A smart advertising board and intelligent control method

Abstract

The present invention belongs to the technical field of advertising equipment and systems, and specifically relates to a smart advertising board and an intelligent control method, comprising an advertising board body and an advertising control system, wherein the advertising control system comprises a pedestrian path prediction module, a crowd feature recognition module, a data storage module, a data analysis module, and an advertising adjustment module; the pedestrian path prediction module is configured to observe the walking path of pedestrians, including the direction and speed of pedestrians walking, and infer pedestrians who are about to enter the advertising range of the advertising board based on the direction, speed and distance between the pedestrians and the advertising board, and mark them as target pedestrians; the crowd feature recognition module is configured to analyze the pedestrian information characteristics of the target pedestrians and output the pedestrian information characteristics of the target pedestrians; through the design of the intelligent control system, customized advertisements are recommended to users according to the different needs of different users, thereby improving the accuracy of advertising delivery and improving the effectiveness of advertising delivery.

Description

Technical Field

[0001] The present invention belongs to the technical field of advertising equipment and systems, and in particular to a smart advertising board and an intelligent control method. Background Art

[0002] Traditional billboards can only display one type of advertisement, that is, the content of the advertisement is printed and then hung on the billboard. With the development of the times, this advertising method has gradually been replaced, especially in some commercial areas. Commercial areas often have a large flow of people and advertising resources are very limited. Therefore, electronic billboards are usually used to display advertisements in these areas.

[0003] For shopping malls, the purpose of their advertising is often to display their internal promotional products in order to attract more customers to buy. Although existing electronic billboards can display multiple advertisements, most of them adopt a sequential display method. This cyclic display method is that multiple advertisements are played in a loop according to a specified duration and specified order, that is, the next advertisement is played after the previous advertisement is completed. However, most pedestrians are not interested in all product advertisements, and the audience groups of different products are also different. This leads to pedestrians passing by an advertisement location. If the current advertisement is not the product they are interested in, they will directly ignore the billboard. After walking through the public display area of ​​the billboard, they will not be able to see the subsequent advertisement content, and will not be able to know about the mall's promotional activities, which will cause the mall to miss customers and consume resources to play advertisements but do not generate positive benefits, thus failing to achieve the purpose of the mall's advertising.

[0004] To this end, the present invention provides a smart advertising board and an intelligent control method. Summary of the Invention

[0005] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0006] The technical solution adopted by the present invention to solve the technical problem is as follows: the smart advertising board of the present invention includes an advertising board body and an advertising control system, wherein the advertising control system includes a pedestrian path prediction module, a crowd feature recognition module, a data storage module, a data analysis module, and an advertising adjustment module;

[0007] The pedestrian path prediction module is configured to observe pedestrian paths, including their walking direction and speed. Based on the pedestrian's walking direction, speed, and distance from the billboard, it estimates pedestrians who are about to enter the billboard's advertising area and marks them as target pedestrians.

[0008] The crowd feature recognition module is configured to analyze pedestrian information features of a target pedestrian and output the pedestrian information features of the target pedestrian;

[0009] The data storage module is configured to store information characteristics of audiences of different advertisements, information of advertisements to be played, and information characteristics of VIP customers of the shopping mall;

[0010] The data analysis module is configured to obtain the pedestrian information features output by the crowd feature recognition module, and simultaneously output a mapping result of the pedestrian information features and the audience crowd information features of different advertisements in the data storage module;

[0011] The advertisement adjustment module is configured to obtain the mapping results output by the data analysis module and the prediction results of the pedestrian path prediction module, and adjust the order and duration of advertisement delivery.

[0012] Preferably, the advertising control system also includes an audience population identification module, which is configured to identify pedestrians watching advertisements and calibrate these pedestrians, while transmitting the video information of these pedestrians to the crowd feature identification module, which identifies the pedestrian information features of the audience population and stores the pedestrian information features of the advertising audience population in the data storage module as the audience population information features of the advertisement.

[0013] Preferably, the method for identifying pedestrians viewing advertisements is as follows:

[0014] First, the visual analysis camera is used to identify the gazes of pedestrians within the advertising range, and the pedestrians whose gazes are focused on the advertising board are selected as the people to be screened. Then, the time when the gazes of the people to be screened are focused on the advertising board is recorded as T. a , and then compare the time of advertising delivery to T b , if T a Greater than or equal to 0.5T b , indicating that the pedestrian has watched at least more than half of the advertisements, and the pedestrian is marked as the audience of the advertisement.

[0015] Preferably, the advertisement control system further includes an abnormal data elimination module, which is configured to eliminate abnormal data in the audience identification module. The specific method is as follows:

[0016] Determine the time that the eyes of the people to be screened are focused on the billboard as T a , if T a The value is greater than 3 or more advertising time T b and, if there are at least two types of these three or more advertisements, it is determined that the pedestrian is not the target audience of all the above advertisements.

