An elevator media delivery management system and method based on the Internet of Things

Abstract

The invention discloses an elevator media delivery management system and method based on the Internet of Things, which relates to the technical field of elevator media management, and comprises a data acquisition module, a data analysis module, an advertisement marking module and an advertisement application module: the data acquisition module is used to collect economic data of the elevator's location, elevator operation data and face data inside the elevator during the elevator operation; the data analysis module comprises a periodic analysis unit and a real-time analysis unit; the technical key points are that when analyzing the information of people inside the elevator, only age and gender data are analyzed, which can be achieved only through face data without comprehensive analysis, and when in use, there is no need to store picture data of people inside the elevator, no need to compare pictures, the overall processed data volume is small, the data processing is fast, it can be applied to areas with large flow of people, the use effect is good, and it has a good use prospect.

Description

Technical Field

[0001] The present invention relates to the technical field of elevator media management, and in particular to an elevator media delivery management system and method based on the Internet of Things. Background Art

[0002] Elevator advertising is a new type of advertising media, which refers to the advertising carrier produced and published on the inner wall of urban building elevators. There are various types of elevator advertising, mainly in business elevators and building elevators, and the main form of expression is the form of photo frames installed in the elevator car.

[0003] With the development of technology, elevator advertising has gradually developed into elevator electronic advertising, that is, installing elevator display screens inside the elevator.

[0004] Elevator electronic advertising is an advertisement shown to people through video when they are riding in an elevator. Compared with print advertising, elevator video advertising has the characteristics of high target audience concentration, mandatory exposure, high advertising reach, and strong visual impact, and can effectively attract the audience's attention in a closed elevator space.

[0005] The existing application publication number is CN110766451B, and the invention patent named "An elevator advertising delivery method and delivery system based on human static tags" records the elevator advertising delivery method, including: S1. Collecting images in the elevator and obtaining face data based on the images; S2. Querying the person ID in the data server based on the face data, and obtaining the attribute label stored in the data server based on the person ID; S3. Sort the value of the people in the elevator according to the attribute label, and obtain the target group with the highest value according to the sorting result; S4. Playing a pre-set advertising category for the target group. According to the present invention, accurate delivery of advertisements in scenes such as elevators with frequent flow of people and small space is achieved.

[0006] However, based on the above content and combined with the prior art, the central solution of the solution recorded in the above patent.

[0007] However, the above solution has major defects in actual use. The above solution mainly obtains face data, compares it with the data stored in the database, determines the attribute label, and then implements accurate advertising according to the attribute label;

[0008] Since it requires a comprehensive analysis of age, gender, clothing and other data, and needs to be compared and identified with existing stored data, it takes a long time to obtain the attributes of people who enter the elevator for the first time, and cannot achieve precise advertising in a short time. At the same time, its use requires the storage of a large number of pictures. Therefore, the overall data is large, and it can only be used for elevators in areas with small population mobility. The overall applicability is not high and does not meet people's usage requirements. For this reason, we have developed an elevator media delivery management system and method based on the Internet of Things. Summary of the invention

[0009] 1. Technical issues to be resolved

[0010] In view of the deficiencies in the prior art, the present invention provides an elevator media delivery management system and method based on the Internet of Things, which solves the problems raised in the background technology.

[0011] (II) Technical solution

[0012] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0013] An elevator media delivery management system based on the Internet of Things includes a data acquisition module, a data analysis module, an advertisement marking module and an advertisement application module:

[0014] A data collection module is used to collect economic data of the elevator's location, elevator operation data, and facial data inside the elevator during the elevator's operation;

[0015] A data analysis module, which includes a periodic analysis unit and a real-time analysis unit;

[0016] The cycle analysis unit obtains economic data according to the set cycle, analyzes the economic data of the elevator location, and generates economic values ;

[0017] The real-time analysis unit obtains the elevator operation data and the face data during the elevator operation when the elevator door closes for the first time and before the advertisement ends, and analyzes the obtained data to obtain the real-time value of the number of people riding the elevator and the single operation time of the elevator;

[0018] Advertisement marking module, used to analyze the advertisements in the advertisement library and assign advertisement filter sets according to the analysis results. The advertisement filter sets include economic values. , advertising duration Ggsc, advertising applicable age and advertising applicable gender, based on economic values Calculate the economic level range and compare the economic values ​​in the ad filter set Whether it is within the economic level range, extract the elevator advertisement list whose economic value is within the economic level range;

[0019] The advertising application module is used to obtain the number of people riding the elevator and the single running time of the elevator, and sort the order in the elevator advertising list according to the number of people riding the elevator, the single running time of the elevator and the number of advertisement playbacks to obtain the final advertising list, and play the advertisement ranked first in the final advertising list.

[0020] Furthermore, the economic data of the location of the elevator include the level of the city where the elevator is located, the echelon of the unit price of the building to which the elevator belongs, and the usage nature of the building to which the elevator belongs;

[0021] Elevator operation data includes the normal operating speed of the elevator, the acceleration of the elevator, the length of time the elevator stays on each floor, and the operation strategy of the elevator.

[0022] Furthermore, the economic value The calculation formula is as follows:

[0023] ;

[0024] In the formula, The usage data of the building to which the elevator belongs. is the city level where the elevator is located, The unit price of the building to which the elevator belongs is in the tier. is the influence coefficient of city level on economic value, is the influence coefficient of the unit price of the building to which the elevator belongs on the economic value, .

