A risk identification method and system for self-service cashier
By integrating components such as regional monitoring terminals, smart shopping carts and quantity detection devices in the self-service cashier system, we analyze customers' shopping behavior in real time, identify and deal with potential risks, and solve the lack of systematicity and integrity of self-service cashier anti-theft loss solution in the existing technology. , The problem of poor risk warning capabilities has achieved more efficient risk warning and handling.
Patent Information
- Application Number
- CN202411285064.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-13
- Publication Date
- 2024-12-10
- Estimated Expiration
- 2044-09-13
AI Technical Summary
The existing self-service cashier anti-theft loss plan lacks systematicity and integrity, has poor risk warning capabilities, and cannot effectively identify and deal with missed goods and theft.
A risk identification system for self-service cashiers is designed, including servers, regional monitoring terminals, smart shopping carts, quantity detection devices and self-service cashiers equipment, through infrared cameras, conventional cameras, RFID card readers, pressure sensors, identity recognition cameras and millimeters Components such as wave radar monitor and analyze customers' shopping behavior in real time, judge and identify potential risks of theft, scandal and emotional abnormalities, and process them through light warnings and prompt information.
Through systematic risk identification and judgment, the risk warning capabilities of self-service cashier equipment are significantly improved, the risk identification process is simplified, and the prevention and handling efficiency of commodity missed scans and thefts are improved.
Smart Images
Figure CN118823929B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computer technology, and in particular to a risk identification method and system for self-service cashiers. Background Art
[0002] In recent years, as the pace of retail digitalization has accelerated, supermarkets, convenience stores, and discount store chains in my country have gradually adopted self-service checkout systems to replace manual checkouts. Self-service checkouts are more convenient and faster, which not only improves customers' shopping experience, but also saves labor costs for companies. However, the losses caused by missed scanning and theft of goods have increased significantly.
[0003] Currently, the problem of theft is usually solved based on the loss prevention function of self-service checkout equipment. However, this method only focuses on specific links, obtains limited information, lacks systematicity and integrity, and has poor risk warning capabilities. Summary of the invention
[0004] In order to solve the technical problems that the self-service cashier anti-theft and loss solutions in the prior art lack systematicity and integrity and have poor risk early warning capabilities, the risk identification system for self-service cashier proposed in the present invention includes a server, an area monitoring terminal, an intelligent shopping cart, a quantity detection device and a self-service cashier device, wherein the server is respectively connected to the area monitoring terminal, the intelligent shopping cart, the quantity detection device and the self-service cashier device for communication;
[0005] The regional monitoring terminal includes an infrared camera, a conventional camera and a third processor, and the regional monitoring terminal is arranged in a key area;
[0006] The quantity detection device is arranged at the entrance of the self-service checkout area, and is used to judge the quantity of goods purchased by customers and their specific commodity information. The quantity detection device includes a sampling camera and multiple RFID card readers. Passive RFID tags are arranged on the goods in the supermarket, and the corresponding commodity information is stored inside the passive RFID tags.
[0007] The self-service cashier device includes a second processor, an identity recognition camera, an infrared camera, a millimeter wave radar and a light warning device.
[0008] Preferably, the smart shopping cart is provided with a pressure sensor, a camera and a first processor, the pressure sensor is arranged at a central position of the bottom of the smart shopping cart, and the first processor and the camera are integrated together and arranged at a middle position of the handle of the smart shopping cart.
[0009] The risk identification method for self-service cashier proposed in the present invention is based on the risk identification system for self-service cashier, and the risk identification method specifically includes the following steps:
[0010] S1. The regional monitoring terminal monitors the shopping process of customers, identifies customers with theft risks as target customers, and transmits the target customers' characteristic data and theft intention information to the server;
[0011] S2. Before the customer enters the self-service checkout area, the customer is prompted to stay at the quantity detection device for a set time. The server regards the customer who has the risk of hiding goods as the target customer and stores the characteristic data of the target customer and the information of the hidden goods;
[0012] S3. The customer scans the product at the self-service cashier device. When the customer chooses to settle the bill, the self-service cashier device sends the customer's feature data, the customer's facial temperature data, the customer's heart rate, and the product information scanned by the customer to the server. The server regards customers with a risk of missed scanning or abnormal emotions as target customers, and stores the target customers' feature data and missed scanning information or abnormal emotions information;
[0013] S4. The server integrates information based on the customer's characteristic data and performs risk management based on the integrated information.