[0017] Preferably, the method for obtaining the information characteristics of the advertising audience is as follows:

[0018] The first step is to separate the advertisement for which the audience information characteristics are to be obtained from the advertisement adjustment module so that it is not affected by the advertisement adjustment module;

[0019] The second step is to set the advertising cycle and run ads for the specified duration at different time periods within the cycle. Specifically, the advertising cycle can be 7 days. If there are holidays within the 7 days, the cycle can be shortened appropriately.

[0020] The third step is to identify the target audience of the advertisement through the audience identification module;

[0021] The fourth step is to identify the pedestrian information characteristics of the advertising audience through the crowd feature analysis module and store them in the data storage module.

[0022] Preferably, the specific method for the advertisement adjustment module to adjust the order and duration of advertisement delivery is as follows:

[0023] S1. Obtain information features of pedestrians who are about to enter the advertising area through the pedestrian path prediction module;

[0024] S2. Determine whether there is a VIP customer among the pedestrians about to enter the advertising area. If no VIP customer exists, execute steps S3 to S5. If a VIP customer exists, execute steps S6 to S8.

[0025] S3. Get the time when ordinary users enter the advertising area, and select the time from the advertising board as T k Ordinary users in the target user are taken as target users, and pedestrian information features of the target users are obtained;

[0026] S4. Filtering the advertisement with the largest mapping result between the information characteristics of the audience of each advertisement and the pedestrian information characteristics of the target user;

[0027] S5. Adjust the playback order of the filtered advertisements so that they become the advertisements to be played;

[0028] S6. Obtain the number of VIP customers. If the number of VIP customers is one, execute step S7. If the number of VIP customers is not one, execute step S8.

[0029] S7: Obtain the VIP customer information characteristics and determine whether the time the VIP customer enters the advertising area is greater than the advertising playback time T. s If the time a VIP customer enters the advertising area is less than the advertising playing time T s , then execute S701, if the time when the VIP customer enters the advertising area is greater than the advertising playing time T s , then execute S702 to S704;

[0030] S701, after screening the advertisements with the largest mapping results between the information characteristics of the audience groups and the information characteristics of the VIP customers, execute step S5;

[0031] S702, filter distance advertising board time T k Ordinary users within the company are the first target users, and VIP customers are the second target users;

[0032] S703: Obtain pedestrian information characteristics of the first target user, select the advertisement with the largest mapping result between the information characteristics of each advertising audience and the pedestrian information characteristics of the first target user, and select the advertisement with the largest mapping result between the information characteristics of each advertising audience and the information characteristics of the VIP customer as the second target advertisement;

[0033] S704: Determine whether the playing duration of the first target advertisement is greater than the time it takes for the second target user to walk to the advertisement area. If the playing duration of the first target advertisement is greater than the time it takes for the second target user to walk to the advertisement area, execute S7041; if the playing duration of the first target advertisement is less than the time it takes for the second target user to walk to the advertisement area, execute S7042 to S7043:

[0034] S7041. Compress the playing time of the first target advertisement to make it equal to or longer than the time it takes for the second target user to walk to the advertisement area, and then execute step S5.

[0035] S7042. Filter the advertisements with the top H values ​​of the mapping results of the advertising audience information characteristics and the pedestrian information characteristics of the first target user to form the first target advertisement set, until the sum of the duration of the first target advertisement set is greater than the time it takes for the second target user to walk to the advertisement area.

[0036] S7043, compress the duration of the first target advertisement set to make it equal to or longer than the time it takes for the second target user to walk to the advertisement area, and then execute step S5;

[0037] S8. Filter the VIP customer who takes the longest time to enter the advertising area as the second target user, filter other VIP customers as the first target users, and then execute S703 to S704.

[0038] Preferably, the data analysis module works as follows:

[0039] Assume that n advertisements are stored in the data storage module, and n is a positive integer greater than 1. n The corresponding audience population information feature set; after the data analysis module obtains the feature data output by the crowd feature recognition module, it will also establish a data set for the feature data of the target pedestrian. Suppose there are a total of target pedestrians, and a is a positive integer greater than or equal to 1, and the target pedestrian P is obtained. aThe characteristic data set of the target pedestrian corresponds to the information characteristics of the advertising audience, the characteristic data set of the target pedestrian is mapped with the information characteristics set of the advertising audience, the mapping data result of each advertisement is obtained, and is output to the advertising adjustment module.

[0040] Preferably, the T k The method to obtain is as follows:

[0041] Get Ad A n The corresponding playing time, select the advertisement with the shortest playing time as the shortest advertisement, multiply the playing time of the shortest advertisement by one quarter, and the resulting value is T k .

[0042] Preferably, the advertisement playing time T is the playing time of the shortest advertisement multiplied by half. s .