[0025] Furthermore, the real-time values ​​of the people taking the elevator include the average age and the gender of the main people;

[0026] The steps to analyze the facial data inside the elevator to obtain the average age and gender of the main population are as follows:

[0027] S1 obtains the age of all people in the elevator based on the face age estimation algorithm that integrates LBP and HOG features;

[0028] S2 calculates the average age based on the number and age of people inside the elevator. The specific calculation formula is as follows:

[0029] ;

[0030] In the formula, is the average age, is the number of people inside the elevator, is the age of the i-th person in the elevator;

[0031] S3 extraction age is in the interval facial data, , and based on the Adaboost+SVM face gender classification algorithm analysis, determine the gender of the corresponding face data;

[0032] S4 counts the number of genders to determine the size of the number of males and females. The gender of the main population is the gender corresponding to the large data.

[0033] If the number of males is equal to the number of females, increase to , = +1, repeat steps S3 and S4 until the number of males is not equal to the number of females; when adjusted to , at this time, the gender is randomly output and the repetitive steps are interrupted.

[0034] Furthermore, the calculation formula for the single running time when the elevator closes its door for the first time is as follows:

[0035] ;

[0036] In the formula, is the single running time of the elevator, The time it takes for the elevator to rise and fall. is the length of time the elevator stays on the floor, The floor difference of this elevator operation. is the floor height of the building to which the elevator belongs, The distance the elevator takes to reach its maximum speed. is the acceleration of the elevator.

[0037] Furthermore, the economic level range is , is the preset fluctuation ratio, , For economic values With economic values The conversion factor between .

[0038] Furthermore, the steps of sorting the order of the elevator advertisement list according to the number of people riding the elevator, the single operation time of the elevator and the number of advertisement playbacks are as follows:

[0039] Obtain gender data of the main population, compare the gender data of the population with the gender applicable to the advertisement, move the order of advertisements with the same attributes forward, and obtain a preliminary adjustment list;

[0040] The adaptability of advertisements with the same attributes is calculated according to the average age, the applicable age of advertisements, the single operation time of the elevator and the advertisement time, and the adaptability of advertisements with the same attributes is further adjusted according to the size of the adaptability to obtain a secondary adjustment list;

[0041] Count the number of times an ad is played within a period of T, and get the time since the last time the ad was played;

[0042] When the number of times the advertisement is played reaches the set upper limit or the duration between the last advertisement and the current advertisement is less than the set interval duration, the advertisement is sorted backward to obtain a final advertisement list.

[0043] Furthermore, the fitness calculation formula is as follows:

[0044] ;

[0045] In the formula, To determine the suitability of the advertisement to the people inside the elevator, For the applicable age of advertisement, is the weight coefficient corresponding to the difference between the average age and the applicable age of the advertisement, For the duration of the ad, is the weight coefficient corresponding to the difference between the advertisement duration and the single operation duration of the elevator, , .

[0046] Furthermore, the real-time analysis unit obtains the data collected by the data collection module μ seconds before the end of the advertisement, and re-analyzes it to obtain the real-time value of the number of people riding the elevator and the single running time of the elevator;

[0047] The real-time analysis unit analyzes the single running time of the elevator to determine whether the elevator is in a stopped state. The calculation formula for the single running time of the elevator in the stopped state is as follows:

[0048] ;

[0049] The calculation formula for the single running time of the elevator when it is in motion is as follows:

[0050] ;

[0051] The advertisement application module analyzes and sorts the results re-analyzed by the real-time analysis unit to obtain a new final advertisement list, and plays the advertisement ranked first in the new final advertisement list.

[0052] Furthermore, a method for managing elevator media delivery based on the Internet of Things includes the following steps:

[0053] Collect economic data of the elevator's location, elevator operation data, and facial data inside the elevator during operation;

[0054] Obtain economic data periodically, analyze the economic data of the elevator location, and generate economic values ;

[0055] Analyze the ads in the ad library and assign ad filter sets based on the analysis results. The ad filter sets include economic values. , advertising duration Ggsc, advertising applicable age and advertising applicable gender, based on economic values Calculate the economic level range and compare the economic values ​​in the ad filter set Whether it is within the economic level range, extract the elevator advertisement list whose economic value is within the economic level range;

[0056] When the elevator door is closed for the first time, the elevator operation data and the face data during the elevator operation are obtained, and the obtained data are analyzed to obtain the real-time value of the number of people riding the elevator and the single operation time of the elevator;

[0057] Obtain the number of people riding the elevator and the single running time of the elevator, and sort the order of the elevator advertisement list according to the number of people riding the elevator, the single running time of the elevator and the number of advertisement playbacks to obtain a final advertisement list, and play the advertisement ranked first in the final advertisement list;

[0058] μ seconds before the end of the advertisement, the elevator operation data and the face data during the elevator operation are obtained, and the obtained data are analyzed to obtain the new real-time value of the number of people riding the elevator and the single operation time of the elevator;

[0059] Obtain new values ​​of the number of people riding the elevator and the single operation duration of the elevator, and sort the order of the elevator advertisement list according to the new values ​​of the number of people riding the elevator, the single operation duration of the elevator and the number of advertisement plays to obtain a new final advertisement list, and play the advertisement ranked first in the new final advertisement list.