[0014] Preferably, in S1, the body temperature of customers in the monitoring area is judged according to the image of the infrared camera, and customers with body temperature greater than the temperature judgment threshold are taken as tentative target customers, the concentration coefficient of the tentative target customers is calculated, and the tentative target customers with concentration coefficient lower than the coefficient threshold are taken as final target customers, the facial features of the target customers are collected, and the facial features are privacy-enhanced to obtain feature data, and the feature data and theft intention information are transmitted to the server, and the concentration coefficient is the ratio of the number of image frames in which the tentative target customer's line of sight is focused on the product to the total number of image frames.
[0015] Preferably, in S2, when a customer stops at the quantity detection device, the sampling camera captures the customer's current status image and facial features, the passive RFID tag set on the product approaches the RFID reader, the antenna of the passive RFID tag converts the received electromagnetic wave energy into electrical energy, activates the chip in the RFID tag, and sends the product information in the chip to the RFID reader, the number of products purchased by the customer can be determined based on the number of received product information, the facial features are privacy enhanced to obtain feature data, and the feature data, the customer's current status image, the number of products purchased by the customer, and the specific product information are sent to the server.
[0016] Preferably, in S2, the server receives characteristic data from the quantity detection device, the customer's current status image, the quantity of goods purchased by the customer, and specific product information, and determines whether the customer places the goods in the smart shopping cart based on the customer's current status image. If the customer places the goods in the smart shopping cart, the server obtains a first quantity value of the goods through the smart shopping cart. If the customer does not place the goods in the smart shopping cart, the server determines the first quantity value of the goods by analyzing the current status image, uses the quantity of goods purchased by the customer from the quantity detection device as the second quantity value of the goods, and determines whether the customer hides the goods based on the first quantity value and the second quantity value.
[0017] Preferably, in S3, the self-service checkout device performs privacy enhancement processing on the facial features of the customer captured by the identity recognition camera to obtain feature data, and when the customer chooses to settle the bill, the feature data, the customer's facial temperature data, the customer's heart rate and the product information scanned by the customer are transmitted to the server.
[0018] Preferably, in S3, the process of judging the risk of missed scanning is to obtain the corresponding product information uploaded by the quantity detection device from the server according to the customer's characteristic data, and compare the product information with the product information scanned by the customer. If the two are consistent, it is determined that no missed scanning has occurred; if the two are inconsistent, it is determined that a missed scanning has occurred; the process of judging the risk of abnormal emotions is to judge that the customer's emotional state is normal if the customer's facial temperature data and the customer's heart rate are both within the normal range; otherwise, the customer's emotional state is determined to be abnormal.
[0019] Preferably, in S4, the server integrates the risk information corresponding to the customer's characteristic data. If the customer is at risk of theft or concealment of goods, the server controls the light warning device of the self-service checkout equipment to sound an alarm, instructing the staff to go to a specific self-service checkout equipment for processing. If the customer is at risk of missing a scan, the server controls the self-service checkout equipment to display a prompt message to guide the customer to re-scan. If the customer is at risk of abnormal emotions, the server controls the self-service checkout equipment to display a comforting statement to calm the customer's emotions.
[0020] Preferably, the privacy enhancement processing of facial features is to convert the grayscale value of each pixel in the facial feature image into a binary representation, and form the binary representation of the grayscale value of each pixel into a row of transition data in order from left to right and from top to bottom, and scramble the transition data according to the set encryption rules to obtain feature data.
[0021] Compared with the prior art, the present invention has the following beneficial effects:
[0022] Risk identification and judgment are carried out throughout the entire shopping process of customers. Through systematic risk identification, the risk identification process of self-service checkout equipment is simplified, and the overall risk warning capability is improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 It is a flow chart of the risk identification method of the present invention. DETAILED DESCRIPTION
[0024] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0025] The risk identification system for self-service cashier provided by the present invention comprises a server, a regional monitoring terminal, a smart shopping cart, a quantity detection device and a self-service cashier device. The server is respectively connected to the regional monitoring terminal, the smart shopping cart, the quantity detection device and the self-service cashier device for communication.