[0043] A method for intelligently controlling a smart advertising board, the method comprising the following steps:

[0044] Q1. Obtain the walking direction, speed, and distance between pedestrians and the billboard around the billboard, analyze and identify pedestrians about to enter the billboard's advertising range, and mark them as target pedestrians;

[0045] Q2. Analyze and obtain the information characteristics of the target pedestrian;

[0046] Q3. Map the target pedestrian's information characteristics with the audience information characteristics of the existing advertisements to obtain mapping data for each advertisement;

[0047] Q4. Based on the mapping data of each advertisement, select the advertisement with the largest mapping data;

[0048] Q5. Adjust the play order of the advertisement with the largest mapping data so that it becomes the next advertisement to be played.

[0049] The beneficial effects of the present invention are as follows:

[0050] 1. The smart advertising board and intelligent control method described in the present invention, through the design of an intelligent control system, recommend customized advertisements to users based on the different needs of different users, thereby improving the accuracy of advertising delivery and improving the effectiveness of advertising delivery.

[0051] 2. The smart advertising board and intelligent control method described in the present invention use visual analysis cameras to identify the gazes of pedestrians within the advertising range, and screen out pedestrians whose gazes are focused on the advertising board, screen the target audience of the advertisement, and analyze and statistically store them, accurately determining the target audience of the advertisement, laying the foundation for subsequent advertising delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The present invention will be further described below with reference to the accompanying drawings.

[0053] Figure 1 It is a system framework diagram of the present invention;

[0054] Figure 2 It is a flow chart of the advertisement delivery method of the present invention. DETAILED DESCRIPTION

[0055] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0056] like Figures 1 to 2 As shown, a smart advertising board and intelligent control method according to an embodiment of the present invention include an advertising board body and an advertising control system. The advertising control system includes a pedestrian path prediction module, a crowd feature recognition module, a data storage module, a data analysis module, and an advertising adjustment module.

[0057] The pedestrian path prediction module is configured to observe pedestrian paths, including their walking direction and speed. Based on the pedestrian's walking direction, speed, and distance from the billboard, it estimates pedestrians who are about to enter the billboard's advertising area and marks them as target pedestrians.

[0058] The crowd feature recognition module is configured to analyze pedestrian information features of a target pedestrian and output the pedestrian information features of the target pedestrian;

[0059] The data storage module is configured to store information characteristics of audiences of different advertisements, information of advertisements to be played, and information characteristics of VIP customers of the shopping mall;

[0060] The information of VIP customers of a shopping mall includes their gender, age, facial information and past consumption information. It should be noted that the facial information of VIP customers should be entered in advance with the consent of the VIP customers. VIP customers are generally the main consumer group of a shopping mall. By entering the facial information of VIP customers, corresponding advertisements can be displayed to VIP customers when they enter the shopping mall. At the same time, the probability of VIP customers purchasing products through advertisements after viewing advertisements of interest is also higher, which can effectively improve the accuracy of advertising delivery.

[0061] The data analysis module is configured to obtain the pedestrian information features output by the crowd feature recognition module, and simultaneously output a mapping result of the pedestrian information features and the audience crowd information features of different advertisements in the data storage module;

[0062] The advertisement adjustment module is configured to obtain the mapping results output by the data analysis module and the prediction results of the pedestrian path prediction module, and adjust the order and duration of advertisement delivery.

[0063] Before using the smart advertising board described in the embodiment of the present invention, the installation location must be determined. When selecting the installation location, care should be taken to avoid a wide field of view on both sides of the advertising board and the absence of large obstructions. After the advertising board is installed, it can be automatically operated by connecting it to a power source.

[0064] It should be noted that the billboard body includes a board body, a display screen is installed inside the board body, an acrylic protective plate is fixed to the surface of the display screen on the outside of the board body, a visual analysis camera is fixed to the top of the board body, and predictive analysis cameras are rotatably connected to both sides of the board body;

[0065] In the embodiment of the present invention, through the design of the advertising control system, when the advertising board is in use, before the previous advertisement is played, the passing pedestrians will first be recorded through the prediction analysis cameras on both sides of the board, and the recorded images will be transmitted to the pedestrian path prediction module in real time. After receiving the images recorded by the prediction analysis cameras, the pedestrian path prediction module will calculate how long it will take for different pedestrians to enter the display range of the display screen based on the walking paths and walking speeds of different people, and transmit these data to the crowd feature analysis module. The crowd feature analysis module will perform information feature analysis on these pedestrians, analyze the gender, age, and facial information of these pedestrians, and obtain the information of these pedestrians. The pedestrian information characteristics are used to determine whether there are VIP customers among them, and then the pedestrian information characteristics and judgment results of this part of pedestrians are transmitted to the data analysis module. After obtaining the pedestrian information characteristics and judgment results of this part of pedestrians, the data analysis module will give priority to serving VIP customers if there are VIP customers, and give priority to pushing advertisements to VIP customers; if there are no VIP customers, the audience information characteristics of different advertisements in the data storage module will be retrieved according to the pedestrian information characteristics of this part of pedestrians, and then the two sets of data will be mapped, and the mapping data of different advertisements will be output at the same time, and then the mapping results will be output to the advertisement adjustment module. The advertisement adjustment module will then adjust the advertisements to be broadcast on the billboard according to the mapping results in order to maximize the advertising effectiveness.

