Intelligent monitoring system and method based on 5G Internet of Things

Abstract

The invention discloses an intelligent monitoring system and method based on the 5G Internet of Things, and relates to the technical field of intelligent monitoring, and the system comprises a data collection module, a data processing platform, a cloud server, an alarm module, a monitoring center and a mobile APP, and is used for solving the defects in the aspects of customer behavior prediction and supermarket safety in the prior art. A cloud platform management tool of the cloud server is used for monitoring the system state and the application program performance in real time, quickly responding to problems and regularly executing data cleaning and backup operations, so that the storage space is prevented from being excessively occupied; according to the method, time series, circulating network model deep learning and Internet of Things technologies are adopted, real-time monitoring and alarm mechanisms are combined, customer action tracks and customer behaviors are predicted and analyzed, appropriate supermarket layout and marketing strategies are formulated, meanwhile, abnormal behaviors can be rapidly and accurately recognized and processed, and the service life of the supermarket is prolonged. Therefore, the safety and management efficiency of supermarkets and supermarkets are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to an intelligent monitoring system and method based on 5G Internet of Things. Background Art

[0002] With the continuous popularization and development of modern supermarkets, the internal security management of supermarkets has also received increasing attention. Traditional monitoring systems can only provide monitoring images. Although they can provide certain security guarantees, they have certain deficiencies in abnormal behavior identification, real-time monitoring and early warning. When abnormal behavior occurs and data is called, it is impossible to quickly locate the abnormal event screen due to the redundant data, so it cannot meet the actual needs of supermarket security management.

[0003] On the other hand, traditional video surveillance lacks intelligent video analysis technology and intelligent data mining methods, and cannot predict customer behavior or sales behavior, and cannot meet the needs of business operation management. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent monitoring system and method based on 5G Internet of Things to address the deficiencies in the background technology.

[0005] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: an intelligent monitoring system based on 5G Internet of Things, characterized in that it includes a data acquisition module, a data processing platform, a cloud server, an alarm module, a monitoring center and a mobile APP application; The data collection module consists of several data collection devices including cameras, Wi-Fi access points and Wi-Fi positioning servers, which are installed in various areas of the supermarket to obtain video images, customer trajectories and supermarket environment data in the supermarket; The data processing platform is used to perform algorithmic processing and analysis on customer trajectory data and supermarket environment data, respectively predict customer behavior and issue alarm signals for abnormal situations; Cloud servers are used to store, transmit and manage data; The alarm module is used to receive the alarm signal sent by the data processing platform and send out alarm information; The monitoring center is used to centrally manage and monitor all data acquisition equipment; The mobile APP application is used to provide mobile applications for supermarket managers and employees to view internal video images and alarm information of supermarkets in real time, and remotely control and manage data acquisition equipment.

[0006] In a preferred embodiment, the customer's location is located through a Wi-Fi access point to form a customer trajectory route: and the customer's purchasing behavior is predicted through the customer trajectory route: Wi-Fi access points are installed in supermarkets. Each access point broadcasts a unique identifier (SSID). When customers enter a supermarket, their mobile phones automatically search for available Wi-Fi access points and scan their signal strength and SSID information. Collect the location and signal strength of each Wi-Fi access point in the supermarket; The positioning server calculates the distance between the customer and the Wi-Fi access point based on the signal strength value transmitted by the customer and the access point's own signal strength value, thereby determining the customer's current location; as the customer moves, the signal transmitted by the customer to the Wi-Fi access point becomes weaker and weaker, which means that the customer is getting farther and farther away from the access point.

[0007] Collect supermarket maps and mark the locations where customers move on the supermarket maps to obtain customer trajectory routes.

[0008] In a preferred embodiment, the data processing platform processes the customer trajectory route to predict the customer behavior in the following process: Obtain customer historical action trajectory data in advance; Establish a two-dimensional coordinate system of customer location and time to obtain the location coordinates at each time point; take the location coordinates at each time point as a feature, and then arrange them in chronological order to form a location time series; The location time series is used as the first input and input into the learning model RNN recurrent neural network model to obtain a trained purchase behavior prediction model; The learning model RNN recurrent neural network model consists of memory units and neural networks; The memory unit is used to store the calculation result X of the first input t ; The neural network is used to calculate the result X t As the next input, calculate and get the output value X t+1， and is stored in the memory unit; Again, the output value X t+1 Input to the neural network and get the output value X t+2 , X t+2 Input it into the neural network again, thus forming a cyclic calculation in the RNN recurrent neural network model. When the last time series is input, the trained customer behavior prediction model is obtained. Input the location time series of the customer's action trajectory to be detected into the customer behavior prediction model to obtain the customer behavior prediction result; By converting the customer's trajectory into a time series and then training it with the RNN deep learning network model, we can make predictions on supermarket marketing data and the following indicators: Sales: By analyzing the time customers spend in different areas and their purchasing behavior, sales trends can be predicted.