[0026] The regional monitoring terminal includes an infrared camera, a conventional camera and a third processor, and the third processor is connected to the infrared camera and the conventional camera respectively. The working process of the third processor is to judge the body temperature of customers in the monitoring area according to the image of the infrared camera, take customers whose body temperature is greater than the temperature judgment threshold as temporary target customers, calculate the concentration coefficient of the temporary target customers, take the temporary target customers whose concentration coefficient is lower than the coefficient threshold as the final target customers, collect the facial features of the target customers, obtain feature data after privacy enhancement processing of the facial features, and transmit the feature data and theft intention information to the server. The concentration coefficient is the ratio of the number of image frames of the temporary target customer's line of sight to the total number of image frames. Normal customers will pay more attention to the goods themselves during shopping. Customers with the intention of stealing will look around and observe the position of the camera during shopping, and pay less attention to the goods themselves. Therefore, the concentration coefficient can identify whether the customer has the intention of stealing. The regional monitoring terminal is set in the key area, which is the commodity area where theft is prone to occur according to historical data statistics and the commodity area where the average price of the goods is greater than the price threshold.
[0027] The smart shopping cart is provided with a pressure sensor, a camera and a first processor. The pressure sensor is provided at the central position of the bottom of the smart shopping cart. The first processor and the camera are integrated together and provided at the middle position of the handle of the smart shopping cart. During the shopping process of the customer, if the data of the pressure sensor changes, the camera is started to capture the image inside the smart shopping cart, and the first processor determines the real-time quantity of goods inside the smart shopping cart through image recognition. After the shopping is completed, the real-time quantity of goods is transmitted to the server. The specific process of determining the quantity of goods through image recognition is that the first processor determines the change trend of the quantity of goods according to the data of the pressure sensor. If the pressure value becomes smaller, it means that the quantity of goods has decreased. If the pressure value becomes larger, it means that the quantity of goods has increased. The current image is compared with the previous image to determine the difference area, the quantity change value is determined according to the difference area, and the real-time quantity of goods is adjusted according to the change trend.
[0028] The quantity detection device is set at the entrance of the self-service checkout area to determine the number of goods purchased by customers and their specific product information. Passive RFID tags are set on the goods in the supermarket, and the corresponding product information is stored inside the passive RFID tags. The quantity detection device includes a sampling camera and multiple RFID readers. When a customer stops at the quantity detection device, the sampling camera takes a picture of the customer's current state image and the customer's facial features. The passive RFID tag set on the product approaches the RFID reader. The antenna of the passive RFID tag converts the received electromagnetic wave energy into electrical energy, activates the chip in the RFID tag, and sends the product information in the chip to the RFID reader. The number of goods purchased by the customer can be determined based on the number of received product information. The facial features are processed for privacy enhancement to obtain feature data. The feature data, the customer's current state image, the number of goods purchased by the customer, and the specific product information are sent to the server.
[0029] The self-service cashier equipment is evenly distributed in the self-service cashier area. The self-service cashier equipment includes a second processor, an identity recognition camera, an infrared camera, a millimeter-wave radar, and a light warning device. The second processor is arranged inside the self-service cashier equipment. The identity recognition camera is arranged at the top of the self-service cashier equipment, which can clearly capture the facial features of the customer. The infrared camera and the millimeter-wave radar are arranged in the middle of the self-service cashier equipment, facing the customer. The infrared camera is used to detect the temperature data of the customer's face, and the millimeter-wave radar is used to detect the customer's heart rate. The self-service cashier equipment obtains feature data after privacy enhancement processing of facial features. When the customer chooses to settle, the feature data, the temperature data of the customer's face, the customer's heart rate, and the product information scanned by the customer are transmitted to the server.
[0030] The privacy enhancement processing of facial features is to convert the grayscale value of each pixel in the facial feature image into a binary representation, and form a row of transition data from the binary representation of the grayscale value of each pixel in the order from left to right and from top to bottom. The transition data is scrambled according to the set encryption rules to obtain the feature data.