[0060] (III) Beneficial effects

[0061] The present invention provides an elevator media delivery management system and method based on the Internet of Things, which has the following beneficial effects:

[0062] 1. The present invention records an elevator media delivery management system based on the Internet of Things. When analyzing the information of people inside the elevator, it only analyzes the age and gender data, which can be achieved only through face data, without the need for comprehensive analysis. Moreover, when in use, it is not necessary to store the image data of people inside the elevator, and it is not necessary to perform image comparison. The overall amount of data processed is small, the data processing is fast, and it can be applied to areas with large traffic volume, has good use effect, and has a good use prospect;

[0063] 2. The present invention records an elevator media delivery and management system based on the Internet of Things, which combines economic data with elevator internal personnel data to make advertising delivery more accurate, and further analyzes the single running time of the elevator, effectively avoiding the situation where the relevant audience inside the elevator leaves the elevator while the advertisement is less than half played, so that the pushed advertisements can be fully presented to the relevant audiences and can be fully displayed, so that the advertising effect is better and has a good prospect for use. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a flowchart of an elevator media delivery management system based on the Internet of Things according to the present invention;

[0065] Figure 2 The present invention is a flow chart of an elevator media delivery management method based on the Internet of Things. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] Research reasons

[0068] The present invention discloses an elevator media delivery management system and method based on the Internet of Things, which mainly aims at obtaining face data during the existing elevator advertising delivery, and then comparing it with the data stored in the database, determining the attribute label, and then realizing accurate delivery of the advertisement according to the attribute label; in the delivery process, since it needs to comprehensively analyze various data such as age, gender, and clothing, and needs to compare and identify with the existing stored data, it takes a long time to acquire the attributes of people who step into the elevator for the first time, and the amount of data processed is relatively large, and it is impossible to realize accurate delivery of advertisements in a short time. At the same time, its use requires the storage of a large number of pictures, and its usage data increases sharply with the increase of the number of people using the elevator. Therefore, the system is prone to jamming when used in the later stage, so it can only be used for elevators in areas with small population mobility, and its overall applicability is not high.

[0069] In addition, the clothing styles of office workers are uniform, and many clothes do not have labels (such as pajamas). Moreover, the prices of clothes of the same brand vary greatly. Therefore, it is impossible to judge the economic situation by directly analyzing the clothes, and the effect is not good. At the same time, the length of advertisements varies. There are cases where people in the elevator only ride a few floors and the time is short, but a longer advertisement is played, which greatly reduces the effectiveness of the advertisement and fails to achieve the desired effect.

[0070] Design ideas

[0071] Most existing systems obtain advertising screening conditions through face and clothing recognition. In this process, clothing recognition is difficult and time-consuming. In addition, clothing is affected by the external seasons, and corresponding advertisements cannot be pushed in a short period of time. It is also prone to lag after being used for a period of time.

[0072] Therefore, during the research and development process, the initial idea was to correct the defects of high difficulty and inaccuracy in clothing recognition, and susceptibility to external environmental influences, and to study relatively stable data that can reflect the economic situation of relevant users. After repeated investigations, discussions and analyses, the easiest, simplest and most intuitive data is the economic attributes of cities and communities.

[0073] However, after a period of experimentation and trial operation, it was accidentally discovered that when people exited the elevator, the advertisement was only halfway played and the brand and product information in the advertisement had just begun to be introduced, resulting in poor overall advertising effect.

[0074] Therefore, in the middle of the research and development, we changed our thinking and added a step to analyze the single operation time of the elevator. We combined the single operation time of the elevator with the advertising time, conducted a comprehensive analysis, and adjusted the plan to make the advertisements play during the operation as much as possible and present the best advertising effect.

[0075] However, after using it for a period of time, it was found that when something happened that required people to take the elevator for a short time, the same advertisement would be played, causing passengers to feel disgusted and disgusted.

[0076] Therefore, in the later stage of research and development, the previous solutions were fully combined to limit the number of times. For specific solutions, see Example 1 and Example 2.

[0077] Example 1

[0078] See also Figure 1 , this embodiment provides an elevator media delivery management system based on the Internet of Things, including a data acquisition module, a data analysis module, an advertisement marking module and an advertisement application module.

[0079] Data acquisition: To achieve accurate elevator advertising, it is necessary to understand the situation of the corresponding personnel inside the elevator and select appropriate advertisements based on the situation of the people inside the elevator. The acquisition of data on the corresponding personnel inside the elevator is based on the data collection module.

[0080] The data acquisition module uses a camera component currently installed on the inner top surface of the elevator car and the front of the advertising machine, which can directly capture the facial data of people entering the elevator.

[0081] The camera component on the front of the advertising machine is mainly used to collect data on the reactions of people inside the elevator to the advertisements, making it easier for subsequent staff to further screen the advertising content.

[0082] The camera component on the inner top surface of the elevator car uses a 360-degree panoramic camera, which can capture high-definition images of people inside the elevator and collect facial data of people inside the elevator.

[0083] The data collection module is used to collect the economic data of the elevator's location, the elevator's operation data, and the face data inside the elevator during the elevator's operation;

[0084] Economic data can be obtained directly through government websites and existing real estate sales websites without the need for separate analysis, which can greatly reduce the workload when the system is running.