[0066] The advertising control system also includes an audience recognition module, which is configured to identify pedestrians watching the advertisement and calibrate these pedestrians. At the same time, the video information of these pedestrians is transmitted to the crowd feature recognition module. The crowd feature recognition module identifies the pedestrian information characteristics of the audience and stores the pedestrian information characteristics of the advertisement audience in the data storage module as the audience information characteristics of the advertisement.

[0067] Since the advertising range of a billboard has a certain range, only pedestrians within the public range of the billboard can observe the billboard. In an embodiment of the present invention, a visual analysis camera is installed on the top of the board body to identify pedestrians within the public range of the billboard, thereby distinguishing the target audience of the advertisement. It should be noted that the shooting range of the visual analysis camera is a 120-degree fan-shaped area, and the center line of the fan-shaped area is in the same straight line as the center line of the board body. The fan-shaped area is the best viewing area of ​​the billboard, that is, the advertising range of the billboard.

[0068] The method for identifying pedestrians viewing advertisements is as follows:

[0069] First, the visual analysis camera is used to identify the gazes of pedestrians within the advertising range, and the pedestrians whose gazes are focused on the advertising board are selected as the people to be screened. Then, the time when the gazes of the people to be screened are focused on the advertising board is recorded as T. a , and then compare the time of advertising delivery to T b , if T a Greater than or equal to 0.5T b , indicating that the pedestrian has watched at least more than half of the advertisements, and the pedestrian is marked as the audience of the advertisement.

[0070] By capturing the gazes of pedestrians, we can analyze the people whose eyes are focused on the billboards, screen this group of people, and determine the real advertising audience, thereby improving the accuracy of audience identification and laying the foundation for the precise delivery of subsequent advertisements.

[0071] The advertising control system further includes an abnormal data elimination module, which is configured to eliminate abnormal data in the audience identification module. The specific method is as follows:

[0072] Determine the time that the eyes of the people to be screened are focused on the billboard as T a , if T a The value is greater than 3 or more advertising time T b and, if there are at least two types of these three or more advertisements, it is determined that the pedestrian is not the target audience of all the above advertisements.

[0073] After the audience recognition module identifies the audience, there are cases where pedestrians are bored and watch advertisements. This part of the people is not the audience group of a certain advertisement. By eliminating them through the abnormal data elimination module, the accuracy of the information characteristics of the audience groups corresponding to different advertisements can be effectively improved.

[0074] The method for obtaining the information characteristics of the advertising audience is as follows:

[0075] The first step is to separate the advertisement for which the audience information characteristics are to be obtained from the advertisement adjustment module so that it is not affected by the advertisement adjustment module;

[0076] The second step is to set the advertising cycle and run ads for the specified duration at different time periods within the cycle. Specifically, the advertising cycle can be 7 days. If there are holidays within the 7 days, the cycle can be shortened appropriately.

[0077] The third step is to identify the target audience of the advertisement through the audience identification module;

[0078] The fourth step is to identify the pedestrian information characteristics of the advertising audience through the crowd feature analysis module and store them in the data storage module.

[0079] It is understandable that the initial target audience of an advertisement has been identified during the filming process, but some advertisements will have other potential users when they are actually released. For example, cosmetics advertisements are mainly aimed at women who are well-dressed, but when they are actually released, some men who pay attention to their personal appearance are also interested. In order to make the characteristic data of the audience corresponding to each advertisement more accurate, it is necessary to conduct trial releases of advertisements to expand the characteristic database, so as to facilitate the subsequent precise delivery of advertisements.

[0080] By trial-running new ads in different time periods, and without being affected by the ad adjustment module during the trial run, the basic data of the ad's audience can be effectively obtained. After obtaining the basic data, the ad can be brought under the control of the ad adjustment module and controlled for playback.