[0031] like Figure 1 As shown, based on the above risk identification system, the risk identification method for self-service checkout specifically includes the following steps:
[0032] S1. The regional monitoring terminal monitors the shopping process of customers, identifies customers with theft risks as target customers, and transmits the target customers' feature data and theft intention information to the server.
[0033] S2. Before the customer enters the self-service checkout area, the customer is prompted to stay at the quantity detection device for a set time. The server takes the customer who has the risk of hiding goods as the target customer, and stores the characteristic data and hidden goods information of the target customer. Specifically, the server receives the characteristic data from the quantity detection device, the current state image of the customer, the number of goods purchased by the customer, and the specific product information, and determines whether the customer places the goods in the smart shopping cart according to the current state image of the customer. If the customer places the goods in the smart shopping cart, the server obtains the first quantity value of the goods through the smart shopping cart. If the customer does not place the goods in the smart shopping cart, the server determines the first quantity value of the goods by analyzing the current state image, and takes the number of goods purchased by the customer from the quantity detection device as the second quantity value of the goods, and determines whether the customer hides the goods according to the first quantity value and the second quantity value. The specific process of judging whether the customer hides the goods according to the first quantity value and the second quantity value is that if the difference between the second quantity value and the first quantity value is greater than the error threshold, the customer has the risk of hiding the goods, otherwise, the customer does not have the risk of hiding the goods.
[0034] S3. The customer scans the product at the self-service checkout device. When the customer chooses to settle the bill, the self-service checkout device sends the customer's characteristic data, the customer's facial temperature data, the customer's heart rate, and the product information scanned by the customer to the server. The server regards customers with the risk of missed scanning or emotional abnormality as target customers, and stores the target customers' characteristic data and missed scanning information or emotional abnormality information. The process of judging the risk of missed scanning is to obtain the corresponding product information uploaded by the quantity detection device from the server according to the customer's characteristic data, and compare the product information with the product information scanned by the customer. If the two are consistent, it is determined that there is no missed scanning. If the two are inconsistent, it is determined that there is a missed scanning. The process of judging the risk of emotional abnormality is that if the customer's facial temperature data and the customer's heart rate are both within the normal range, the customer's emotional state is determined to be normal, otherwise, the customer's emotional state is determined to be abnormal.
[0035] S4. The server integrates information based on the customer's characteristic data and performs risk processing based on the integrated information. Specifically, the server integrates the risk information corresponding to the customer's characteristic data. If the customer is at risk of theft or concealment of goods, the server controls the light warning device of the self-service cash register to sound an alarm and instruct the staff to go to a specific self-service cash register for processing. If the customer is at risk of missing a scan, the server controls the self-service cash register to display a prompt message to guide the customer to re-scan. If the customer is at risk of abnormal emotions, the server controls the self-service cash register to display a comforting statement to calm the customer's emotions.
[0036] The above disclosure is only the preferred embodiment of the present invention, and certainly cannot be used to limit the scope of the present invention. It should be pointed out that for those skilled in the art, any equivalent changes made to the scheme of the present invention without departing from the design structure and principle of the present invention are considered to be within the protection scope of the present invention.
