. If an advertisement is placed in the program segment, the user may leave the TV or perform other activities, or even disgust the audience because of the interruption of the program, so the effect of the advertisement is limited. In addition, users often misidentify products when using placement marketing (for example, Korean dramas often include Korean car brands, but many viewers mistake them for Japanese car brands) or ignore products that are not of interest. Moreover, the titled advertisement is only a manifestation of a noun, and it does not correspond to the type and content of the program, nor does it lack detailed product descriptions.

In order to improve the effectiveness of advertisement playing based on the content of the program, the existing technology usually judges the timing of advertisement playing in a manual manner. For example, advertisers may choose to show advertisements related to sporting goods between sections of a ball game program. However, there are a large number of programs nowadays, and manual monitoring of programs to determine the timing of advertisements is very labor-intensive.

The invention provides a device for triggering advertisements of digital television programs, which includes a storage unit and a processor. The storage unit records multiple modules. The processor is coupled to the storage unit, and accesses and executes the modules stored in the storage unit. The modules include an image capture module, an object creation module, and an object comparison module. The image capturing module captures the first static image of the program content and the second static image of the advertisement content. The object creation module generates a program segment object based on the first static image, and generates an advertisement object based on the second static image. The object comparison module compares the relevance between the program segment object and the advertisement object, and based on the relevance decides to place the advertisement content on the program content when the program content is played.

The present invention provides a method for triggering an advertisement of a digital television program, which includes: capturing a first static image of the program content and a second static image of the advertisement content. A program segment object is generated based on the first static image, and an advertisement object is generated based on the second static image. Compare the relevance of the program segment object and the advertisement object. And based on the correlation, it is determined to place the advertisement content on the program content when the program content is played.

Based on the above, the present invention can create image objects of each program by analyzing programs, and analyze advertisements using the same technology to create image objects of each advertisement. Then, by comparing the shape and color attributes of each program and the advertisement image object, it is determined whether the program and the advertisement have a high degree of positive correlation. When the specific advertisement has a high correlation with the program, the present invention can place the advertisement in the program at the same time in a way that the content of the program is not shielded.

In order to make the above-mentioned features and advantages of the present invention more comprehensible, the following specific embodiments are described in detail in conjunction with the accompanying drawings.

The present invention provides a device and method for triggering advertisements on digital televisions. The method is a new type of advertising technology, and can be used to broadcast advertisements in video streams such as digital television or Internet Protocol Television (IPTV).

FIG. 1 illustrates a schematic diagram of an apparatus 100 for triggering an advertisement of a digital television program according to an embodiment of the present invention. The apparatus 100 may include, but is not limited to, a processor 110 and a storage unit 130.

The storage unit 130 can be any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash memory, traditional hard disk drive, solid state hard drive A disk (solid-state drive) or similar components or a combination of the above components are used to record multiple modules, including: an image capture module 131, an object creation module 133, and an object comparison module 135. The processor 110 is coupled to the storage unit 130, and can be a central processing unit (CPU), or other programmable general-purpose or special-purpose microprocessor (Microprocessor), digital signal processor (DSP), programmable Controller, application-specific integrated circuit (ASIC) or other similar components or a combination of the above components. In the embodiment of the present invention, the processing unit 110 is used to execute all the functions of the device 100 for triggering the advertisement of the digital TV, and can access and execute the modules recorded in the storage unit 130.

The image capturing module 131 can be used to capture static images of program content and static images of advertisement content. The object creation module 133 can generate a program segment object based on the static image of the program content, and generate an advertisement object based on the static image of the advertisement content. The object comparison module 135 can compare the relevance of the program segment object and the advertisement object, and based on the relevance, determine whether to place the advertisement content on the program content and not cover the program content when the program content is being played. Presented to viewers.

FIG. 2 shows a flowchart of a method 200 for triggering an advertisement of a digital television program according to an embodiment of the present invention. The method 200 may be implemented by the aforementioned apparatus 100. The implementation process is as follows.

Step S211, step S212, step S213, and step S214 describe the process of generating program segment objects.

In step S211, the device 100 may analyze the content of the program through a terminal device (for example, a television). The analysis method can be, for example, to analyze the program content data by pre-loading the video stream, as shown in FIG. 3. In the example of FIG. 3, the terminal device starts streaming program content from n1 seconds, and extracts the program content from n1+k to n2 seconds of streaming for analysis.

Returning to FIG. 2, in step S212, the image capturing module 131 captures one or more static images of the program content. The sampling interval may affect the accuracy of associating the program content with the advertisement content in the subsequent steps. The interval time can be set according to the needs of the user.

