TWI490803B - Methods and system for monitoring of people flow - Google Patents

Description

Human flow monitoring advertising method and system thereof

An advertisement flow monitoring technology, in particular, a method and system for monitoring a flow of people who actively follow a face by integrating a design pattern.

The ad flow monitoring technology is proposed to solve the difficulty of traditional advertising efficiency calculation, and it is not easy to evaluate the types of viewers who are interested in advertising. Traditional advertising flow monitoring technology uses human resources for interviews and questionnaires, and requires a large number of human visits and questionnaires. However, due to the large number of people who refuse to access and reject the questionnaire, not only will the complexity of the statistics be increased, but also the results of statistical distortion will be derived.

Therefore, there is now an advertising and broadcasting system that uses push-pull technology to apply to human flow identification. It mainly uses the camera to record in the advertisement billboard or the number of people in the calculation, and then through the flow identification system to calculate and automatically identify whether it is a human face, and can lock The face and the current number of people are calculated, and the information is stored in the database. The number of people is counted by the flow of people, and the information is actively transmitted to other users; thus, it is convenient to count the number of people and automatically recognize the face of the person. At the same time, it can actively transmit information to other users, so as to facilitate the counting of statistics and automatically recognize the function of recording the face.

Another active intelligent flow monitoring system can use smart point identification technology, business intelligence analysis data and data exploration using point-of-sale servers, cameras, advertising machines, shelf advertising machines, electronic billboards and servers. The camera of the advertising machine, the shelf advertising machine and the electronic billboard automatically captures the customer image data, and the camera lens that detects the flow of people also transmits the number of people to the server, and the smart identification technology is used in the server to automatically identify the customer's face, Gender and age, and the appropriate ads are sent to the advertising machine for viewing to increase sales performance, and then the data is sent to the server to store intelligent attribute identification data and consumer characteristics and other multi-dimensional benefit analysis. This prior art can play information on specific products for individual consumers, making the advertisements become customized intelligent marketing methods, and increasing the customer's consumption motivation by collecting and cross-analyzing the consumer's propensity to consume.

However, regardless of the prior art, for the viewing customer, the customer does not clearly feel the relevance of the content of the advertisement to itself, causing the customer to quickly pass the photographic lens. In fact, perhaps the audience does not face the advertising content at all, and the conventional technology cannot Let customers interact with ads. Therefore, the prior art relies only on face recognition, which leads to the same distortion of the flow statistics, and the aforementioned conventional technology cannot greatly improve the advertising efficiency.

The invention provides a method and system for monitoring a person flow monitoring, which mainly uses a camera to record an image on an advertisement billboard, and then through the calculation and automatic identification of whether it is a human face, and can lock the face and calculate the current number of people. According to the information, the information is stored in the database. Then, the preset design pattern is automatically transmitted on the designated person's face, and the advertising requirement integrated design pattern actively follows the face presentation, thereby achieving the purpose of interacting with the human face on the image, so the invention can not only integrate the flow monitoring and management, but also can Improve your advertising performance.

One aspect of the present invention provides a method for monitoring a person flow monitoring. The method for monitoring a person flow includes the following steps: an advertisement recording step, a face recognition step, a flow statistics step, and a mask feedback step.

According to an embodiment of the present invention, the method for monitoring the flow of people includes the following steps: the step of advertising is to use a camera to record an image on an advertisement billboard, and the image is presented by an advertisement billboard; the face recognition step is automatically performed. Identify whether the image has at least one face, and lock the face and calculate the data of the face; the flow statistics step stores and counts the data; the mask feedback step uses a design pattern to follow the display on the designated person's face, and then integrates the design pattern with the data. Interact with the faces on the image.

According to another embodiment of the present invention, in the foregoing advertisement recording step, face recognition step, human flow statistics step, and mask feedback step, a mask changing step may be further included, and the complex design pattern is used to match the face data. Take turns or appear on the person's face at the same time. In the face recognition step, the image is subtracted from the subtraction operation to take out a human form, and then converted by RGB, and calculated as edge capture, and then the face is taken out through image compression, and the ratio calculation or genetic algorithm is used to find the person. face.

In addition, a report output step may be included, and the face feedback data and the mask change step are used to calculate and collect various face related data, for example, calculating the number, time and size of the face for viewing the advertisement billboard, and outputting one person. Flow monitoring report. The report output step can use the face data calculation statistics, and output a one-person flow monitoring report.

Another aspect of the present invention provides a human flow monitoring advertising system that implements the human flow monitoring advertising method of the present invention in conjunction with network and electronic computing technologies.