[0085] Elevator operation data can be obtained from the existing elevator operating system without the need to install separate sensors for auxiliary detection.

[0086] The collected data can reflect the main mobile personnel situation at the location. Advertising can be placed based on the situation of the main mobile personnel, making the advertising effect better, rather than comprehensively collecting the data of a single person and then placing advertisements after comprehensive analysis.

[0087] Data processing

[0088] After the data is collected, it needs to be further processed and analyzed to obtain the desired parameters through the data. This process is based on the data analysis module.

[0089] The data analysis module includes a periodic analysis unit and a real-time analysis unit;

[0090] The cycle analysis unit obtains economic data according to the set cycle, analyzes the economic data of the elevator location, and generates economic values ;

[0091] The periodic analysis unit is analyzed once at a certain interval, usually one quarter, half a year or one year.

[0092] Since economic figures do not fluctuate much in a short period of time, there is no need for real-time analysis. It is sufficient to analyze them once at regular intervals. Economic data can reflect people's consumption concepts and levels.

[0093] Elevator operation data includes the normal operating speed of the elevator, the acceleration of the elevator, the length of time the elevator stays on each floor, and the operation strategy of the elevator.

[0094] Economic Values The calculation formula is as follows:

[0095] ;

[0096] In the formula, The usage data of the building to which the elevator belongs. is the city level where the elevator is located, The unit price of the building to which the elevator belongs is in the tier. is the influence coefficient of city level on economic value, is the influence coefficient of the unit price of the building to which the elevator belongs on the economic value, .

[0097] The value range is 0.5-0.8, The value range is 0.2-0.5.

[0098] After repeated calculations, it is found that cities at the same level will still have certain differences. After testing, under normal circumstances Take 0.7, Take 0.4.

[0099] Currently, city levels are divided into 1-4 levels. This patent directly uses this division and can also further divide the relationship between first-tier cities and new first-tier cities, which is now 5 levels.

[0100] The economic data of the elevator's location include the city level where the elevator is located (there is an official report every year), the echelon of the unit price of the building to which the elevator belongs (each city is divided into 4 levels. For example, if the highest housing price in a city is 40,000 yuan per square meter, then the housing prices between 0-10,000 yuan per square meter are in the 4th level, the housing prices between 10,000-20,000 yuan per square meter are in the 3rd level, the housing prices between 20,000-30,000 yuan per square meter are in the 2nd level, and the housing prices between 30,000-40,000 yuan per square meter are in the 1st level), and the use nature of the building to which the elevator belongs (mainly commercial or civil, commercial use nature data Data on usage characteristics greater than that of civil use , data on the nature of civil use Select 1, commercial use data Data on normal civilian use 1.2 times of that of the city);

[0101] The real-time analysis unit obtains the elevator operation data and the face data during the elevator operation when the elevator door closes for the first time and before the advertisement ends, and analyzes the obtained data to obtain the real-time value of the number of people riding the elevator and the single operation time of the elevator;

[0102] The first closing of the elevator door is the time when the elevator moves from the standby state to the use state, moves to the corresponding floor, and closes the door after people enter.

[0103] This method prevents the real-time analysis unit from starting when the elevator is stationary, eliminating the need for analysis and consuming less energy.

[0104] The real-time data of people riding the elevator include the average age and gender of the main group;

[0105] The steps to analyze the facial data inside the elevator to obtain the average age and gender of the main population are as follows:

[0106] S1 obtains the age of all people in the elevator based on the face age estimation algorithm that integrates LBP and HOG features;

[0107] S2 calculates the average age based on the number and age of people inside the elevator. The specific calculation formula is as follows:

[0108] ;

[0109] In the formula, is the average age, is the number of people inside the elevator, is the age of the i-th person in the elevator;

[0110] For example, there are 5 people in the elevator. After analysis, the ages of all the people in the elevator are 36, 22, 28, 42, and 29 respectively. The average age is calculated It is 31.4.

[0111] S3 extraction age is in the interval facial data, And based on the Adaboost+SVM face gender classification algorithm analysis, the gender of the corresponding face data is determined;

[0112] In general Pick , but when the average age is less than 10 or more than 60, it needs to be adjusted. Pick , when the average age is above 60 Pick , so as to avoid the scope being too large or too small, resulting in too many or too few people, affecting the analysis results.

[0113] The above interval is .

[0114] S4 counts the number of genders to determine the size of the number of males and females. The gender of the main population is the gender corresponding to the large data.

[0115] For example, in the above interval, the number of males is 2 and the number of females is 1. At this time, the number of males is greater than the number of females, so the main gender of the population is male.

[0116] If the number of males is equal to the number of females, increase to , = +1, repeat steps S3 and S4 until the number of males is not equal to the number of females; when adjusted to , at this time, the gender is randomly output and the repetitive steps are interrupted.

[0117] For example, there are 4 people in the elevator. After analysis, the ages of all the people in the elevator are 22, 28, 42, and 29 respectively. The average age is calculated It is 30.25.

[0118] The interval is , at this time, the two people in the interval are a man and a woman, and it is impossible to judge at this time. Therefore, it is necessary to expand the range. After 3 adjustments, the interval is At this time, there are 3 people who can judge the number of men and women.

[0119] For example, there are 4 people in the elevator. After analysis, the ages of all the people in the elevator are 15, 28, 48, and 29 respectively. The average age is calculated is 30.