In step S213, the object creation module 133 generates a program segment object based on the static image of the program content. More specifically, the object creation module 133 may generate the program segment object based on at least one of the similarity of the objects in the static image, the weighted screen-to-body ratio, and the appearance time. The detailed implementation of step S213 is described in relevant paragraphs of FIG. 4 described later.

In step S214, the object creation module 133 creates program segment objects of one or more program segments.

Step S221, step S222, step S223, and step S224 describe the process of generating advertisement objects. The process of generating advertisement objects is very similar to the process of generating program paragraph objects (that is, the processes described in step S211, step S212, step S213, and step S214). The only difference between the two is that the advertisement object is generated based on the advertisement, while the program Paragraph objects are generated based on the content of the program. Based on the above, this specification does not repeat the process of generating advertisement objects.

After one or more program segment objects and one or more advertisement objects are generated, step S231 is entered.

In step S231, the object comparison module 135 compares the association between the one or more program segment objects and the one or more advertisement objects. The object comparison module 135 can generate the association based on at least one of the shape and color of the object in the image. More specifically, the object comparison module 135 can use the Pearson product-moment correlation coefficient to compare the shape and position correlation and the color correlation between the program segment object and the advertisement object.

…..公式1 假設將節目段落物件與廣告物件抽樣m個點，物件之顏色的皮爾遜積差相關係數r_rgb 可由下述的公式2表示：

…..公式2 其中的標籤數學意義為： r_shape ：形狀位置相關係數，-1≦ r_shape ≦+1； r_rgb ：顏色相關係數，-1≦ r_rgb ≦+1； X_i ：取樣位置i之X座標參數；

：取樣位置X座標參數平均值； Y_i ：取樣位置i之Y座標參數；

：取樣位置Y座標參數平均值； R_j ：取樣顏色j之R參數；

：取樣顏色R參數平均值； G_j ：取樣顏色j之G參數；

：取樣顏色G參數平均值； B_j ：取樣顏色j之B參數；

：取樣顏色B參數平均值。Assuming that the program segment object and the advertisement object are sampled at n points, the Pearson product difference correlation coefficient r _shape of the shape position of the object can be expressed by the following formula 1:

….. Formula 1 Assuming that the program segment object and the advertisement object are sampled at m points, the Pearson product difference correlation coefficient r _rgb of the color of the object can be expressed by the following formula 2:

….. The mathematical meaning of the label in formula 2 is: r _shape : shape position correlation coefficient, -1≦r _shape ≦+1; r _rgb : color correlation coefficient, -1≦r _rgb ≦+1; X _i : sampling position The X coordinate parameter of i;

: Average value of X-coordinate parameters of sampling position; Y _i : Y-coordinate parameters of sampling position i;

: The average value of the Y coordinate parameters of the sampling position; R _j : the R parameter of the sampling color j;

: Average value of R parameter of sampled color; G _j : G parameter of sampled color j;

: Average value of sampled color G parameters; B _j : B parameter of sampled color j;

: Average value of sampled color B parameters.

_{When the correlation coefficient r shape} and the correlation coefficient r _rgbr between a program segment object and an advertisement object are both positively correlated, it means that the object comparison module 135 can regard the combination of the program segment object and the advertisement object as Therefore, in step S232, it is further determined whether the positive correlation between the program segment object and the advertisement object is sufficient. If multiple sets of combinations are considered to be related, that is, when the correlation coefficient r _shape and the correlation coefficient r _rgbr are both positively correlated, the object comparison module 135 can select the product of _{r shape} and r _{rgbr that is the largest} And determine whether the combination reaches the relevance threshold in step S232.

In step S232, the object comparison module 135 determines whether the relevance between the program segment object and the advertisement object reaches a relevance threshold. The threshold value can be set to ensure that the associated advertisement has a certain degree of accuracy. If the threshold of relevance is reached, step S233 is entered. If the threshold of relevance is not reached, step S236 is entered.

In step S233, the object comparison module 135 presents the program and the advertisement at the same time in a way that the program content is not shielded, as shown in FIG. 5, please refer to FIG. 2 and FIG. 5 at the same time below. The object comparison module 135 determines that among the program content 510 played by the terminal device of FIG. 5, the most representative object is the object i. Therefore, the object comparison module 135 displays the advertisement content 530 related to the object i on the terminal device in a manner of unmasking the program content 510 while playing the program content 510.

In step S234, the object comparison module 135 determines whether the program segment related to the object i (ie: the m1 second to the m2 second in FIG. 5) is over, if the program segment has not ended, it can return to step S233 and continue Advertisement content 530 is played. If the program segment has ended, step S235 is entered.

In step S235, the object comparison module 135 stops playing the advertisement content 530.