According to an embodiment of the system aspect of the present invention, a human flow monitoring advertisement system includes: an advertisement billboard for viewing by a user. A camera is placed on the billboard and the camera captures an image. A display is disposed on the advertising billboard and presents an image on the display. The one-person flow identification module is connected to the camera, and the flow recognition module calculates and automatically recognizes at least one face in the image, and the flow recognition module can lock the face. The one-person flow statistics module is connected to the human flow identification module, and the human flow statistics module calculates a data according to the human face. A database is connected to the flow statistics module, the database stores at least one design pattern, and the database stores data provided by the flow statistics module. And a facial mask module is connected to the camera, the display, the flow recognition module, the flow statistics module and the database. The facial mask module takes the design pattern from the database according to the face and the data, and the facial mask module order The design pattern follows the face of the person displayed in the image. In this way, the flow monitoring advertising system can integrate the advertising requirements into a preset design pattern to actively follow the face presentation, and achieve the purpose of interacting with the human face on the image.

According to another embodiment of the system aspect of the present invention, the plurality of design patterns can be stored in the database, and the facial mask module alternates the design patterns in accordance with the data or simultaneously displays the faces of the designated persons displayed in the images.

The foregoing flow statistics module calculates the number, time and size of the faces of the advertisement billboard, and stores them in the database according to the preset classification items. In addition, the aforementioned database can be synchronized with a cloud server for cloud synchronization.

100-400‧‧‧Steps

410, 420‧‧‧ design patterns

501‧‧‧ camera

502‧‧‧Advertising board

503‧‧‧ display

5031‧‧ images

504‧‧‧Accident Identification Module

505‧‧‧Civil Flow Statistics Module

506‧‧‧Database

507‧‧‧Face mask module

508‧‧‧Cloud Server

D‧‧‧ data

F‧‧‧Face

FIG. 1 is a diagram showing the steps of an embodiment of the method for monitoring the flow of the present invention.

FIG. 2 is a schematic diagram showing a system embodiment of an embodiment of the present invention.

FIG. 3 is a schematic diagram showing the display image of the human flow monitoring advertising system of the present invention.

Please refer to FIG. 1 to FIG. 3 , which is a step diagram of an embodiment of the method for monitoring the flow of people in the present invention. Figure 2 is a schematic diagram of an embodiment of an embodiment of a flow monitoring advertising system. Figure 3 also shows a schematic image of the display of the flow monitoring advertising system.

The present invention provides a method for monitoring a person flow monitoring comprising the following steps: an advertisement recording step 100, a face recognition step 200, a person flow statistics step 300, and a mask feedback step 400. The advertisement recording step 100 uses a camera 501 to record an image on the advertisement billboard 502, and presents the image 5031 by the advertisement billboard 502. The face recognition step 200 subtracts the image 5031 by subtraction to take out a human figure, and then converts the image by RGB. And the operation is used for edge capture, and the shape of the face is taken out through image compression, and the face F is found through the ratio calculation or the genetic algorithm. Therefore, it is possible to automatically recognize whether the image 5031 has at least one face F and lock the face F and The data D of the face F is calculated; the flow statistics step 300 stores and counts the data D; the mask feedback step 400 follows the display on the designated face F using a design pattern 410, and the design pattern 410 is further displayed. The integrated data interacts with the face F on the image 5031. Accordingly, in the operation of the advertisement recording step 100, the face recognition step 200, the flow statistics step 300, and the mask feedback step 400, the present invention uses the camera 501 to record the image 5031 on the advertisement billboard 502, and then through the calculation and automatic identification. Whether it is a face F, and the face F can be locked and the current flow data D is calculated, and the data D information is stored in the database 506. Then, the preset design pattern 410 is automatically transmitted on the designated face F, and then the advertisement demand integration design pattern 410 is actively followed by the face F of the viewing advertisement billboard 502, and the face F of the viewer on the image 5031 is achieved. The purpose of the interaction is that the method of the invention can not only integrate the flow monitoring and management, but also improve the advertising effect.

According to another embodiment of the present invention, the foregoing method for monitoring the flow of people may further include a mask changing step. Please refer to the schematic image of the display of the flow monitoring advertising system shown in FIG. This step can be performed on the face F by using the second design pattern 420 in conjunction with the face data D or simultaneously with the first design pattern 410. In order to achieve more functional advertising effects, the second design pattern 420 may be a crown design pattern, and will be displayed on the face data D is closest to the advertisement billboard. 502 face F.

The human flow monitoring advertising method provided by the present invention may further comprise a report output step (not shown), using the mask feedback step and the mask changing step to calculate statistics, and outputting a one-person flow monitoring report. The report output step can also use the face data calculation statistics, and output a one-person flow monitoring report.

