TW201405462A - Digital signage push system based on crowd correlation determination - Google Patents

Abstract

A digital signage push system based on relationship between the crowds is disclosed, which is particularly limited to a case when multiple people watch the digital signage at the same time. At the time the digital signage displays advertising, the invention utilizes a camera module to capture images of viewing crowds in front of the digital signage, and performs image identification and determines the relationship between the crowds for the captured images of the crowds. The crowd relationship determination is performed according to respective identified gender and age information, and subjected to further analysis to determine information such as hands holding, hugging, wearing the same clothes style and the same color, and distances and positions among standing groups, such that advertising content suitable for the crowds to watch can be selected by an advertisement control module, thereby improving advertising marketing values and benefits.

Description

Electronic kanban push broadcasting system based on the relationship between the masses

This creation is about an electronic kanban push-and-cast system based on the relationship between the masses. It is a kind of advertisement that actively plays the appropriate audience by comprehensively analyzing the relationship between the gender, age and the masses of the audience when multiple people are watching at the same time. content.

Nowadays, in the public places where the MRT station, subway platform, department store and other crowds gather, multimedia display devices and electronic billboards are set up to broadcast advertisements and various kinds of information to consumers for reference, but the current advertising and information Only the unspecified viewers can play unrestrictedly, and it is impossible to select suitable advertisements for the characteristics of consumers, which reduces the effectiveness of advertising marketing. Therefore, the market is paying more and more attention to smart electronic kanban products. The camera on the electronic billboard captures information such as the image of the human face, and plays suitable advertisement content according to the viewer's gender, age, etc.; this practice has been disclosed in the patent "channel switching module" (certificate number: M342530) and "Integration of Advertising and Payment System" (Certificate No.: M372506); this practice will lead to some problems. If there is only one viewer in front of the kanban, the corresponding advertising content can be broadcasted through the gender and age information, but the electronic billboards are all Set up in a crowded place to increase the exposure of advertising and increase the sales of goods, so the board Many viewers have often appeared, the audience may contain both male and female and of all ages, using the above method that is not suitable for groups of broadcast advertising content.

In addition, many advertisements are set up for marketing purposes rather than individual but related groups. For example, KTV, Internet cafes, and bubble tea shops set up marketing for youth students; children's playgrounds, fast food restaurants, and toy store marketing customers. Groups are children and their parents; couples and couples are the main marketing objects of restaurants, MOTELs, and tour operators, using the methods of the above patents for this type of Types of advertising cannot improve the effective reach and the effectiveness of advertising.

It can be seen that there are still many shortcomings in the above-mentioned methods of use, which is not a good design, but needs to be improved.

In view of the shortcomings derived from the above-mentioned conventional methods, the inventor of the present invention has improved and innovated, and after years of painstaking research, he finally succeeded in researching and developing the electronic kanban push-and-cast system based on the relationship between the masses.

The electronic kanban push-and-cast system based on the relationship between the masses for achieving the above-mentioned creative purposes includes a camera module, an image recognition module, a relationship judgment module between the masses, and an advertisement control module; the main purpose is to install a camera on the electronic kanban. Using the camera module to capture the image of the crowd before capturing the electronic kanban, and playing the advertisement on the electronic kanban, the image of the captured mass image is judged by the relationship between the image and the mass; the image recognition module includes a face Detecting sub-modules, gender and age identification sub-modules and the relationship analysis sub-module between the masses, using the face detection sub-module to obtain the face position of the viewing mass, and obtaining the audience through the gender and age identification sub-module Gender and age information, and the inter-mass relationship analysis sub-module is used to detect whether there is any hand, hug, wear the same clothes style and color, the distance and position between the groups, etc. The module receives the result of the image recognition module, and analyzes the information by using the relationship between the identified gender, age information and the mass to obtain a partner. The results of the relationship between the father and son, the mother and the child, and the students; if it is judged that there is a certain relationship between the current audience, the advertisement corresponding to the group relationship is played; if there is no specific relationship, the gender and age information of the audience are counted, and the number of people is selected. The appropriate advertising of a certain gender and age group, in this way to enhance the value and effectiveness of advertising marketing; in addition, the results of the relationship between the masses can be transmitted to the back end for statistical analysis, breaking through the previous restrictions on single-person customer marketing analysis To provide advertisers with information on potential customers targeting the masses.