[0009] Traffic flow: By analyzing the number of customers entering and leaving the supermarket and their activity tracks within the supermarket, changes in traffic flow can be predicted.

[0010] Hot Items: Analyze the sales trends and correlations of hot items based on the customer’s location time series, and predict future hot items in order to develop promotion plans and increase sales revenue.

[0011] Inventory requirements: By analyzing the relationship between sales and inventory levels, future inventory requirements can be predicted.

[0012] Promotion effectiveness: By analyzing customer behavior and purchasing patterns during a specific campaign, you can evaluate the effectiveness of your promotion and make adjustments accordingly.

[0013] Customer satisfaction: Analyze customers’ shopping experience and satisfaction based on customer location time series and individual behavior data, and predict future customer satisfaction in order to improve service quality and increase customer satisfaction.

[0014] In a preferred embodiment, the process of collecting supermarket environment data is as follows: Supermarket environment data includes: supermarket traffic, shelf merchandise status, and cashier status; Supermarket traffic: collect Wi-Fi access point signal connection times; Shelf product status: the number of shelf products and the number of reductions; Cashier status: collect the number of people queuing at the cashier; The data processing platform processes supermarket environment data as follows: Count the number of Wi-Fi access point signal connections, calculate the human traffic, and set a threshold for the human traffic in supermarkets. If the human traffic is greater than or equal to the threshold, it means that the human traffic exceeds the capacity of the supermarket. At this time, the alarm system is triggered and unblocking management must be taken; The goods on the shelf have an initial quantity value before opening. The number of goods reduced is obtained through monitoring, and the reduction speed of goods is calculated in combination with time. A threshold is set for the reduction speed. When the reduction speed is greater than the threshold, the alarm system is triggered to check whether there is any abnormal situation and determine whether it is a hot sale or theft. Set a threshold for the number of people queuing at the cashier counter. When the number of people queuing is greater than the threshold, an alarm is triggered, indicating that the number of people queuing for checkout exceeds the normal level. It is necessary to adopt unblocking management or temporarily add cashier counters. In a preferred embodiment, the alarm module issues an alarm when the following anomalies occur: The flow of people in the supermarket exceeds the preset threshold or increases suddenly; Customers spend too much time in a particular area; The number of customers waiting in line at the checkout counter exceeded normal levels; The merchandise on a certain shelf is decreasing at an unusually fast rate, which may involve theft; Abnormal incidents such as malicious destruction, fighting, etc. occurred in the supermarket; These situations are all abnormal phenomena and will trigger the alarm system and generate an alarm.

[0015] In a preferred embodiment, the cloud server stores, transmits and manages the data collected by intelligent monitoring as follows: Use cloud storage services to save the collected data on the cloud; The HTTPS protocol is used to encrypt and transmit the compressed data. At the same time, the message queue is used to realize asynchronous data transmission, reducing the direct interaction between the client and the server. When this data is called, the data is decoded to restore the original data information. Use the cloud platform management tools that come with the cloud server to monitor system status and application performance in real time, respond to problems quickly, and perform data cleanup and backup operations regularly to avoid excessive storage space occupation.

[0016] In a preferred embodiment, the monitoring center centrally manages and monitors all monitoring equipment and data, provides real-time monitoring, early warning and analysis services, and quickly identifies and resolves security issues. At the same time, the monitoring center maintains and upgrades the monitoring system to ensure that the system is always in a reliable working state.

[0017] In a preferred embodiment, the intelligent monitoring system communicates with the mobile APP application by using an application program interface. The mobile application sends a request to the monitoring system and obtains a response. The mobile APP application provides the following functions: Real-time monitoring: allows users to watch live video streams and other sensor data updated in real time; Alarm notification: When the monitoring system detects abnormal activities or events, it sends alarm notifications to users via push notifications or SMS; Remote control: Allows users to remotely start / shut down devices and monitor systems through the APP as needed; Data query: allows users to view historical monitoring data and event records; Setting adjustment: allows users to set up equipment and adjust monitoring parameters; Position tracking: provides positioning and trajectory tracking functions; User management: Administrator users manage users and permissions.

