CN107978229B - A face recognition-based advertising machine and its control method - Google Patents

Abstract

The invention relates to the technical field of advertising equipment and an advertisement pushing method, in particular to an advertising machine based on face recognition and a control method thereof, wherein the advertising machine comprises a display, a camera which is arranged at the lower side edge of the display and used for capturing a face, a base and a supporting rod, the upper end of the supporting rod is connected with the display, the lower end of the supporting rod is fixed on the base, and a non-stick film is attached to the lower surface of the base; the advertising machine can effectively overcome the defects that the existing floor type advertising machine and the splicing type advertising machine occupy large space and are not suitable for being placed on the tabletop of a cash desk with narrow space, and the placement position is convenient to adjust by fixing the non-stick film on the edge of the desk; the control method can summarize the playing times of the single advertisements, perform sequencing statistics, further identify the popular advertisements, increase and promote the popular advertisements, improve the advertising efficiency of the advertisements, manage the advertisement updating work of the advertisement player through the server, and has the advantages of quick response, convenient control and high efficiency.

Description

A face recognition-based advertising machine and its control method

technical field

The present invention relates to the technical field of advertising equipment and advertising push methods, in particular to an advertising machine based on face recognition and a control method thereof.

Background technique

With the increasingly developed domestic business and consumption environment, the demand for advertising is also growing, and digital, networked, and information-based multimedia advertising machines have also become a bright spot in the advertising media market. In residential areas, schools and prosperous areas with large passenger flow, convenience stores have been greatly developed in my country. However, if you want to place advertisements in such small-scale convenience stores, traditional floor-standing advertising machines, wall-mounted advertising machines, etc. Advertising machines and spliced advertising machines are not suitable for use in convenience stores with large palms. The traditional bulky advertising machines limit the scope of activities. In order to address this situation, a new type of advertising machine that can be placed on the cashier desk without being used has been invented. An advertising machine that takes up space and will not leave foreign objects on the desktop, so that the advertising machine can be flexibly placed on the desktop, and the placement of the advertising machine can be adjusted at any time according to the actual needs of the situation.

The control method of the existing advertising machine is to arrange the advertisements, and then play the corresponding advertisements in the arranged order without interruption. In order to attract consumers, it is difficult to improve the efficiency of advertising in a targeted manner.

SUMMARY OF THE INVENTION

In order to overcome the shortcomings and deficiencies in the prior art, the purpose of the present invention is to provide an advertising machine based on face recognition, which can effectively overcome the existing floor-standing advertising machines and wall-mounted advertising machines. . The splicing type advertising machine occupies a large space and is not suitable for placing it on the desk of the cash register with a narrow space. It is fixed on the edge of the desk by a non-stick film, which is easy to adjust the placement position.

Another object of the present invention is to provide a control method of an advertising machine based on face recognition, the control method can summarize the playing times of a single advertisement, perform sorting and statistics, and then identify more popular advertisements and increase the popularity of advertisements. advertising and improve the efficiency of advertising.

The object of the present invention is achieved through the following technical solutions: a face recognition-based advertising machine, including a display and a camera positioned on the lower edge of the display for capturing human faces, also including a base and a support rod, the upper end of the support rod Connected with the display, the lower end of the support rod is fixed on the base, and the lower surface of the base is pasted with a non-stick film.

The non-stick film is nanometer micro-adhesive, and the display is also provided with a face recognition system for identifying whether someone enters the face capture area.