In step S236, the flow of the method 200 ends. The terminal equipment will play the program in a general and conventional manner.

FIG. 4 further illustrates a flowchart of generating program segment objects (ie, step S213) according to an embodiment of the present invention.

In step S2131, the object creation module 133 determines that the screens of the program content are similar and continue to exceed a threshold value for a period of time. If it exceeds, the mechanism of step S2132 and step S2133 is performed.

…..公式3 其中i代表靜態影像中的物件的索引、N代表靜態影像被分割成的總畫面分格數、P_j 代表物件在第j個畫面分格的加權屏占比、並且W_j 第j個畫面分格的佔所有畫面分格的階層權重。物件建立模組133可基於形狀或顏色等參數而由靜態影像辨識出物件。舉例來說，假設一靜態影像被分割為2個畫面分格，其中第1個畫面分格佔據靜態影像較中心的位置，其階層權重為0.8，而第2個畫面分格佔據靜態影像較週邊的位置，其階層權重為0.2。此外，假設某人物A在第1個畫面分格佔據了50% 的畫面，在第二個畫面分格佔據了20% 的畫面，如此，則人物A在所述靜態影像的物件融合值Δ_A 可表示如公式4：

…..公式4In step S2132, the object creation module 133 formulates the frame of the static image of the program content and the level weight corresponding to each frame. Specifically, the object creation module 133 can divide a static image of the program content into N frame divisions and assign different hierarchical weights to each frame division, wherein the frame division closer to the center of the static image is allocated The higher the level weight assigned, the lower the level weight assigned to the screen grids closer to the four corners of the static image. Then, the object creation module 133 can calculate the object fusion value Δ in the static image based on the hierarchy weight. The object fusion value Δ represents the weighted screen-to-body ratio of the object in the screen division and the hierarchical weight of the screen division. The formula 3 of the object fusion value Δ is as follows:

….. Formula 3 where i represents the index of the object in the static image, N represents the total number of frame divisions into which the static image is divided, P _j represents the weighted screen ratio of the object in the j-th frame division, and W _j The hierarchical weight of the j-th frame in all frames. The object creation module 133 can identify the object from the static image based on parameters such as shape or color. For example, suppose a still image is divided into two frame divisions, where the first frame frame occupies the center of the static image, and its level weight is 0.8, and the second frame frame occupies the periphery of the static image The position of, its class weight is 0.2. In addition, suppose that a character A occupies 50% of the screen in the first frame and 20% of the screen in the second frame. In this case, the object fusion value Δ _{A of the character A in the static image} It can be expressed as formula 4:

….. Formula 4

…..公式5In step S2133, the object creation module 133 determines the appearance time of the object from one or more static images, and calculates and selects the best object according to the appearance time of the object. For example, if in the second paragraph of program content m1 to m2 sec, the article appeared i S _i seconds. The object creation module 133 can select a best object as the program segment object k representing the program segment (ie: m1 second to m2 second) based on the following formula 5, so as to determine the program segment (ie: From the m1 second to the m2 second), the advertisement of the item k of the program segment is played, as shown in FIG. 5.

….. Formula 5

Fig. 6 illustrates a schematic diagram of playing an advertisement according to an embodiment of the present invention. As shown in FIG. 6, the present invention can use the context of program creation to place complementary advertisements, thereby arousing the resonance of the audience. The program plot of FIG. 6 is a food feast, and the present invention can place food advertisements similar to the program content. Compared with the traditional video and audio stimulation, the advertisement of the present invention can be injected into the program context, and the natural hunger of the audience can be used to enhance the advertisement effect.

In summary, the present invention provides a device and method for triggering advertisements on digital televisions. The method is a new type of advertising technology, and can be used to broadcast advertisements in video streams such as digital TV or IPTV. The present invention can create the image objects of each program by analyzing the programs, and analyze the advertisements with the same technology to create the image objects of each advertisement. Then, by comparing the shape and color attributes of each program and the advertisement image object, it is determined whether the program and the advertisement have a high degree of positive correlation. When the specific advertisement has a high correlation with the program, the present invention can place the advertisement in the program at the same time in a way that the content of the program is not shielded. Advertisements are broadcast in real-time in association with the content of the program, so that the product appears to the person (for example: idol endorsement) or thing (for example: food causes hunger) corresponding to the content of the program. In this way, the advertisement is continuously watched in accordance with the program context, thereby creating a high degree of relevance and positive attraction, thereby improving the effectiveness of the advertisement.