[0018] In a preferred embodiment, a smart monitoring method based on 5G Internet of Things includes the following steps: Step 1: Obtain various data of monitoring equipment through the data acquisition module; Step 2: Analyze and process data through the data processing platform; Step 3: Compress the collected data through the cloud server and upload it to the cloud for sharing and remote control; Step 4: Set threshold rules. When the collected data exceeds the threshold range, the alarm system is triggered to issue warnings and alarms, and relevant personnel are notified to handle the situation. Step 5: Operation personnel monitor and manage the system in real time through the monitoring center; Step 6: Users can check the system status and perform operations at any time through the mobile app.

[0019] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. The present invention calculates the customer's movement trajectory and residence time through intelligent monitoring and data analysis technology, and inputs it into the neural network model through location time series, so as to predict the sales volume, traffic volume, popular commodities, inventory demand, promotion effect and customer satisfaction of supermarkets, and provide a scientific basis for managers to adjust the layout and marketing strategy of supermarkets in time, thereby improving the customer shopping experience and conversion rate (that is, the proportion of all consumers who visit the store who can eventually be converted into successful customers); on the other hand, by monitoring the supermarket environment, it can automatically detect abnormal behaviors, such as theft, fighting, etc., and effectively prevent and reduce the occurrence of safety accidents.

[0020] 2. The present invention can better manage equipment and assets and save labor costs through technical means such as the Internet of Things. In addition, the intelligent monitoring system can collect a large amount of data, and through data analysis and mining, provide more accurate decision support for large supermarkets, thereby realizing intelligent management. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0022] Figure 1 is a system block diagram of the present invention; Figure 2 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0024] Example 1, please refer to Figure 1 As shown, the intelligent monitoring system based on 5G Internet of Things described in this embodiment is characterized by including a data acquisition module, a data processing platform, a cloud server, an alarm module, a monitoring center and a mobile APP application; The data collection module consists of several data collection devices including cameras, Wi-Fi access points and Wi-Fi positioning servers, which are installed in various areas of the supermarket to obtain video images, customer trajectories and supermarket environment data in the supermarket; The data processing platform is used to perform algorithmic processing and analysis on customer trajectory data and supermarket environment data, respectively predict customer behavior and issue alarm signals for abnormal situations; Cloud servers are used to store, transmit and manage data; The alarm module is used to receive the alarm signal sent by the data processing platform and send out alarm information; The monitoring center is used to centrally manage and monitor all data acquisition equipment; The mobile APP application is used to provide mobile applications for supermarket managers and employees to view internal video images and alarm information of supermarkets in real time, and remotely control and manage data acquisition equipment.

[0025] In supermarkets, customers are located through Wi-Fi access points to form customer trajectory routes. The steps are as follows: Wi-Fi access points are installed in supermarkets. Each access point broadcasts a unique identifier (SSID). When customers enter a supermarket, their mobile phones automatically search for available Wi-Fi access points and scan their signal strength and SSID information. Collect the location and signal strength value PT of each Wi-Fi access point in the supermarket; The positioning server calculates the distance L between the customer and the Wi-Fi access point based on the signal strength value PR sent by the customer and the signal strength value of the access point itself: L=PT / PR*n Where n is the propagation factor, the size of which depends on the external environment in which the wireless signal propagates; As customers move, the signal strength value PR transmitted by their mobile phones to the Wi-Fi access point also changes. When customers are farther and farther away from the Wi-Fi access point, the signal transmitted to the Wi-Fi access point will become smaller and smaller, and obviously the L value will become larger and larger. These L values ​​also represent the customer's mobile location; Collect supermarket maps and mark the customer's movement locations on the map to obtain the customer's trajectory route.