An electronic kanban push-and-cast system based on the relationship between the masses, comprising a camera module, the camera module comprising a camera and an image capturing software, wherein the image of the mass is viewed before the electronic kanban is photographed; the image recognition analysis host, the image The identification analysis host includes an image recognition module, a relationship determination module between the masses, and an advertisement control module. The image recognition module receives the image of the mass of the person captured by the camera module for identification, outputs the image of the viewing mass, and recognizes the image. The information of the relationship between the position of the face and the relationship between gender, age and mass, the inter-maritime relationship judgment module receives the identification result of the image recognition module and the position and gender, age and relationship between the masses of the image recognition module. And comprehensively judging whether there is a specific relationship between the viewing and the masses, the advertising control module analyzes the comprehensive information of the relationship between the gender, age and the mass of the viewer identified by the relationship judgment module between the masses, and plays on the electronic billboard for the masses to watch. Advertising content; the electronic signboard receives video from the image recognition and analysis host No, and play for the people viewing the ad content.

The camera is a video camera or a plurality of cameras. The video image captured by the camera is transmitted to the image recognition and analysis host through a coaxial cable and a network. The network transmission of the video image is wired or wireless. The image recognition analysis host is a front-end host, an embedded device or a remote server, and is configured to determine and output suitable advertisement content by image recognition and relationship between the people, and the image recognition module is a face detection sub-module. The gender and age identification sub-module and the mass relationship analysis sub-module are formed. The face detection sub-module obtains the face position of the viewing mass through image block analysis, and the gender and age identification sub-module system utilizes The facial feature analysis obtains the gender and age information of the audience. The inter-person relationship analysis sub-module detects whether the audience has hands, hugs, wears the same clothes style and color, and the distance and position between the groups to obtain the masses. The relationship between the masses, the interpersonal relationship judgment module is based on the identified gender, age information and analysis of the posture between the masses Behavior define their relationship as a relationship partner, father and son, mother and student, the relationship between the project The definition can be added or deleted. The electronic signage is LED marquee, LCD or outdoor large-scale TV. The analysis result of the inter-person relationship judgment module is transmitted to the back end for statistical analysis to provide advertisers to judge potential customers for the masses. News.

Please refer to FIG. 1 for the architecture diagram of the electronic kanban push-pull system based on the relationship between the masses, and a camera 11 mounted on an electronic kanban 13 for viewing the electronic kanban 13 by using the camera 11 The mass image transmits the captured mass image to the image recognition and analysis host 12 for image identification and relationship judgment between the masses. According to the comprehensive information of the relationship between gender, age and mass, the corresponding information is provided to suit the masses. Advertising content is played to enhance the value and effectiveness of advertising marketing.