The characteristics and effects of the present invention: the new-style IPTV advertising method provided by the embodiments of the present invention, when compared with other conventional technologies, has the following advantages: 1. The present invention provides a new-style IPTV advertisement, which makes use of viewers’ viewing Programs are mostly interesting content, and timely advertisements that complement the content can increase user interest and deepen advertising effects. Design advertisements to be broadcast in real time in association with the content of the program, so that the product appears to the person (such as: idol endorsement) or object (such as: food causes muscle hunger) corresponding to the program, creating a high degree of relevance and positive attraction, and improving advertising efficiency . 2. Traditional advertisements are played randomly and cannot complement the program context. The advertisement placed by the present invention is highly related to the content of the program, and can use the appeal of the idol or the plot guide to enhance the cohesion of the user on the advertisement product. 3. Traditional advertisements are broadcast in program segments, and users may leave the TV without effect. Even because advertisements block program viewing, users are disgusted. The invention designs programs and advertisements to be presented at the same time without covering the contents of the programs, so that users can continuously watch, integrate into the program and watch the advertisements at the same time, thereby enhancing the advertisement effect. 4. In addition to being applied to the IPTV field, the present invention can also be widely used in the program production and advertising design industries. Different from traditional programs and advertisements with low relevance in their respective design and production, the present invention can strengthen the horizontal industry chain, make programs and advertisements complement each other, and create new application categories.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention. Anyone with ordinary knowledge in the relevant technical field can make some changes and modifications without departing from the spirit and scope of the present invention. The scope of protection of the present invention shall be determined by the scope of the attached patent application.

100‧‧‧Device for triggering advertisements of digital TV programs 110‧‧‧Processor 130‧‧‧Storage unit 131‧‧‧Image capture module 133‧‧‧Object creation module 135‧‧‧Object comparison module 510 ‧‧‧Program content 530‧‧‧Advertising content S211, S212, S213, S214, S221, S222, S223, S224, S231, S232, S233, S234, S235, S236, S2131, S2132, S2133, and steps

FIG. 1 illustrates a schematic diagram of an apparatus for triggering advertisements of digital television programs according to an embodiment of the present invention. Fig. 2 shows a flow chart of a method for triggering an advertisement of a digital television program according to an embodiment of the present invention. Fig. 3 illustrates a schematic diagram of analyzing program content according to an embodiment of the present invention. Fig. 4 further illustrates a flow chart of generating program segment objects according to an embodiment of the present invention. FIG. 5 further illustrates a schematic diagram of playing an advertisement corresponding to a program segment object according to an embodiment of the present invention. Fig. 6 illustrates a schematic diagram of playing an advertisement according to an embodiment of the present invention.

100‧‧‧A device that triggers advertisements for digital TV programs

110‧‧‧Processor

130‧‧‧Storage Unit

131‧‧‧Image capture module

133‧‧‧Object creation module

135‧‧‧Object Comparison Module

Claims (6)

A device for triggering advertisements of digital television programs includes: a storage unit that records a plurality of modules; and a processor, coupled to the storage unit, and accesses and executes the modules stored in the storage unit, and the modules The group includes: an image capture module that captures a first static image of the program content and a second static image of the advertisement content; an object creation module that generates a program segment object based on the first static image, and based on the second static image Static images to generate advertising objects; and an object comparison module that compares the shape and color between the program segment object and the advertising object to generate shape correlation coefficients and color correlation coefficients, and based on the shape correlation coefficients and the The color correlation coefficients generate relevance, and based on the relevance, it is determined to place the advertising content on the program content when the program content is played, wherein the object creation module is based on the first static image and the first static image. 2. The appearance time of the object in the static image generates the program segment object and the advertisement object. The device according to claim 1, wherein the object creation module generates the program segment object and the advertisement object based on at least one of the similarity of the object and the weighted screen-to-body ratio. The device according to the first item of the scope of patent application, wherein the object comparison module generates the correlation based on a Pearson product-moment correlation coefficient. A method for triggering an advertisement of a digital television program includes: capturing a first static image of the program content and a second static image of the advertisement content; generating a program segment object based on the first static image, and based on the second static image Generating an advertising object includes: generating the program segment object and the advertising object based on the appearance time of the object in the first static image and the second static image; comparing the program segment object with the advertising object To generate a shape correlation coefficient and a color correlation coefficient between the shape and color; and generate a correlation based on the shape correlation coefficient and the color correlation coefficient, and then determine the placement of the program content when the program content is played based on the correlation. The advertising content is in the program content. The method described in item 4 of the scope of patent application further includes generating the program segment object and the advertising object based on at least one of the similarity of the object and the weighted screen-to-body ratio. The method described in item 4 of the scope of patent application further includes generating the correlation based on the Pearson product-moment correlation coefficient.

Images