[0081] It should be noted that the specific implementation method of the second step is as follows:

[0082] The trial delivery time of the advertisement to be delivered is set to one week, and the corresponding delivery time period is selected according to the installation location of the billboard. In the embodiment of the present invention, a shopping mall with a large flow of people is taken as an example. The business hours of the shopping mall are generally from 10 am to 10 pm, a total of 12 hours, and the passenger flow of the shopping mall on weekdays is often less than the passenger flow on non-working days. In this way, on weekdays, the advertisement to be delivered is delivered for 3 hours a day. When delivering, it should be noted that it should be distinguished from the delivery time of the previous few days to reduce the overlap of delivery time. For example, on Monday, the delivery time node of the advertisement to be delivered is from 10 am to 1 pm, and on Tuesday, the delivery time node of the advertisement to be delivered is from 1 pm to 4 pm; and so on. Push, thereby improving the accuracy of audience identification; on weekends, it is necessary to roll out the delivery mode of the advertisement to be delivered and the existing advertisements. For example, according to actual needs, a new advertisement can be played after playing X old advertisements, where X is a positive integer greater than 1 and is manually set according to actual conditions. It should be noted that it should be ensured that the advertisement is broadcast at different time periods on weekends; in this way, on weekends with peak customer traffic, the accurate delivery of existing advertisements can be guaranteed and the trial delivery of new advertisements can be completed, thereby effectively identifying the audience of the advertisement to be delivered and extracting the basic data of the new advertisement audience, which is convenient for subsequent further accurate delivery of the new advertisement.

[0083] The specific method of the advertisement adjustment module to adjust the order and duration of advertisement delivery is as follows:

[0084] S1. Obtain information features of pedestrians who are about to enter the advertising area through the pedestrian path prediction module;

[0085] S2. Determine whether there is a VIP customer among the pedestrians about to enter the advertising area. If no VIP customer exists, execute steps S3 to S5. If a VIP customer exists, execute steps S6 to S8.

[0086] S3. Get the time when ordinary users enter the advertising area, and select the time from the advertising board as T k Ordinary users in the target user are taken as target users, and pedestrian information features of the target users are obtained;

[0087] S4. Filtering the advertisement with the largest mapping result between the information characteristics of the audience of each advertisement and the pedestrian information characteristics of the target user;

[0088] S5. Adjust the playback order of the filtered advertisements so that they become the advertisements to be played;

[0089] S6. Obtain the number of VIP customers. If the number of VIP customers is one, execute step S7. If the number of VIP customers is not one, execute step S8.

[0090] S7: Obtain the VIP customer information characteristics and determine whether the time the VIP customer enters the advertising area is greater than the advertising playback time T. s If the time a VIP customer enters the advertising area is less than the advertising playing time T s , then execute S701, if the time when the VIP customer enters the advertising area is greater than the advertising playing time T s , then execute S702 to S704; It should be noted that the advertising playing time T s The duration of the shortest ad multiplied by half;

[0091] S701, after screening the advertisements with the largest mapping results between the information characteristics of the audience groups and the information characteristics of the VIP customers, execute step S5;

[0092] S702, filter distance advertising board time T k Ordinary users in the company are the first target users, and VIP customers are the second target users. It should be noted that T k The duration of the shortest ad multiplied by one-fourth;

[0093] S703: Obtain pedestrian information characteristics of the first target user, select the advertisement with the largest mapping result between the information characteristics of each advertising audience and the pedestrian information characteristics of the first target user, and select the advertisement with the largest mapping result between the information characteristics of each advertising audience and the information characteristics of the VIP customer as the second target advertisement;

[0094] S704: Determine whether the playing duration of the first target advertisement is greater than the time it takes for the second target user to walk to the advertisement area. If the playing duration of the first target advertisement is greater than the time it takes for the second target user to walk to the advertisement area, execute S7041; if the playing duration of the first target advertisement is less than the time it takes for the second target user to walk to the advertisement area, execute S7042 to S7043:

[0095] S7041. Compress the playback duration of the first target advertisement so that it is equal to or longer than the time required for the second target user to walk to the advertisement area, and then execute step S5. It should be noted that, during playback, the playback order should be to play the compressed first target advertisement first, followed by the second target advertisement. That is, the system will not run after the first target advertisement is played, and will only perform the next round of target determination when the second target advertisement is played. The specific compression method may be to fast-forward the playback speed of the advertisement or to delete part of the advertisement content.

[0096] S7042. Filter the advertisements with the top H values ​​of the mapping results of the advertising audience information characteristics and the pedestrian information characteristics of the first target user to form the first target advertisement set, until the sum of the duration of the first target advertisement set is greater than the time it takes for the second target user to walk to the advertisement area.

[0097] S7043, compress the duration of the first target advertisement set to make it equal to or longer than the time it takes for the second target user to walk to the advertisement area, and then execute step S5;

[0098] S8. Filter the VIP customer who takes the longest time to enter the advertising area as the second target user, filter other VIP customers as the first target users, and then execute S703 to S704.

[0099] The working process of the data analysis module is as follows:

[0100] Assume that n advertisements are stored in the data storage module, and n is a positive integer greater than 1. n The corresponding audience population information feature set; after the data analysis module obtains the feature data output by the crowd feature recognition module, it will also establish a data set for the feature data of the target pedestrian. Suppose there are a total of target pedestrians, and a is a positive integer greater than or equal to 1, and the target pedestrian P is obtained. a The characteristic data set of the target pedestrian corresponds to the information characteristics of the advertising audience, the characteristic data set of the target pedestrian is mapped with the information characteristics set of the advertising audience, the mapping data result of each advertisement is obtained, and is output to the advertising adjustment module.