Claims
1. A risk identification method for self-service cashier, based on a risk identification system for self-service cashier, the risk identification system comprising a server, an area monitoring terminal, an intelligent shopping cart, a quantity detection device and a self-service cashier device, the server being respectively connected to the area monitoring terminal, the intelligent shopping cart, the quantity detection device and the self-service cashier device for communication; The regional monitoring terminal includes an infrared camera, a conventional camera and a third processor, and the regional monitoring terminal is arranged in a key area; The quantity detection device is arranged at the entrance of the self-service checkout area, and is used to judge the quantity of goods purchased by customers and their specific commodity information. The quantity detection device includes a sampling camera and multiple RFID card readers. Passive RFID tags are arranged on the goods in the supermarket, and the corresponding commodity information is stored inside the passive RFID tags. The self-service cashier device includes a second processor, an identity recognition camera, an infrared camera, a millimeter wave radar and a light warning device; Features ,The risk identification method specifically includes the following steps: S1. The regional monitoring terminal monitors the shopping process of customers, identifies customers with theft risks as target customers, and transmits the target customers' characteristic data and theft intention information to the server; S2. Before the customer enters the self-service checkout area, the customer is prompted to stay at the quantity detection device for a set time. The server regards the customer who has the risk of hiding goods as the target customer and stores the characteristic data of the target customer and the information of the hidden goods; S3. The customer scans the product at the self-service cashier device. When the customer chooses to settle the bill, the self-service cashier device sends the customer's feature data, the customer's facial temperature data, the customer's heart rate, and the product information scanned by the customer to the server. The server regards customers with a risk of missed scanning or abnormal emotions as target customers, and stores the target customers' feature data and missed scanning information or abnormal emotions information; S4. The server integrates information based on the customer's characteristic data and performs risk processing based on the integrated information; In S1, the body temperature of customers in the monitoring area is judged according to the image of the infrared camera, customers whose body temperature is greater than the temperature judgment threshold are taken as tentative target customers, the concentration coefficient of the tentative target customers is calculated, and the tentative target customers whose concentration coefficient is lower than the coefficient threshold are taken as final target customers, facial features of the target customers are collected, and feature data is obtained after the facial features are subjected to privacy enhancement processing, and the feature data and theft intention information are transmitted to the server, and the concentration coefficient is the ratio of the number of image frames in which the tentative target customer's line of sight focuses on the product to the total number of image frames; In S2, when a customer stops at the quantity detection device, the sampling camera captures the customer's current state image and the customer's facial features, the passive RFID tag set on the product approaches the RFID card reader, the antenna of the passive RFID tag converts the received electromagnetic wave energy into electrical energy, activates the chip in the RFID tag, and sends the product information in the chip to the RFID card reader. The number of products purchased by the customer can be determined based on the number of received product information, the facial features are privacy enhanced to obtain feature data, the feature data, the customer's current state image, the number of products purchased by the customer, and the specific product information are sent to the server, and the server receives the information from the digital camera. The server uses characteristic data of the quantity detection device, the customer's current state image, the quantity of goods purchased by the customer, and specific product information to determine whether the customer places the goods in the smart shopping cart according to the customer's current state image. If the customer places the goods in the smart shopping cart, the server obtains a first quantity value of the goods through the smart shopping cart. If the customer does not place the goods in the smart shopping cart, the server determines the first quantity value of the goods by analyzing the current state image, and uses the quantity of goods purchased by the customer from the quantity detection device as the second quantity value of the goods. If the difference between the second quantity value and the first quantity value is greater than the error threshold, the customer is at risk of hiding the goods. Otherwise, the customer is not at risk of hiding the goods.
2. The risk identification method according to claim 1, characterized in that: In S3, the self-service checkout device performs privacy enhancement processing on the facial features of the customer captured by the identity recognition camera to obtain feature data. When the customer chooses to settle the bill, the feature data, the customer's facial temperature data, the customer's heart rate, and the product information scanned by the customer are transmitted to the server.
3. The risk identification method according to claim 2, characterized in that: In S3, the process of judging the risk of missed scanning is to obtain the corresponding product information uploaded by the quantity detection device from the server according to the customer's characteristic data, and compare the product information with the product information scanned by the customer. If the two are consistent, it is determined that no missed scanning has occurred; if the two are inconsistent, it is determined that a missed scanning has occurred; the process of judging the risk of abnormal emotions is to judge that the customer's emotional state is normal if the customer's facial temperature data and the customer's heart rate are both within the normal range, otherwise, the customer's emotional state is determined to be abnormal.
4. The risk identification method according to claim 1, characterized in that: In S4, the server integrates the risk information corresponding to the customer's characteristic data. If the customer is at risk of theft or concealment of goods, the server controls the light warning device of the self-service checkout equipment to sound an alarm, instructing the staff to go to a specific self-service checkout equipment for processing. If the customer is at risk of missing a scan, the server controls the self-service checkout equipment to display a prompt message to guide the customer to re-scan. If the customer is at risk of abnormal emotions, the server controls the self-service checkout equipment to display a comforting statement to calm the customer's emotions.
5. The risk identification method according to any one of claims 1-2, characterized in that: The privacy enhancement processing of facial features is to convert the grayscale value of each pixel in the facial feature image into a binary representation, and form a row of transition data from the binary representation of the grayscale value of each pixel in the order from left to right and from top to bottom. The transition data is scrambled according to the set encryption rules to obtain the feature data.