[0120] The interval is , at this time, the two people in the interval are a man and a woman, and it is impossible to judge at this time, so it is necessary to expand the range. After 4 adjustments, the interval is , at this time there are still 2 people, and at this time Therefore, it is not within the adjustment range and a gender is randomly selected for output (it can also be set to female, as female consumption power is higher than male).

[0121] The face age estimation algorithm based on the fusion of LBP and HOG features and the face gender classification algorithm based on Adaboost+SVM are relatively common algorithms in the prior art for identifying age and gender based on face data. Since they are prior art, this patent only applies them and does not improve them. Therefore, this patent does not make too much description on the algorithm part.

[0122] The calculation formula for the single running time when the elevator closes its door for the first time is as follows:

[0123] ;

[0124] In the formula, is the single running time of the elevator, The time it takes for the elevator to rise and fall. is the length of time the elevator stays on the floor, The floor difference of this elevator operation. is the floor height of the building to which the elevator belongs, The distance the elevator takes to reach its maximum speed. is the acceleration of the elevator.

[0125] For example, the acceleration of an elevator is 0.25 m / s 2 , the floor height is 3 meters, the distance the elevator takes to reach its maximum speed is 8 meters, and the elevator stays for 15 seconds.

[0126] Single operation duration when running on the 4th floor Second.

[0127] Single operation duration when running on the 10th floor Second.

[0128] When judging the operating floor, the operating strategy of the elevator door closing time is received, and the operating strategy at the door closing time is the floor to which the elevator descends.

[0129] For example: There are 2 people going downstairs on the 30th floor. One person enters the elevator and presses the button for the 20th floor, and the other presses the button for the 1st floor. There is a pedestrian going downstairs on the 26th floor and presses the down button outside the elevator. At this time, the operation strategy at the door closing moment is to move to the 26th floor to open the door, which means going down 4 floors.

[0130] When the advertisement is playing, there is a pedestrian going downstairs on the 27th floor and pressing the down button outside the elevator. At this time, the operation strategy is to move to the 27th floor and open the door. This does not belong to the operation strategy of closing the door for the first time and the operation strategy μ seconds before the end of the advertisement, so it is not analyzed.

[0131] This step ensures the continuity of advertisement broadcasting and ensures that the advertisement can be broadcast completely, while avoiding over-analysis. It can greatly reduce the amount of data calculation and has good use effect.

[0132] The real-time analysis unit obtains the data collected by the data collection module μ seconds before the end of the advertisement, and re-analyzes it to obtain the real-time value of the number of people riding the elevator and the single running time of the elevator;

[0133] μ is mainly set according to the system's computing speed, signal transmission speed and control speed. The value range of μ is 3-5 seconds, and 3 seconds is normally taken.

[0134] When obtaining data μ seconds before the end of the advertisement, it is determined whether there is a person. If there is no face data or the single running time of the elevator is calculated to be 0, that is, there is no one in the elevator, there is no need to play the advertisement, and the next closing of the elevator door is regarded as the first closing.

[0135] The real-time analysis unit analyzes the single running time of the elevator to determine whether the elevator is in a stopped state. The calculation formula for the single running time of the elevator in the stopped state is as follows:

[0136] ;

[0137] This situation is caused by pressing the elevator button outside the elevator during operation, which causes the elevator operation strategy to change.

[0138] For example, when the elevator descends to the 15th floor, there are 3 seconds left for the advertisement. The facial data and elevator operation data collected by the data collection module are obtained 3 seconds in advance. After further analysis, the operation strategy of the elevator is to descend to the 1st floor.

[0139] At this time, calculate the single running time of the elevator Second.

[0140] Since we don't know how long the elevator will stay, the calculated single running time of the elevator is too long, and the corresponding advertising time is too long. The elevator will be in operation during the next analysis, and the advertising time will be corrected accordingly.

[0141] It is also possible to use a photosensitive sensor and a calculator to count when the elevator stops. When calculating the duration of a single run of the elevator, the timer can be removed (for example, if the elevator stops for 10 seconds when collecting data, the single run duration of the elevator is calculated to be 47.5 seconds). However, the need to further install supporting devices will increase costs. Therefore, the implementation of this solution is still under discussion, but after testing, the solution can be implemented.

[0142] The calculation formula for the single running time of the elevator when it is in motion is as follows:

[0143] ;

[0144] For example: there are two people going downstairs on the 30th floor. One person enters the elevator and presses the button for the 20th floor, and the other presses the button for the 1st floor. At this time, the operation strategy at the closing moment is to move to the 20th floor to open the door, which is to go down 10 floors. During this process, no one interferes. After analysis, the duration of the advertisement played is a 45-second advertisement with a single operation duration of 38 seconds, which is closest to the 10th floor (advertisements are usually 15 seconds, 30 seconds, 45 seconds and 60 seconds). The face data and elevator operation data collected by the data acquisition module are obtained 3 seconds in advance. After further analysis, the operation strategy time of the elevator is obtained as 42 seconds of the advertisement playing. At this time, the operation strategy of the elevator is to run from the 18th floor (the elevator descended 4 meters from 38 to 42 seconds, and descended 2 floors from the floor) to the 1st floor, descending 18 floors.