A control method for an advertising machine based on face recognition, comprising the following steps: A. The server classifies advertisements according to gender and age group, establishes and stores classification labels, places the advertising machine in places such as cashiers or counters, and adjusts the Face capture angle, the advertising machine and the server are connected through the network, the server transmits the classified advertisements to the advertising machine, and the advertising machine stores the advertisements in the corresponding label directory; B. The advertising machine captures the facial expressions of the people who enter the capture area. Perform gender and age analysis on the captured facial expressions, analyze and complete the playback of advertisements under the corresponding category tags, and accumulate the number of advertisement playbacks corresponding to the category tags, and record the total playback times of the corresponding advertisements; C. Finally, the total number of advertisements of each advertisement is recorded in real time The number of playbacks is sent back to the server, and the server counts the proportion of the total number of playbacks of each advertisement to the sum of the number of playbacks of all advertisements. When updating the advertisement, the server increases the number of advertisements in the classification label category corresponding to the top-ranked advertisement according to the statistical result. The number of advertisements is sent to the advertising machine, and the advertisements ranked lower in the total number of plays are removed.

The following steps are also included between the step A and the step B: the advertising machine senses whether there is a person's face in the capture area facing the display through the face recognition system, if not, the advertising machine automatically enters the standby state, if there is, then enter. Step B.

In the step B, before playing the advertisement under the corresponding classification label, the following steps are further included: the advertisement machine judges whether the classified label contains an advertisement corresponding to the analysis result; and send the corresponding classification label to the server immediately, and the server immediately sends the advertisement corresponding to the classification label to the advertising machine. If there is, enter the advertisement playing step in step B.

In the step B, when the advertisement machine plays an advertisement, if the advertisement machine captures the faces of other personnel facing the display, the advertisement machine automatically captures the facial expressions of the personnel and conducts gender and age analysis, and a person leaves at present. The advertisement under the corresponding category label is automatically played. When the advertisement machine plays the advertisement, the advertisement machine does not capture the face of other people facing the display, then the corresponding advertisement playback is completed and the advertisement machine enters the standby state.

The step B also includes a recording step of recording the total playing time of a single advertisement on a regular basis. The statistical method of the single playing time of a single advertisement is: from the time interval from the start of the advertisement to the single stop of the broadcast, the advertising machine returns the recorded information in real time. to the server.

Described step B also includes advertisement playing time period control step: each advertisement under the classification label is provided with corresponding sub-label, and this sub-label is marked with the playing time period of the advertisement, such as the broadcast time and the sub-label of a certain advertisement. For the recorded playback time period, the advertisement will be played preferentially under the premise that the playback conditions are met.

The step B also includes the step of controlling the advertisements to stop broadcasting: each advertisement under the classification label is provided with a corresponding total playback duration label and a total playback count label, the total playback duration label is set with the total playback duration, and the total playback count label is set. There is a total number of playbacks. When the total playback duration or total playback times of the corresponding advertisement reaches the duration recorded in the total playback duration label or the number of times recorded in the total playback times label, the advertisement will be stopped.

It also includes a step of advertising playback volume control: the advertising machine senses the size of ambient noise through the volume sensing module, and if the environmental noise is large, the playback volume of the advertising machine is increased, and if the environmental noise is small, the playback volume of the advertising machine is reduced.

The beneficial effects of the present invention are: the advertising machine based on face recognition of the present invention can effectively overcome the large space occupied by existing floor-standing advertising machines, wall-mounted advertising machines and splicing advertising machines, and is not suitable for being placed in a cash register with a narrow space The defect of the desktop, it is fixed on the edge of the table by non-stick film, which is easy to adjust the placement position.

The control method of the present invention can summarize the playing times of a single advertisement, perform sorting and statistics, and then identify the more popular advertisements, increase the popular advertisements, improve the publicity efficiency of advertisements, and manage the advertisement update work of the advertisement machine through the server. , fast response, convenient control and high efficiency.

Description of drawings

FIG. 1 is a perspective view of the advertising machine of the present invention.

FIG. 2 is a schematic diagram of the control method of the present invention.

Reference numerals are: display 1 , camera 2 , base 3 , support rod 4 .

Detailed ways

In order to facilitate the understanding of those skilled in the art, the present invention will be further described below with reference to the embodiments and accompanying drawings 1 to 2 , and the content mentioned in the embodiments does not limit the present invention.