[0026] The data processing platform processes the customer trajectory route and predicts the customer behavior in the following process: Obtain customer historical action trajectory data in advance; Establish a two-dimensional coordinate system of customer location and time to obtain the location coordinates at each time point; take the location coordinates at each time point as a feature, and then arrange them in chronological order to form a location time series; The location time series is used as the first input and input into the learning model RNN recurrent neural network model to obtain a trained purchase behavior prediction model; The learning model RNN recurrent neural network model consists of memory units and neural networks; The memory unit is used to store the calculation result X of the first input t ; The neural network is used to calculate the result X t As the next input, calculate and get the output value X t+1， and is stored in the memory unit; Again, the output value X t+1 Input to the neural network and get the output value X t+2 , X t+2 Input it into the neural network again, so that a cyclic calculation is formed in the RNN recurrent neural network model. When the last time series X is input t+n When the trained purchase behavior prediction model is obtained; Input the location time series of the customer's action trajectory to be detected into the purchase behavior prediction model to obtain the customer behavior prediction result; By converting the customer's trajectory into a time series and then training it with the RNN deep learning network model, we can make predictions on supermarket marketing data and the following indicators: Sales: By analyzing the time customers spend in different areas and their purchasing behavior, sales trends can be predicted.

[0027] Traffic flow: By analyzing the number of customers entering and leaving the supermarket and their activity tracks within the supermarket, changes in traffic flow can be predicted.

[0028] Hot Items: Analyze the sales trends and correlations of hot items based on the customer’s location time series, and predict future hot items in order to develop promotion plans and increase sales revenue.

[0029] Inventory requirements: By analyzing the relationship between sales and inventory levels, future inventory requirements can be predicted.

[0030] Promotion effectiveness: By analyzing customer behavior and purchasing patterns during a specific campaign, you can evaluate the effectiveness of your promotion and make adjustments accordingly.

[0031] Customer satisfaction: Analyze customers’ shopping experience and satisfaction based on customer location time series and individual behavior data, and predict future customer satisfaction in order to improve service quality and increase customer satisfaction.

[0032] For example, to predict future popular products, the specific steps are as follows: 1. Divide the time series data into sequences of equal length; 2. Combine customer purchase behavior, product attributes, and whether the product is on promotion to extract features and build an RNN recurrent network model. Here, the LSTM model training data set under the RNN category is used; 3. Use historical customer trajectory data to train the model, input the time series into the model, obtain the output result, keep the output result in a memory unit, and then use the result as the next input, input the neural network, and repeat the cycle. Finally, get the prediction result through fitting. 4. Evaluate and tune the model to improve prediction accuracy and stability while avoiding overfitting and underfitting; 5. Use charts to present the forecast results and explain the reasons and influencing factors behind the forecast results so that decision makers can formulate specific business plans and marketing strategies.

[0033] The process of collecting supermarket environment data: Supermarket environment data includes: supermarket traffic, shelf merchandise status, and cash register status.

[0034] Traffic volume in supermarkets: the number of connections to Wi-Fi signal access points; Shelf product status: the number of shelf products and the number of reductions; Cashier status: number of people queuing at the cashier; The data processing platform processes supermarket environment data as follows: Count the number of connections to Wi-Fi signal access points, calculate the flow of people, set a threshold for the flow of people in supermarkets, and trigger the alarm module if the flow of people is greater than or equal to the threshold; The shelf goods have an initial value before opening. The number of goods reduced is obtained through monitoring, and the reduction rate is calculated. For example, if the initial value of the goods is P1, within time t, the quantity is reduced to P 2， Then the commodity reduction rate is: V = (P1-P2) / t; From the formula, we can see that the shorter the time is and the greater the number of goods reduced, the greater the reduction rate V, that is, the faster the goods are reduced. When the reduction rate is too fast, management personnel need to find out whether it is a normal promotion or theft.

[0035] Set a threshold for the reduction speed. When the reduction speed is greater than the threshold, the alarm module is triggered. Set a threshold for the number of people queuing at the cashier counter. When the number of people queuing exceeds the threshold, an alarm is triggered. It is also possible to obtain the customer's facial expressions and body language, identify the customer's expression through image recognition technology, and combine the body language and the state of the goods to identify whether the goods have been maliciously damaged. At the same time, the customer's expression can be identified through image recognition technology, and combined with the body language to identify whether a fight has occurred.

[0036] The alarm module issues an alarm when the following abnormal conditions occur: The flow of people in the supermarket exceeds the preset threshold; The customer stays in a fixed area for a time greater than the threshold; The number of customers waiting in line at the checkout counter is greater than the threshold; The rate of decrease of goods on a certain shelf is greater than the threshold; Abnormal incidents such as malicious destruction and fighting occurred in the supermarket.