[0145] Recalculate the single running time of the elevator The acceleration phase of the elevator is shortened, so the calculated duration is longer, but the actual running distance is two meters more, and this time is partially offset by the acceleration phase. Therefore, the calculated duration difference normally does not exceed 3 seconds, which is relatively accurate and allows people to grasp the core content of the advertisement.

[0146] Since people can only be accurately identified when they enter the elevator, the previous advertisement is designed to be played until the elevator door closes. The load design has no effect on longer advertisements, but has a certain impact on advertisements of 30 seconds or less. Therefore, the subsequent improvement direction is to set up a separate push plan for short advertisements.

[0147] The advertisement application module analyzes and sorts the results re-analyzed by the real-time analysis unit (the advertisement application module is started after the real-time analysis unit analyzes), obtains a new final advertisement list, and plays the advertisement ranked first in the new final advertisement list.

[0148] For this purpose, a 60-second advertisement may be replaced by a 15-second advertisement and a 45-second advertisement.

[0149] When no facial data is detected or the single running time of the elevator is calculated to be 0, the elevator is in a stationary state. At this time, no analysis or advertising is required, which can save energy.

[0150] Advertisement marking module, used to analyze the advertisements in the advertisement library and assign advertisement filter sets according to the analysis results. The advertisement filter sets include economic values. , advertising duration Ggsc, advertising applicable age and advertising applicable gender, based on economic values Calculate the economic level range and compare the economic values ​​in the ad filter set Whether it is within the economic level range, extract the elevator advertisement list whose economic value is within the economic level range.

[0151] The ad marking module is activated only when a new ad is received and the price of the product within the ad is adjusted.

[0152] The applicable age and gender of advertisements are obtained directly from advertisers. Advertisers will have statistics on this aspect, so you can directly contact the merchants to obtain the relevant information.

[0153] For example, alcohol advertisements are suitable for men aged 20-60.

[0154] The economic level range is , is the preset fluctuation ratio, , The value range of is the more appropriate range obtained after computer simulation. For economic values With economic values The conversion factor between .

[0155] Economic values ​​in the ad filter set It is the ratio of commodity price to average price.

[0156] For example, in alcohol advertising, the average price of alcohol is 150 yuan. There are three alcohol advertisements, and the prices of alcohol in the three advertisements are 100 yuan, 200 yuan, and 300 yuan respectively. The economic value of the last alcohol is The economic value of the first two wines is 300÷150=2. is 0.67 and 1.34, and the economic level range corresponding to a certain elevator is , then the elevator advertising list will only include an advertisement for the third wine.

[0157] The advertising application module is used to obtain the number of people riding the elevator and the single running time of the elevator, and sort the order in the elevator advertising list according to the number of people riding the elevator, the single running time of the elevator and the number of advertisement playbacks to obtain the final advertising list, and play the advertisement ranked first in the final advertising list.

[0158] The steps to sort the order of the elevator advertisement list according to the number of people riding the elevator, the single operation time of the elevator and the number of advertisement playbacks are as follows:

[0159] Obtain gender data of the main population, compare the gender data of the population with the applicable gender of the advertisement, and move the order of advertisements with the same attributes forward to obtain a preliminary adjustment list;

[0160] The order of advertisements with different attributes is moved back as a whole, without considering sorting, or advertisements with the same attributes are directly selected to form a preliminary adjustment list.

[0161] The adaptability of advertisements with the same attributes is calculated according to the average age, the applicable age of advertisements, the single operation time of the elevator and the advertisement time, and the adaptability of advertisements with the same attributes is further adjusted according to the size of the adaptability to obtain a secondary adjustment list;

[0162] The formula for calculating the degree of fitness is as follows:

[0163] ;

[0164] In the formula, To determine the suitability of the advertisement to the people inside the elevator, For the applicable age of advertisement, is the weight coefficient corresponding to the difference between the average age and the applicable age of the advertisement, For the duration of the ad, is the weight coefficient corresponding to the difference between the advertisement duration and the single operation duration of the elevator, , .

[0165] Filter ads by distance difference method.

[0166] Count the number of times an ad is played within a period of T, and get the time since the last time the ad was played;

[0167] Most advertisements are played according to the number of times and are played within a period of time. Therefore, the general period T is one day, and it is determined whether the advertisement broadcast on that day exceeds the upper limit.

[0168] When the number of times the advertisement is played reaches the set upper limit or the time between the last advertisement and the current advertisement is less than the set interval, the advertisement is sorted backward to obtain a final advertisement list.

[0169] For example, for a certain advertisement, the upper limit of the number of times it is played per day is set to 20 times, and the interval is set to 5 minutes (that is, it will only be played once within 5 minutes to avoid excessive playback in a short period of time, which will result in insufficient publicity and cause people in the elevator to see the same advertisement repeatedly and feel bored).

[0170] After the ad is played, it will be temporarily marked and the end time of the play will be recorded. During the marking period, the ad will no longer be considered when the ad list is generated.

[0171] When the number of plays reaches the set upper limit, the ad will be marked as a period, and will no longer be considered when subsequent ad lists are generated within the period T.