A face recognition-based advertising machine, comprising a display 1 and a camera 2 arranged on the lower edge of the display 1 for capturing human faces, also comprising a base 3 and a support rod 4, the upper end of the support rod 4 is connected with the display 1, The lower end of the support rod 4 is fixed on the base 3, and the lower surface of the base is attached with a non-stick film.

The non-stick film is nanometer micro-adhesive, and the display is also provided with a face recognition system for identifying whether someone enters the face capture area.

A control method for an advertising machine based on face recognition, comprising the following steps: A. The server classifies advertisements according to gender and age group, establishes and stores classification labels, places the advertising machine in places such as cashiers or counters, and adjusts the Face capture angle, the advertising machine and the server are connected through the network, the server transmits the classified advertisements to the advertising machine, and the advertising machine stores the advertisements in the corresponding label directory; B. The advertising machine captures the facial expressions of the people who enter the capture area. Perform gender and age analysis on the captured facial expressions, analyze and complete the playback of advertisements under the corresponding category tags, and accumulate the number of advertisement playbacks corresponding to the category tags, and record the total playback times of the corresponding advertisements; C. Finally, the total number of advertisements of each advertisement is recorded in real time The number of playbacks is sent back to the server, and the server counts the proportion of the total number of playbacks of each advertisement to the sum of the number of playbacks of all advertisements. When updating the advertisement, the server increases the number of advertisements in the classification label category corresponding to the top-ranked advertisement according to the statistical result. The number of advertisements is sent to the advertising machine, and the advertisements ranked lower in the total number of plays are removed.

The following steps are also included between the step A and the step B: the advertising machine senses whether there is a person's face in the capture area facing the display through the face recognition system, if not, the advertising machine automatically enters the standby state, if there is, then enter. Step B.

In the step B, before playing the advertisement under the corresponding classification label, the following steps are further included: the advertisement machine judges whether the classified label contains an advertisement corresponding to the analysis result; and send the corresponding classification label to the server immediately, and the server immediately sends the advertisement corresponding to the classification label to the advertising machine. If there is, enter the advertisement playing step in step B.

In the step B, when the advertisement machine plays an advertisement, if the advertisement machine captures the faces of other personnel facing the display, the advertisement machine automatically captures the facial expressions of the personnel and conducts gender and age analysis, and a person leaves at present. The advertisement under the corresponding category label is automatically played. When the advertisement machine plays the advertisement, the advertisement machine does not capture the face of other people facing the display, then the corresponding advertisement playback is completed and the advertisement machine enters the standby state.

The step B also includes a recording step of recording the total playing time of a single advertisement on a regular basis. The statistical method of the single playing time of a single advertisement is: from the time interval from the start of the advertisement to the single stop of the broadcast, the advertising machine returns the recorded information in real time. to the server.

Described step B also includes advertisement playing time period control step: each advertisement under the classification label is provided with corresponding sub-label, and this sub-label is marked with the playing time period of the advertisement, such as the broadcast time and the sub-label of a certain advertisement. For the recorded playback time period, the advertisement will be played preferentially under the premise that the playback conditions are met.

The step B also includes the step of controlling the advertisements to stop broadcasting: each advertisement under the classification label is provided with a corresponding total playback duration label and a total playback count label, the total playback duration label is set with the total playback duration, and the total playback count label is set. There is a total number of playbacks. When the total playback duration or total playback times of the corresponding advertisement reaches the duration recorded in the total playback duration label or the number of times recorded in the total playback times label, the advertisement will be stopped.

It also includes a step of advertising playback volume control: the advertising machine senses the size of ambient noise through the volume sensing module, and if the environmental noise is large, the playback volume of the advertising machine is increased, and if the environmental noise is small, the playback volume of the advertising machine is reduced.

The above-mentioned embodiment is a preferred implementation scheme of the present invention. In addition, the present invention can also be implemented in other ways, and any obvious replacements are within the protection scope of the present invention without departing from the concept of the present invention.