[0037] In addition, for example, if the temperature changes beyond the specified range, such as the temperature of the refrigerator or freezer is too high or too low, or if there is a sudden major equipment failure, such as air conditioning system failure, ventilation system failure, lighting system flickering, etc., these situations will also trigger the alarm system and sound an alarm so that managers or staff can take timely measures to resolve unsafe factors.

[0038] For the video image data collected by the acquisition device, the process of storage, transmission and management using the cloud server is as follows: Use cloud storage services to save the collected data on the cloud; The HTTPS protocol is used to encrypt and transmit the compressed data. At the same time, the message queue is used to realize asynchronous data transmission, reducing the direct interaction between the client and the server. When this data is called, the data is decoded to restore the original data information. Use the cloud platform management tools that come with the cloud server to monitor system status and application performance in real time, respond to problems quickly, and perform data cleanup and backup operations regularly to avoid excessive storage space occupation.

[0039] The monitoring center centrally manages and monitors all monitoring equipment and data, provides real-time monitoring, early warning and analysis services, and quickly identifies and resolves security issues. At the same time, the monitoring center maintains and upgrades the monitoring system to ensure that the system is always in a reliable working state.

[0040] The intelligent monitoring system communicates with the mobile APP application by using the application program interface. The mobile application sends requests to the monitoring system and obtains responses. The mobile APP application provides the following functions: Real-time monitoring: allows users to watch live video streams and other sensor data updated in real time; Alarm notification: When the monitoring system detects abnormal activities or events, it sends alarm notifications to users via push notifications or SMS; Remote control: Allows users to remotely start / shut down devices and monitor systems through the APP as needed; Data query: allows users to view historical monitoring data and event records; Setting adjustment: allows users to set up equipment and adjust monitoring parameters; Position tracking: provides positioning and trajectory tracking functions; User management: Administrator users manage users and permissions.

[0041] In a preferred embodiment, a smart monitoring method based on 5G Internet of Things includes the following steps: Step 1: Obtain various data of monitoring equipment through the data acquisition module; Step 2: Analyze and process data through the data processing platform; Step 3: Compress the collected data through the cloud server and upload it to the cloud for sharing and remote control; Step 4: Set threshold rules. When the collected data exceeds the threshold range, the alarm system is triggered to issue warnings and alarms, and relevant personnel are notified to handle the situation. Step 5: Operation personnel monitor and manage the system in real time through the monitoring center; Step 6: Users can check the system status and perform operations at any time through the mobile app.

[0042] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present 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. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent monitoring system based on 5G Internet of Things, characterized in that: Including data acquisition module, data processing platform, cloud server, alarm module, monitoring center and mobile APP application; The data collection module consists of several data collection devices including cameras, Wi-Fi access points and Wi-Fi positioning servers, which are installed in various areas of the supermarket to obtain video images, customer trajectories and supermarket environment data in the supermarket; The data processing platform is used to perform algorithmic processing and analysis on customer trajectory and supermarket environment data, respectively predict customer behavior and issue alarm signals for abnormal situations; Cloud servers are used to store, transmit and manage video image data; The alarm module is used to receive the alarm signal sent by the data processing platform and send out alarm information; The monitoring center is used to centrally manage and monitor all data collection devices and push alarm information to the mobile APP; The mobile APP application is used to provide mobile applications for supermarket managers and employees to view internal video images and alarm information of supermarkets in real time, and remotely control and manage data acquisition equipment.

2. The intelligent monitoring system based on 5G Internet of Things according to claim 1 is characterized in that: Locate the customer through Wi-Fi access points to form customer tracks: Collect the location and signal strength of each Wi-Fi access point in the supermarket; When a customer enters a supermarket, the mobile phone they carry will automatically search for available Wi-Fi access points. After the Wi-Fi access point receives the signal from the customer's mobile phone, it will transmit the signal to the positioning server. The positioning server calculates the distance between the customer's mobile phone and the Wi-Fi access point based on the signal strength value transmitted by the customer's mobile phone and the signal strength value of the Wi-Fi access point itself, thereby obtaining the customer's current location; Collect supermarket maps, mark the customer's current location on the supermarket map, and obtain the customer's trajectory.