Please refer to Figure 2 for the flow chart of the electronic kanban push-casting system based on the relationship between the masses. The camera module 21 (including the camera and the image-taking software) is used to view the images of the crowd and transmit the video images. To the image recognition and analysis host 12, the image recognition analysis host 12 can be a front-end host, an embedded device, or a remote server for image recognition, relationship determination between the masses, and output of suitable advertisement content. First, image recognition is performed. The face detection sub-module 23 in the module 22 obtains the number and position of the face of the crowd, and the face detection sub-module 23 is used to calculate the characteristic parameters of the Haar Feature, and then utilizes a kind of The machine learning algorithm (Adaboost) detects the position of the face, and uses the pre-trained Adaboost model to determine whether the square frame of different positions and sizes in the whole image is a human face, and finally obtains the face and the face without the face in the image. If the face is not detected, the advertisement control module 27 plays the content according to the scheduled or randomly selected advertisement content; otherwise, if the detected face number Greater than or equal one, based on the sex of the resulting image recognition analysis, analysis of the relationship between age and masses of integrated information appropriate advertising content; wherein, the number of people face is greater than one, the process of image recognition analysis via the The gender and age identification sub-module 24 of the image recognition module 22 obtains the sex and age information of the detected personnel, and then detects the relationship between the persons through the inter-personal relationship analysis sub-module 25, the gender and age identification. The sub-module 24 recognizes the process of extracting facial features by using a local binary pattern by eye detection and face normalization, and adopts a support vector machine (Support Vector Machine). The data collected in advance is generated by the SVM classification model, and the gender is selected by the model: male, female, and age groups: infant, adolescent, adult, middle-aged and elderly; the inter-person relationship analysis sub-module 25 includes the following images. Identification and analysis technology: detect the presence or absence of hands, shoulders, hugs, wearing the same clothes style and color, the distance between the group standing and the status information of the position, the implementation process of holding hands, shoulders, hugs, etc. are as follows, first Obtain a foreground person image by using a Gaussian background model, and input the foreground image into a previously trained SVM classifier, the SVM score The device can distinguish four categories, namely, hand, shoulder, hug, and other categories; and the clothing style comparison uses a Scale-invariant feature transform (SIFT) to detect important feature points on clothes. Calculate the similarity degree by comparing the SIFT feature points on the clothes between people through the Euclidean distance method. If the similarity is greater than a certain threshold value, judge the clothes of the same style between the persons; in addition, the color comparison of the clothes uses the clothes of the statistical staff. The histogram distribution of the three primary colors (Red, Green, Blue), the result of the histogram comparison to get the same color clothes between the people; finally, the distance and position between the groups can be detected through the face generated by the face detection The central coordinates analyze the results; the inter-mass relationship judgment module 26 uses the identified gender, age information and relationship between the masses to analyze the message to obtain the partner (satisfying three conditions: (1) the gender identification result is 1 male and 1 female; (2) The age identification result is the same age range; (3) the person has hands, shoulders, hugs, etc.), father or son (satisfying the two conditions) (1) the age of the identification results were infants, adults or teenagers, the elderly and so on; (2) staff in hand, stand on another station or Positions close to the posture and behavior), students (satisfying the two conditions: (1) age identification results for teenagers; (2) personnel wearing the same clothes style and color) and other mass relationship results; If a certain relationship has been set, the advertisement content corresponding to the group relationship is selected by the advertisement control module 27 to be played on the electronic board (for example, the student group recommends KTV, the tutoring class, the internet cafe; the parent-child relationship marketing child park, the fast food restaurant, Toys; couples, couples choose restaurants, MOTEL, tourism industry to recommend, etc.); if it is judged that there is no relationship item or the number of faces detected is equal to one person, then directly view the gender and age information of people by statistics. Promote the advertising content of a certain number of genders and age groups (such as adult women recommend cosmetics, adult males recommend motorhomes, elderly people marketing health care products, etc.).

The electronic kanban push-casting system based on the relationship between the masses is used in addition to the electronic kanban service for product marketing. The comprehensive information of the relationship between the masses can also be transmitted to the back-end database for statistical analysis to provide advertisers. Accurately research the information of potential customers.

The detailed description of the present invention is intended to be illustrative of a preferred embodiment of the invention, and is not intended to limit the scope of the invention. The patent scope of this case.

To sum up, this case is not only innovative in terms of technical thinking, but also has many of the above-mentioned functions that are not in the traditional methods of the past. It has fully complied with the statutory invention patent requirements of novelty and progressiveness, and applied for it according to law. Approved this invention patent application, in order to invent invention, to the sense of virtue.

11‧‧‧ camera

12‧‧‧Image Identification Analysis Host

13‧‧‧Electronic board

21‧‧‧ camera module

22‧‧‧Image recognition module

23‧‧‧Face Detection Sub-module

24‧‧‧Gender and age identification sub-module

25‧‧‧Interpersonal relationship analysis sub-module

26‧‧‧The relationship judgment module between the masses

27‧‧‧Advertising Control Module

28‧‧‧Is the face detected?

29‧‧‧ Is it judged to be relevant?

30‧‧‧Electronic Kanban playback schedule or randomly selected advertising content

31‧‧‧Electronic kanban to broadcast advertising content of a certain age group

32‧‧‧Electronic Kanban Plays the advertising content corresponding to the group relationship

Please refer to the detailed description of this creation and its drawings for further understanding of the technical content of this creation and its purpose; the relevant drawings are: Figure 1 is a structural diagram of the electronic kanban push-casting system based on the relationship between the masses; Figure 2 is a flow chart of the electronic kanban push-casting system based on the relationship between the masses.