Claims (7)

1. a control method of an advertising machine based on face recognition, is characterized in that: comprise the following steps: A, server classify advertisement according to gender and age group, establish classification label and store, advertising machine is placed on cash register or The counter and adjust the face capture angle, the advertising machine and the server are connected through the network, the server transmits the classified advertisement to the advertising machine, and the advertising machine stores the advertisement in the corresponding label directory; B. The advertising machine captures the information of the person entering the capture area. Facial expressions and gender and age analysis of the captured facial expressions are performed, and the advertisements under the corresponding classification labels are analyzed and played, and the number of times of advertisement playbacks corresponding to the corresponding classification labels is accumulated, and the total number of times of corresponding advertisements is recorded; C. Finally, in real time, The total playing times of each advertisement is sent back to the server, and the server counts the ratio of the total playing times of each advertisement to the sum of the total playing times of all advertisements. When updating advertisements, the server increases the classification label category corresponding to the top advertisement according to the statistical result. The number of advertisements in the advertisement machine is sent to the advertisement machine, and the advertisements ranked at the bottom of the total number of broadcasts are removed. In the step B, when the advertisement machine plays the advertisement, if the advertisement machine captures the faces of other people facing the display, then The advertising machine automatically captures the facial expression of the person and analyzes the gender and age. When the previous person leaves, the advertisement under the corresponding category label is automatically played. When the advertising machine plays the advertisement, the advertising machine does not capture the faces of other people. display, the corresponding advertisement playback is completed and the advertisement machine enters the standby state. 2. the control method of a kind of advertising machine based on face recognition according to claim 1, it is characterized in that: between described step A and step B, also comprise the following steps: advertising machine is induced to capture area by face recognition system Whether there is a person's face facing the display, if not, the advertising machine will automatically enter the standby state, if so, go to step B. 3. the control method of a kind of advertising machine based on face recognition according to claim 1, it is characterized in that: before playing the advertisement under the corresponding classification label in the described step B, also comprise the following steps: the advertisement machine judges that under the classification label Whether there is an advertisement corresponding to the analysis result, if not, the advertising machine will continue to remain in the standby state, and record the information about the lack of the corresponding advertisement and immediately send the classification label corresponding to the information back to the server, and the server will immediately send the advertisement corresponding to the classification label to the server. If there is an advertisement machine, enter the advertisement playing step in step B. 4. the control method of a kind of advertisement machine based on face recognition according to claim 1, is characterized in that: described step B also comprises the recording step of recording the total playing duration in the regular period of the single advertisement, the single advertisement single playing duration The statistical method is as follows: the time interval from the start of the advertisement to the single stop of the broadcast, the advertisement machine transmits the recorded information back to the server in real time. 5. the control method of a kind of advertising machine based on face recognition according to claim 4, it is characterized in that: described step B also comprises advertisement playing time period control step: each advertisement under the classification label is provided with corresponding Sub-tag, the sub-tag is marked with the playback time period of the advertisement. If the playback time corresponds to the playback time period recorded in the sub-tag of an advertisement, the advertisement will be played preferentially under the premise that the playback conditions are met. 6. the control method of a kind of advertising machine based on face recognition according to claim 4, is characterized in that: described step B also comprises the step of advertisement off-air control: each advertisement under the classification label is provided with corresponding total Playing time label and total playing times label, the total playing time is set in the total playing time label, and the total playing times is set in the total playing times label. Stop the ad when it reaches the number of times recorded in the Duration or Total Plays tab. 7. a kind of control method of advertising machine based on face recognition according to claim 1, it is characterized in that: also comprise advertising playback volume control step: advertising machine senses the size of surrounding environmental noise by volume sensing module, and the environmental noise is large Then increase the playback volume of the advertising machine, and reduce the playback volume of the advertising machine if the ambient noise is small.

Images