Referring again to FIG. 2, in accordance with an embodiment of the system aspect of the present invention, a flow monitoring system includes: an advertising billboard 502 for viewing by a user. A camera 501 is disposed on the advertising billboard 502 and photographed The machine 501 records an image 5031. A display 503 is disposed on the advertising billboard 502, and the image 5031 is presented by the display 503. The one-person flow identification module 504 is connected to the camera 501, and the flow recognition module 504 calculates and automatically recognizes at least one face F of the image 5031, and the flow recognition module 504 can lock the face F. The one-person flow statistics module 505 is connected to the flow identification module 504, and the flow statistics module 505 calculates a data D based on the face F. A database 506 is coupled to the flow statistics module 505, the database 506 stores the plurality of design patterns 410, 420, ..., and the database 506 stores the data D provided by the flow statistics module 505. The face mask module 507 is connected to the camera 501, the display 503, the flow identification module 504, the flow statistics module 505, and the database 506. The facial mask module 507 has a different design pattern 410 according to the face F and the data D. Or 420 is taken from the database 506, and the face mask module 507 causes each design pattern 410 or 420 to follow the face F displayed in the image 5031. Thereby, the flow monitoring advertising system can integrate the advertising requirement into a preset design pattern 410 or 420 to actively follow the face F presentation, and achieve the purpose of interacting with the face F of the viewing billboard 502 on the image 5031.

It is worth mentioning that the foregoing database 506 can store the complex design patterns 410, 420, ..., and the facial mask module 507 displays the complex design patterns 410, 420, ... according to the data D, in turn or simultaneously. The designated face F in 5031. The foregoing flow statistics module 504 calculates the number, time and size of the face F of the advertisement billboard 502, and stores them in the database 506 according to the preset classification items. The foregoing database 506 can perform cloud synchronization with a cloud server 508.

In summary, the present invention provides a method for monitoring pedestrian flow and Its system has the following advantages:

1. It has the purpose of integrating the design pattern of advertising requirements to actively follow the face presentation, and achieves the purpose of interacting with the human face on the image. Therefore, the invention can integrate the flow monitoring management to improve the advertising effect.

2. It has the function of performing specific flow statistics and advertisement evaluation for individual consumers, and collects and cross-analyzes the propensity of potential customers to investigate the effect of advertising and the motivation of customers.

100-400‧‧‧Steps

Claims (10)

A method for monitoring a person flow monitoring includes the following steps: an advertisement recording step of recording an image on an advertisement billboard by using a camera, and presenting the image by the advertisement billboard; the face recognition step automatically identifying whether the image has at least one face And locking the face and calculating a data of the face, the data including the number of the face, the time of the face viewing, and the size of the face; the flow statistics step, storing and counting the data; a mask feedback step of displaying at least one design pattern on the face of the designated person, and integrating the design pattern with the data to interact with the face, and the mask changing step determines the face and the advertisement by using the size of the face The distance of the kanban, and the data determines whether the design pattern is displayed on the face of the face corresponding to the advertisement kanban. The method for monitoring the advertisement of the flow of claim 1 further includes: a mask changing step of using the plurality of design patterns to match the face data in turn or simultaneously displaying on the face of the person. The method for monitoring the advertisement of the flow of claim 2, further comprising: a report output step, using the mask feedback step and the mask change step operation statistics, and outputting a one-person flow monitoring report. The method for monitoring the advertisement of the flow of the item of claim 1, further comprising: a report output step, using the face data to calculate statistics, and outputting a one-person flow monitoring report. The method for monitoring the advertisement of the flow of the item of claim 1, wherein the face recognition step subtracts the image from the subtraction operation and takes out a human figure, and then converts the image by RGB, and calculates the edge for capturing, and then extracts the face through image compression. Find the face through ratio calculation or genetic algorithm. A person flow monitoring advertisement system includes: an advertisement billboard; a camera disposed on the advertisement billboard, wherein the camera captures an image; a display is disposed on the advertisement billboard, and the image is presented by the display; An identification module is connected to the camera, and the human flow identification module calculates and automatically recognizes at least one face of the image, and the human flow identification module can lock the face; a person flow statistics module is connected to the flow recognition module And the person flow statistics module calculates a data according to the face, the data includes the number of the face, the time of the face viewing, and the size of the face; a database connecting the flow statistics module, The database stores at least one design pattern, and the database stores the data provided by the human flow statistics module; and a facial mask module that connects the camera, the display, the human flow identification module, the human flow statistics module, and In the database, the facial mask module extracts the design pattern from the database according to the face and the data, and the facial mask module makes the design pattern follow Displaying the face of the person in the image, and the facial mask module determines the distance between the face and the advertising board by using the size of the face, and determines whether the design pattern is displayed on the advertising board by using the data. The position of the face corresponding to the middle. The person flow monitoring advertisement system of claim 6, wherein the database stores a plurality of design patterns, and the facial mask module sequentially follows the data on the face of the person displayed in the image according to the data. The person flow monitoring advertisement system of claim 6, wherein the database stores a plurality of design patterns, and the facial mask module simultaneously follows the plurality of design patterns on the face of the person displayed in the image according to the data. The person flow monitoring advertisement system of claim 6, wherein the person flow statistics module calculates the number of faces of the advertisement viewing board, the time of the face viewing, and the size of the face, and stores the same according to the preset classification item. Within the database. The person flow monitoring advertisement system of claim 6, wherein the database is additionally synchronized with a cloud server for cloud synchronization.