[0101] For example: Each ad A n The corresponding audience information feature set can be expressed in the following way:

[0102] A1={M、W、Y}

[0103] A2={M、O}

[0104] A3={W、O}

[0105]

[0106] A n ={M, W, Y}

[0107] Among them, male is M, female is W, child is Y, and adult is O;

[0108] Target pedestrian P a The feature data set can be expressed by the following formula:

[0109] P1={M, Y}

[0110] P2={M、O}

[0111] P3={W、O}

[0112]

[0113] P a ={W, Y}

[0114] The characteristic data set of the target pedestrian is mapped with the characteristic data set of the advertising audience to obtain mapping data.

[0115] Example 1: Assume that there are 10 target pedestrians, including 2 adult males, 4 adult females, 3 male children, and 1 female child. The data storage module stores 5 advertisements:

[0116] The target audience of the first advertisement is adult males, so the mapping data is 2.

[0117] The target audience information characteristics of the second advertisement are: adult females, so the mapping data is 4;

[0118] The target audience information feature of the third advertisement is: children, so the mapping data is 4;

[0119] The target audience information characteristic of the fourth advertisement is: adults, so the mapping data is 6;

[0120] The audience information characteristics of the fifth advertisement are: adult women, female children, so the mapping data is 5;

[0121] The advertisement with the most matching mapping data is the fourth advertisement, and this advertisement will be played. It should be noted that if there are multiple advertisements with the same maximum mapping data, one of them will be played randomly.

[0122] Example 2: Assume that there are 8 target pedestrians, including 4 adult males, 2 adult females (one of whom is a VIP), 1 male child, and 1 female child. The data storage module stores 4 advertisements:

[0123] The first advertisement's target demographic is adult males, so the mapping data is 4.

[0124] The second ad's target demographic is adult women, so the mapping data is 2. The VIP user's historical consumption history indicates they have purchased similar products to those advertised.

[0125] The target audience information feature of the third advertisement is: children, so the mapping data is 2;

[0126] The target audience information characteristics of the fourth advertisement are: adult women and female children, so the mapping data is 3;

[0127] At this time, it is judged whether the time when the VIP customer enters the advertising area is greater than the advertising playing time T s , if it is less than, play the second advertisement first;

[0128] If the time a VIP customer enters the advertising area is greater than the advertising playing time T s , then further determine whether the playing time of the first advertisement is greater than the time it takes for the VIP customer to enter the advertising area. If it is less, compress the playing time of the first advertisement to make it equal to the time it takes for the VIP customer to enter the advertising area, and then play the second advertisement;

[0129] If the playing duration of the first ad is less than the time it takes for the VIP customer to enter the ad area, determine whether the playing duration of the first and fourth ads is greater than the time it takes for the VIP customer to enter the ad area. In this case, the playing duration generally exceeds the time it takes for the VIP customer to enter the ad area. In this case, compress the first and fourth ads respectively to make them equal to the time it takes for the VIP customer to enter the ad area. Then play the compressed first and fourth ads in sequence, and then play the second ad. If the playing duration of the first and fourth ads is less than the time it takes for the VIP customer to enter the ad area, add the third ad and compress it in the same way.

[0130] Example 3: Assume that there are 12 target pedestrians, including 4 adult males, 2 of whom are VIPs and one of whom takes the longest time to enter the advertising area, 6 adult females, 3 of whom are VIPs, 1 male child, and 1 female child. The data storage module stores 5 advertisements:

[0131] The target demographic of the first ad is adult males, so the mapping data is 4, and the VIP user has purchased similar products in their historical consumption history.

[0132] The second ad's target demographic is adult women, so the mapping data is 6. The VIP user's historical consumption history indicates they have purchased similar products to those advertised.

[0133] The target audience information feature of the third advertisement is: children, so the mapping data is 2;

[0134] The fourth ad's target demographics are: adult women and female children. The mapping value is 7, and the VIP user's historical consumption history indicates they have purchased similar products to those advertised.

[0135] The target audience information feature of the fifth advertisement is: adults, so the mapping data is 10;

[0136] Identify the VIP customer who takes the longest time to enter the advertising area, and determine whether the time the VIP customer takes to enter the advertising area is greater than the advertising playback time T s , if it is less than, the first advertisement will be played first;

[0137] If the time the VIP customer enters the advertising area is greater than the advertising playing time T s, then further determine whether the playing time of the fourth advertisement is greater than the time when the VIP customer enters the advertising area. If it is less, compress the playing time of the fourth advertisement to make it equal to the time when the VIP customer enters the advertising area, and then play the first advertisement;

[0138] If the playing duration of the fourth advertisement is less than the time when the VIP customer enters the advertising area, then determine whether the playing duration of the fourth advertisement and the second advertisement is greater than the time when the VIP customer enters the advertising area. At this time, the playing duration generally exceeds the time when the VIP customer enters the advertising area. Compress the fourth advertisement and the second advertisement respectively to make them equal to the time when the VIP customer enters the advertising area. Then play the compressed fourth advertisement and the second advertisement in sequence, and then play the first advertisement. If the playing duration of the fourth advertisement and the second advertisement is less than the time when the VIP customer enters the advertising area, add the third advertisement and the fifth advertisement for compression in the same way.