[0172] The weight coefficient is determined by the coefficient of variation method, which is a method of weighting each indicator according to the degree of variation between the current value of each evaluation indicator and the target value. If the numerical difference of an indicator is large, it can clearly distinguish the evaluated objects, indicating that the indicator has rich discrimination information, and thus the indicator should be given a larger weight. On the contrary, if the numerical difference of each evaluated object on a certain indicator is small, then the ability of this indicator to distinguish the evaluation objects is weak, and thus the indicator should be given a smaller weight. This method directly uses the information contained in each indicator and obtains the weight of the indicator through calculation, so it is objective.

[0173] The present invention records an elevator media delivery management system based on the Internet of Things, which only analyzes age and gender data when analyzing the information of people inside the elevator. This can be achieved only through facial data, without the need for comprehensive analysis. Moreover, when in use, there is no need to store picture data of people inside the elevator, and no need to perform picture comparison. The overall amount of data processed is small, the data processing is fast, and it can be applied to areas with large traffic flow, has good use effect, and has a good prospect for use.

[0174] The present invention records an elevator media delivery management system based on the Internet of Things, which combines economic data with elevator internal personnel data to make advertising delivery more accurate, and further analyzes the single running time of the elevator, effectively avoiding the situation in which the relevant audience inside the existing elevator leaves the elevator while the advertisement is less than half played, so that the pushed advertisement can be fully presented to the relevant audience and can be fully displayed, so that the advertising effect is better and has a good prospect of use.

[0175] Example 2

[0176] See also Figure 2 Based on Example 1, this embodiment also provides an elevator media delivery management method based on the Internet of Things, comprising the following steps:

[0177] Collect economic data of the elevator's location, elevator operation data, and facial data inside the elevator during operation;

[0178] Obtain economic data periodically, analyze the economic data of the elevator location, and generate economic values ;

[0179] Analyze the ads in the ad library and assign ad filter sets based on the analysis results. The ad filter sets include economic values. , advertising duration Ggsc, advertising applicable age and advertising applicable gender, based on economic values Calculate the economic level range and compare the economic values ​​in the ad filter set Whether it is within the economic level range, extract the elevator advertisement list whose economic value is within the economic level range;

[0180] When the elevator door closes for the first time, the elevator operation data and the face data during the elevator operation are obtained, and the obtained data is analyzed to obtain the real-time value of the number of people riding the elevator and the single operation time of the elevator;

[0181] Obtain the number of people riding the elevator and the single running time of the elevator, and sort the order of the elevator advertisement list according to the number of people riding the elevator, the single running time of the elevator and the number of advertisement playbacks to obtain a final advertisement list, and play the advertisement ranked first in the final advertisement list;

[0182] The steps to sort the order of the elevator advertisement list according to the number of people riding the elevator, the single operation time of the elevator and the number of advertisement playbacks are as follows:

[0183] S1 obtains the age of all people in the elevator based on the face age estimation algorithm that integrates LBP and HOG features;

[0184] S2 calculates the average age based on the number and age of people inside the elevator. The specific calculation formula is as follows:

[0185] ;

[0186] In the formula, is the average age, is the number of people inside the elevator, is the age of the i-th person in the elevator;

[0187] S3 extraction age is in the interval facial data, , and based on the Adaboost+SVM face gender classification algorithm analysis, determine the gender of the corresponding face data;

[0188] S4 counts the number of genders to determine the size of the number of males and females. The gender of the main population is the gender corresponding to the large data.

[0189] If the number of males is equal to the number of females, increase to , = +1, repeat steps S3 and S4 until the number of males is not equal to the number of females; , at this time, the gender is randomly output and the repetition step is interrupted;

[0190] μ seconds before the end of the advertisement, the elevator operation data and the face data during the elevator operation are obtained, and the obtained data are analyzed to obtain the new real-time value of the number of people riding the elevator and the single operation time of the elevator;

[0191] Obtain new values ​​of the number of people riding the elevator and the single operation duration of the elevator, and sort the order of the elevator advertisement list according to the new values ​​of the number of people riding the elevator, the single operation duration of the elevator and the number of advertisement plays to obtain a new final advertisement list, and play the advertisement ranked first in the new final advertisement list.