21‧‧‧ camera module

22‧‧‧Image recognition module

23‧‧‧Face Detection Sub-module

24‧‧‧Gender and age identification sub-module

25‧‧‧Interpersonal relationship analysis sub-module

26‧‧‧The relationship judgment module between the masses

27‧‧‧Advertising Control Module

28‧‧‧Is the face detected?

29‧‧‧ Is it judged to be relevant?

30‧‧‧Electronic Kanban playback schedule or randomly selected advertising content

31‧‧‧Electronic kanban to broadcast advertising content of a certain age group

32‧‧‧Electronic Kanban Plays the advertising content corresponding to the group relationship

Claims (12)

An electronic kanban push-and-cast system based on the relationship between the masses, comprising: a camera module, the camera module comprising a camera and an image capturing software, wherein the image of the mass is viewed before the electronic kanban is photographed; the image recognition analysis host, The image recognition and analysis host includes an image recognition module, a relationship determination module between the masses, and an advertisement control module. The image recognition module receives the image of the mass image captured by the camera module for identification, and outputs the viewed mass image. The information of the relationship between the location of the face and the relationship between gender, age and the mass is identified. The relationship judgment module between the people receives the image of the viewer and the relationship between the face image and the gender, age and mass of the image recognition module. Identifying the results and comprehensively judging whether there is a specific relationship between the viewing and the masses, the advertising control module analyzes the comprehensive information of the relationship between the gender, age and the mass of the audience identified by the relationship judgment module between the masses, and plays the suitable mass on the electronic billboard. The advertisement content viewed; the electronic signboard receives the image recognition analysis host Hearing the signal, and plays for the masses to watch the ad content. The electronic kanban push-and-cast system based on the judgment of the relationship between the masses, as described in claim 1, wherein the camera is a one or a plurality of cameras. For example, the electronic kanban push-and-cast system based on the relationship between the masses described in the first aspect of the patent application, wherein the video image captured by the camera is transmitted to the image recognition and analysis host through a coaxial cable and a network. For example, the electronic kanban push-and-cast system based on the relationship between the masses described in claim 3, wherein the network transmission of the video image is wired or wireless. For example, the electronic kanban push-pull system based on the relationship between the masses mentioned in the first application of the patent scope, wherein the image recognition analysis host is a front-end host, an embedded device or a remote server, which is characterized by image recognition and mass relationship. Judge and output the appropriate advertising content. For example, the electronic kanban push-and-cast system based on the relationship between the masses mentioned in the first application of the patent scope, wherein the image recognition module is analyzed by the face detection sub-module, the gender and age identification sub-module and the mass relationship Sub-module. For example, the electronic kanban push-and-cast system based on the relationship between the masses described in claim 6 of the patent application, wherein the face detection sub-module obtains the face position of the viewing mass through image block analysis. For example, the electronic kanban push-pull system based on the relationship between the masses mentioned in the sixth paragraph of the patent application scope, wherein the gender and age identification sub-module uses the facial feature analysis to obtain the gender and age information of the viewer. For example, the electronic kanban push-and-cast system based on the relationship between the masses described in claim 6 of the patent application, wherein the inter-person relationship analysis sub-module detects whether the hand, the hug, and the same clothes are in the style and color. The distance and position between the groups to obtain the relationship between the masses. For example, the electronic kanban push-casting system based on the relationship between the masses mentioned in the first paragraph of the patent application, wherein the inter-maritime relationship judgment module defines the relationship according to the identified gender, age information, and analysis of the posture and behavior among the masses. Definition of relationship project for relationship between partner, father and son, mother and child, and student Can be added or deleted. For example, the electronic kanban push-pull system based on the relationship between the masses mentioned in the first application of the patent scope, wherein the electronic kanban is an LED marquee, an LCD or an outdoor large-scale television. For example, the electronic kanban push-casting system based on the relationship between the masses mentioned in the first paragraph of the patent application, wherein the analysis result of the inter-person relationship judgment module is transmitted to the back end for statistical analysis, so as to provide advertisers to judge and target the masses. Lead customer information.