[0139] The T k The method to obtain is as follows:

[0140] Get Ad A n The corresponding playing time, select the advertisement with the shortest playing time as the shortest advertisement, multiply the playing time of the shortest advertisement by one quarter, and the resulting value is T k .

[0141] By obtaining T k Specific value of the distance from the advertising board T k For pedestrians within the advertisement board, it can effectively ensure that when pedestrians reach the advertisement board, there are still three-quarters of the remaining broadcast time, so that pedestrians can watch the advertisement content to the maximum extent without missing most of the content when watching the advertisement.

[0142] The playing time of the shortest advertisement multiplied by half is the advertising playing time T. s .

[0143] If the VIP customer is too far away from the billboard, the advertisement for the VIP customer may be played halfway before the VIP customer arrives. At this time, part of the advertisement content cannot be displayed, resulting in a poor experience. s The value of is set to half the duration of the shortest ad, which can leave enough time to compress other ads and play them until the VIP customer reaches the range of the billboard.

[0144] A method for intelligently controlling a smart advertising board, the method comprising the following steps:

[0145] Q1. Obtain the walking direction, speed, and distance between pedestrians and the billboard around the billboard, analyze and identify pedestrians about to enter the billboard's advertising range, and mark them as target pedestrians;

[0146] Q2. Analyze and obtain the information characteristics of the target pedestrian;

[0147] Q3. Map the target pedestrian's information features with the audience information features of existing advertisements to obtain mapping data for each advertisement;

[0148] Q4. Based on the mapping data of each advertisement, select the advertisement with the largest mapping data;

[0149] Q5. Adjust the play order of the advertisement with the largest mapping data so that it becomes the next advertisement to be played.