[0192] The above formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0193] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0194] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0195] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. An elevator media delivery management system based on the Internet of Things, characterized in that: include: A data collection module is used to collect economic data of the elevator's location, elevator operation data, and facial data inside the elevator during the elevator's operation; A data analysis module, which includes a periodic analysis unit and a real-time analysis unit; The cycle analysis unit obtains economic data according to the set cycle, analyzes the economic data of the elevator location, and generates economic values ; The economic data of the elevator location include the city level where the elevator is located, the unit price echelon of the building to which the elevator belongs, and the use nature of the building to which the elevator belongs; Elevator operation data includes the normal operating speed of the elevator, the acceleration of the elevator, the length of time the elevator stays on each floor, and the operation strategy of the elevator; The real-time analysis unit obtains the elevator operation data and the face data during the elevator operation when the elevator door closes for the first time and before the advertisement ends, and analyzes the obtained data to obtain the real-time value of the number of people riding the elevator and the single operation time of the elevator; The real-time values ​​of the people taking the elevator include the average age and the gender of the main people; The steps to analyze the facial data inside the elevator to obtain the average age and gender of the main population are as follows: S1 obtains the age of all people in the elevator based on the face age estimation algorithm that integrates LBP and HOG features; S2 calculates the average age based on the number and age of people inside the elevator. The specific calculation formula is as follows: ； In the formula, is the average age, is the number of people inside the elevator, is the age of the i-th person in the elevator; S3 extraction age is in the interval facial data, , and based on the Adaboost+SVM face gender classification algorithm analysis, determine the gender of the corresponding face data; S4 counts the number of genders to determine the size of the number of males and females. The gender of the main population is the gender corresponding to the large data. If the number of males is equal to the number of females, increase to , = +1, repeat steps S3 and S4 until the number of males is not equal to the number of females; when adjusted to , at this time, the gender is randomly output and the repetition step is interrupted; The calculation formula for the single running time of the elevator when the door is closed for the first time is as follows: ； In the formula, is the single running time of the elevator, The time it takes for the elevator to ascend or descend. is the length of time the elevator stays on the floor, is the floor difference of this elevator operation, is the floor height of the building to which the elevator belongs, The distance the elevator takes to reach its maximum speed. is the acceleration of the elevator; Advertisement marking module, used to analyze the advertisements in the advertisement library and assign advertisement filter sets according to the analysis results. The advertisement filter sets include economic values. , advertising duration Ggsc, advertising applicable age and advertising applicable gender, based on economic values Calculate the economic level range and compare the economic values ​​in the ad filter set Whether it is within the economic level range, extract the elevator advertisement list whose economic value is within the economic level range; The advertising application module is used to obtain the number of people riding the elevator and the single running time of the elevator, and sort the order in the elevator advertising list according to the number of people riding the elevator, the single running time of the elevator and the number of advertisement playbacks to obtain the final advertising list, and play the advertisement ranked first in the final advertising list.

2. According to the Internet of Things-based elevator media delivery management system of claim 1, it is characterized by: The economic values The calculation formula is as follows: ； In the formula, The usage data of the building to which the elevator belongs. is the city level where the elevator is located, The unit price of the building to which the elevator belongs is in the tier. is the influence coefficient of city level on economic value, is the influence coefficient of the unit price of the building to which the elevator belongs on the economic value, .

3. According to the Internet of Things-based elevator media delivery management system of claim 2, it is characterized by: The economic level range is , is the preset fluctuation ratio, , For economic values With economic values The conversion factor between .

4. According to claim 3, an elevator media delivery management system based on the Internet of Things is characterized in that: The steps to sort the order of the elevator advertisement list according to the number of people riding the elevator, the single operation time of the elevator and the number of advertisement playbacks are as follows: Obtain gender data of the main population, compare the gender data of the population with the gender applicable to the advertisement, move the order of advertisements with the same attributes forward, and obtain a preliminary adjustment list; The adaptability of advertisements with the same attributes is calculated according to the average age, the applicable age of advertisements, the single operation time of the elevator and the advertisement time, and the adaptability of advertisements with the same attributes is further adjusted according to the size of the adaptability to obtain a secondary adjustment list; Count the number of times an ad is played within a period of T, and get the time since the last time the ad was played; When the number of times the advertisement is played reaches the set upper limit or the time between the last advertisement and the current advertisement is less than the set interval, the advertisement is sorted backward to obtain a final advertisement list.

5. According to claim 4, an elevator media delivery management system based on the Internet of Things is characterized in that: The formula for calculating the degree of fitness is as follows: ； In the formula, To determine the suitability of the advertisement to the people inside the elevator, For the applicable age of advertisement, is the weight coefficient corresponding to the difference between the average age and the applicable age of the advertisement, is the duration of the advertisement, is the weight coefficient corresponding to the difference between the advertisement duration and the single operation duration of the elevator, , .

6. According to the Internet of Things-based elevator media delivery management system of claim 5, it is characterized by: The real-time analysis unit obtains the data collected by the data collection module μ seconds before the end of the advertisement, and re-analyzes it to obtain the real-time value of the number of people riding the elevator and the single running time of the elevator; The real-time analysis unit analyzes the single running time of the elevator to determine whether the elevator is in a stopped state. The calculation formula for the single running time of the elevator in the stopped state is as follows: ； The calculation formula for the single running time of the elevator when it is in motion is as follows: ； The advertisement application module analyzes and sorts the results re-analyzed by the real-time analysis unit to obtain a new final advertisement list, and plays the advertisement ranked first in the new final advertisement list.

7. An elevator media delivery management method based on the Internet of Things, using any system described in claims 1 to 6, characterized in that: The steps include: Collect economic data of the elevator's location, elevator operation data, and facial data inside the elevator during operation; Obtain economic data periodically, analyze the economic data of the elevator location, and generate economic values ; Analyze the ads in the ad library and assign ad filter sets based on the analysis results. The ad filter sets include economic values. , advertising duration Ggsc, advertising applicable age and advertising applicable gender, based on economic values Calculate the economic level range and compare the economic values ​​in the ad filter set Whether it is within the economic level range, extract the elevator advertisement list whose economic value is within the economic level range; The elevator operation data and face data during the operation of the elevator are obtained when the elevator door closes for the first time and before the advertisement ends. The obtained data is analyzed to obtain the real-time value of the number of people riding the elevator and the single operation duration of the elevator. The number of people riding the elevator and the single running time of the elevator are obtained, and the order of the elevator advertisement list is sorted according to the number of people riding the elevator, the single running time of the elevator and the number of advertisement plays to obtain a final advertisement list, and the advertisement ranked first in the final advertisement list is played.

Images