3. The intelligent monitoring system based on 5G Internet of Things according to claim 2 is characterized in that: The data processing platform processes customer trajectories and then predicts customer behavior in the following process: Get the customer's historical customer trajectory; The data processing platform processes the customer trajectory as follows: establish a two-dimensional coordinate system of customer location and time to obtain the location coordinates at each time point; use the location coordinates at each time point as a feature, and then arrange them in chronological order to form a location time series; The process of predicting customer behavior is: The location time series is used as the first input and input into the learning model RNN recurrent neural network model to obtain a trained customer behavior prediction model; The learning model RNN recurrent neural network model consists of memory units and neural networks; The memory unit is used to store the calculation result input for the first time; The neural network is used to use the calculation result as the next input, perform calculation, obtain the output value, and store it in the memory unit; The output value is input into the neural network again to obtain a new output value. In this way, a loop calculation is formed in the RNN recurrent neural network model to obtain a trained customer behavior prediction model; The location time series of the customer trajectory to be detected is input into the customer behavior prediction model to obtain the customer behavior prediction result.

4. The intelligent monitoring system based on 5G Internet of Things according to claim 1 is characterized in that: The collection process of supermarket environment data is as follows: Supermarket environment data includes: supermarket traffic, shelf merchandise status, and cashier status; Supermarket traffic: collect Wi-Fi access point signal connection times; Shelf product status: the number of shelf products and the number of reductions; Checkout counter status: collect the number of people queuing at the checkout counter.

5. The intelligent monitoring system based on 5G Internet of Things according to claim 4 is characterized in that: The data processing platform processes supermarket environment data as follows: Count the number of Wi-Fi access point signal connections and calculate the traffic volume; Get the initial value of shelf goods before opening, obtain the number of goods reduced through monitoring, and calculate the speed of business reduction; When the following abnormal situations occur, an alarm signal is sent to the alarm module: Set a threshold for the flow of people in the supermarket. If the flow of people is greater than or equal to the threshold, an alarm signal will be sent to the alarm module; A threshold is set for the reduction speed of the goods. When the reduction speed is greater than the threshold, an alarm signal is sent to the alarm module; Set a threshold for the number of people queuing at the cashier counter. When the number of people queuing is greater than or equal to the threshold, an alarm signal is sent to the alarm module.

6. The intelligent monitoring system based on 5G Internet of Things according to claim 5 is characterized in that: When the alarm module receives the alarm signal, it sends an alarm message to the monitoring center.

7. The intelligent monitoring system based on 5G Internet of Things according to claim 1 is characterized in that: The cloud server stores, transmits and manages the data collected by intelligent monitoring as follows: Use a cloud server to save the video images collected by the data acquisition device on the cloud; The HTTPS protocol is used to encrypt and transmit the compressed video images; The cloud server includes a cloud platform management tool, which is used to regularly clean up and back up video image data.

8. The intelligent monitoring system based on 5G Internet of Things according to claim 1 is characterized in that: The monitoring center manages and monitors all data acquisition devices and video image data, and pushes received alarm information to the mobile APP.

9. The intelligent monitoring system based on 5G Internet of Things according to claim 8 is characterized in that: The mobile APP application communicates with the intelligent monitoring system by using the application program interface. The mobile APP application has real-time monitoring, alarm notification, remote control, data query and setting adjustment functions: Real-time monitoring: supermarket managers and employees can watch video images in real time; Alarm notification: When the monitoring center issues an alarm, the mobile APP application pushes the alarm notification via SMS on the current interface; Remote control: supermarket managers can remotely start / stop data collection equipment through the APP; Data query: view historical video image data; Settings and adjustments: Supermarket managers set up and adjust parameters of data collection equipment.

10. An intelligent monitoring method based on 5G Internet of Things, used to implement the monitoring system according to any one of claims 1 to 9, characterized in that: The monitoring method includes the following steps: Step 1: Obtain video images, customer trajectories, and supermarket environment data in the supermarket through the data acquisition module; Step 2: Analyze and process data through the data processing platform; Step 3: Compress the collected data through the cloud server and upload it to the cloud for sharing and remote control; Step 4: Set threshold rules. When the collected data exceeds the threshold range, the alarm module is triggered to issue a warning and alarm, and notify relevant personnel to handle the situation. Step 5: Operation personnel monitor and manage the system in real time through the monitoring center; Step 6: Supermarket managers and employees can check the system status and perform operations at any time through the mobile app.

Images