[0150] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligently controlling a smart advertising board, characterized by: The method comprises the following steps: Q1. Obtain the walking direction, speed, and distance between pedestrians and the billboard around the billboard, analyze and identify pedestrians about to enter the billboard's advertising range, and mark them as target pedestrians; Q2. Analyze and obtain the information characteristics of the target pedestrian; Q3. Map the target pedestrian's information characteristics with the audience information characteristics of the existing advertisements to obtain mapping data for each advertisement; Q4. Based on the mapping data of each advertisement, select the advertisement with the largest mapping data; Q5. Adjust the play order of the advertisement with the largest mapping data so that it becomes the next advertisement to be played; The method is applicable to an advertising board body and an advertising control system, wherein the advertising control system includes a pedestrian path prediction module, a crowd feature recognition module, a data storage module, a data analysis module, and an advertising adjustment module; The pedestrian path prediction module is configured to observe pedestrian paths, including their walking direction and speed. Based on the pedestrian's walking direction, speed, and distance from the billboard, it estimates pedestrians who are about to enter the billboard's advertising area and marks them as target pedestrians. The crowd feature recognition module is configured to analyze pedestrian information features of a target pedestrian and output the pedestrian information features of the target pedestrian; The data storage module is configured to store information characteristics of audiences of different advertisements, information of advertisements to be played, and information characteristics of VIP customers of the shopping mall; The data analysis module is configured to obtain the pedestrian information features output by the crowd feature recognition module, and simultaneously output a mapping result of the pedestrian information features and the audience crowd information features of different advertisements in the data storage module; The advertisement adjustment module is configured to obtain the mapping results output by the data analysis module and the prediction results of the pedestrian path prediction module, and adjust the order and duration of advertisement delivery; The advertising control system further includes an audience recognition module, which is configured to identify pedestrians viewing the advertisement and to mark the pedestrians, and transmit video information of the pedestrians to the crowd feature recognition module, which recognizes the pedestrian information features of the audience and stores the pedestrian information features of the advertisement audience in the data storage module as the audience information features of the advertisement; The method for identifying pedestrians viewing advertisements is as follows: First, the visual analysis camera is used to identify the gazes of pedestrians within the advertising range, and the pedestrians whose gazes are focused on the advertising board are selected as the people to be screened. Then, the time when the gazes of the people to be screened are focused on the advertising board is recorded as T. a , and then compare the time of advertising delivery to T b , if T a Greater than or equal to 0.5T b , indicating that the pedestrian has watched at least more than half of the advertisements, and the pedestrian is marked as the audience of the advertisement; The advertising control system further includes an abnormal data elimination module, which is configured to eliminate abnormal data in the audience identification module. The specific method is as follows: Determine the time that the eyes of the people to be screened are focused on the billboard as T a , if T a The value is greater than 3 or more advertising time T b If the sum of 3 or more advertisements exists in at least two types, it is determined that the pedestrian is not the target audience of all the above advertisements; The method for obtaining the information characteristics of the advertising audience is as follows: The first step is to separate the advertisement for which the audience information characteristics are to be obtained from the advertisement adjustment module so that it is not affected by the advertisement adjustment module; The second step is to set the advertising cycle and run ads at different time periods within the cycle according to the specified duration; The third step is to identify the target audience of the advertisement through the audience identification module; The fourth step is to identify the pedestrian information characteristics of the advertising audience through the crowd feature analysis module and store them in the data storage module; The specific method of the advertisement adjustment module to adjust the order and duration of advertisement delivery is as follows: S1. Obtain information features of pedestrians who are about to enter the advertising area through the pedestrian path prediction module; S2. Determine whether there is a VIP customer among the pedestrians about to enter the advertising area. If no VIP customer exists, execute steps S3 to S5. If a VIP customer exists, execute steps S6 to S8. S3. Get the time when ordinary users enter the advertising area, and select the time from the advertising board as T k Ordinary users in the target user are taken as target users, and pedestrian information features of the target users are obtained; S4. Filtering the advertisement with the largest mapping result between the information characteristics of the audience of each advertisement and the pedestrian information characteristics of the target user; S5. Adjust the playback order of the filtered advertisements so that they become the advertisements to be played; S6. Obtain the number of VIP customers. If the number of VIP customers is one, execute step S7. If the number of VIP customers is not one, execute step S8. S7: Obtain the VIP customer information characteristics and determine whether the time the VIP customer enters the advertising area is greater than the advertising playback time T. s If the time a VIP customer enters the advertising area is less than the advertising playing time T s , then execute S701, if the time when the VIP customer enters the advertising area is greater than the advertising playing time T s , then execute S702 to S704; S701, after screening the advertisements with the largest mapping results between the information characteristics of the audience groups and the information characteristics of the VIP customers, execute step S5; S702, filter distance advertising board time T k Ordinary users within the company are the first target users, and VIP customers are the second target users; S703: Obtain pedestrian information characteristics of the first target user, select the advertisement with the largest mapping result between the information characteristics of each advertising audience and the pedestrian information characteristics of the first target user, and select the advertisement with the largest mapping result between the information characteristics of each advertising audience and the information characteristics of the VIP customer as the second target advertisement; S704: Determine whether the playing duration of the first target advertisement is greater than the time it takes for the second target user to walk to the advertisement area. If the playing duration of the first target advertisement is greater than the time it takes for the second target user to walk to the advertisement area, execute S7041; if the playing duration of the first target advertisement is less than the time it takes for the second target user to walk to the advertisement area, execute S7042 to S7043: S7041. Compress the playing time of the first target advertisement to make it equal to or longer than the time it takes for the second target user to walk to the advertisement area, and then execute step S5. S7042. Filter the advertisements with the top H values ​​of the mapping results of the advertising audience information characteristics and the pedestrian information characteristics of the first target user to form the first target advertisement set, until the sum of the duration of the first target advertisement set is greater than the time it takes for the second target user to walk to the advertisement area. S7043, compress the duration of the first target advertisement set to make it equal to or longer than the time it takes for the second target user to walk to the advertisement area, and then execute step S5; S8. Filter the VIP customer who takes the longest time to enter the advertising area as the second target user, filter other VIP customers as the first target users, and then execute S703 to S704.

2. The intelligent control method of a smart advertising board according to claim 1, characterized in that: The working process of the data analysis module is as follows: Assume that n advertisements are stored in the data storage module, and n is a positive integer greater than 1. n The corresponding audience population information feature set; after the data analysis module obtains the feature data output by the crowd feature recognition module, it will also establish a data set for the feature data of the target pedestrian. Suppose there are a total of target pedestrians, and a is a positive integer greater than or equal to 1, and the target pedestrian P is obtained. a The characteristic data set of the target pedestrian corresponds to the information characteristics of the advertising audience, the characteristic data set of the target pedestrian is mapped with the information characteristics set of the advertising audience, the mapping data result of each advertisement is obtained, and is output to the advertising adjustment module.

3. The intelligent control method of a smart advertising board according to claim 2, characterized in that: The T k The method to obtain is as follows: Get Ad A n The corresponding playing time, select the advertisement with the shortest playing time as the shortest advertisement, multiply the playing time of the shortest advertisement by one quarter, and the resulting value is T k .

4. The intelligent control method of a smart advertising board according to claim 3, characterized in that: The playing time of the shortest advertisement multiplied by half is the advertising playing time T. s .

Images