Shopping place management system for unmanned shop operation
Through the unmanned store system, the product shelf life is managed and discount information is provided, and the use of human sensing and deep learning to analyze customer actions, speculate and guide product location, the problem of product shelf life management and customer positioning in unmanned stores is solved, and the product sales rate and shopping experience are improved.
Patent Information
- Application Number
- CN202410802867.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-05-27
- Filing Date
- 2024-06-20
- Publication Date
- 2025-05-20
AI Technical Summary
Products such as fruits and other products with undetermined shelf life in unmanned stores cannot be sold for a long time, resulting in a decrease in freshness, affecting consumer experience, and lack of effective product management and positioning services.
The shelf life information of each product is obtained from the franchise store management server through the unmanned store system and managed on the store management server. Use human sensing sensors and deep learning models to analyze customer actions, infer the products that customers are looking for, and provide guidance information on product location.
Effectively manage the shelf life of products in unmanned stores, provide discount information in advance, and improve product sales rate; through action analysis and location guidance, reduce the trouble of customers not finding products and improve shopping experience.
Smart Images

Figure CN120020850A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a store management system for operating an unmanned store, specifically an unmanned store system for managing the shelf life of products such as fruits with undetermined shelf life in an unmanned store, providing usage methods of products, and providing product positioning information for those who cannot find products. Background Art
[0002] An unmanned store refers to a store that conducts sales operations only with counters and calculators without any staff. Since it is an unmanned system, it is vulnerable to theft, so it is operated by means of closed-circuit television surveillance. Until the 2000s, the only so-called unmanned stores were bank ATMs or vending machines, but recently, familiar unmanned stores have become popular since around 2020. In particular, there are many unmanned ice cream stores. In addition, various stores such as unmanned laundries, unmanned cafes, unmanned convenience stores, or unmanned ramen stores have emerged. There are also many unmanned stores selling overseas cookies that are not sold in ordinary convenience stores.
[0003] Korean Patent Publication No. 10-2023-0080758 is an unmanned store server that stores information on unmanned store counters and products in each unmanned store counter; a radar (lidar) installed in the unmanned store to sense customers in the unmanned store; a load cell installed on the unmanned store counter to detect a change in weight in order to detect product pick-up or product return; an optimization system for unmanned store counter information and customer shopping basket information, including customer information detected by the radar, generating a customer shopping basket for the customer, and a customer processing server that confirms the customer's purchase or return.
[0004] However, for an unmanned store, for products such as fruits without a specified shelf life, if the products are not sold for a long time, there will be a problem of decreased freshness. For consumers, there will be a problem of needing to consume products with decreased freshness, and the sales store will also face a problem of decreased store trust due to providing products of lower quality.
[0005] In addition, since the store is unmanned, if there are difficulties in finding a specific product, there will also be a problem of facing difficulties due to the absence of management staff.
[0006] Prior Art Documents
[0007] Patent Documents
[0008] Patent Document 1: Korean Patent Publication No. 10-2023-0080758 Summary of the Invention
[0009] Problems to be Solved by the Invention
[0010] The purpose of the present invention is to manage the shelf life of products with uncertain shelf life, such as fruits, in unmanned stores. The shelf life of each product can be obtained from the franchise store management server and the shelf life can be managed according to the product.
[0011] In addition, if the person's movements are recognized in an unmanned store and it is believed that the product cannot be found, the purpose is to analyze the movement, estimate what product is being sought, and then provide information to guide the location of the product.
[0012] Methods used to solve the problem
[0013] To solve the above problems, the present invention aims to provide an unmanned store booth 100, that is, an unmanned store booth 100 forming an internal space for displaying products, a product shelf 200 provided for displaying products being sold, and a price indication board 300 provided on one side of the shelf 200 for displaying price information sent from a store management server 700 or for displaying price discount information sent from the store management server 700, and an unmanned store management server 700 for displaying price information on the price indication board 300.
[0014] Effects of the Invention
[0015] The present invention can obtain the shelf life of each product from the franchise store management server, manage the shelf life by product, and manage the shelf life of products with uncertain shelf life such as fruits in unmanned stores, thereby managing the quality of the products. In addition, when the shelf life is approaching, discount information is provided, which has the advantage of purchasing products at a lower price. If the product cannot be found by identifying the movement of a person in an unmanned store, it can be inferred what product is being sought by analyzing the movement, and then information introducing the location of the product can be provided, thereby minimizing the inconvenience of using the store. Brief Description of the Figures
[0016] Figure 1 This is a drawing that reflects the overall appearance of the present invention.
[0017] Figure 2 is a drawing showing the price indicator 300 of the present invention.
[0018] Figure 3 is a drawing showing 600 degrees of the present invention.
[0019] Figure 4 This is a configuration diagram of the store management server 700 of the present invention.
[0020] Description of reference numerals
[0021] 100: Unmanned store booth
[0022] 110: Entrance and Exit
[0023] 120: Cashier desk
[0024] 200: Product shelf
[0025] 300: Price indicator
[0026] 400: QR code book
[0027] 500: Human body induction sensor
[0028] 600: Direction guide part
[0029] 700: Store management server
[0030] 710: Sales expiration date management part classified by product
[0031] 720: Price guide management part
[0032] 730: Motion recognition information receiving and judging part
[0033] 740: Product location information learning
[0034] 800: Franchise store management server Detailed implementation mode
[0035] The terms or words used in this statement and the scope of claims shall not be interpreted in their ordinary or prior meanings. The inventor must, in accordance with the principle that can appropriately define the concept of the term, interpret them as meanings and concepts that conform to the technical idea of the present invention in order to best explain his invention.
[0036] The ideal embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0037] The present invention relates to an unmanned store system, specifically to an unmanned store system, aiming to manage the shelf life of products such as fruits with undetermined shelf life in an unmanned sales area, provide the usage method of the products, and provide information introducing the product location for those who cannot find the products.
[0038] Figure 1 It is a drawing reflecting the overall appearance of the present invention.
[0039] The present invention is composed of an unmanned store booth 100, a product shelf 200, a price indicator 300, a two-dimensional code book 400, a human body induction sensor 500, a direction inside 600, a store management server 700, and a franchise store management server 800.
[0040] Each component is specifically described as follows:
[0041] The unmanned store booth 100 is an internal space formed for displaying products. One side of the unmanned store booth 100 is for using the unmanned store to form an entrance and exit 110.
[0042] The unmanned store booth 100 may form a cashier's desk 120 for consumers to calculate products on one side, and no drawing has been made for this. In addition, a server is built into the cashier's desk 120, and consumer information containing basic matters of consumers can be input to provide points, discount offers, etc. The input information can also be sent to the store management server 700.
[0043] The consumer information can input the information of the products purchased by the consumer when purchasing products, or the product evaluation scores of the purchased products previously purchased by the consumer can be input.
[0044] The product shelf 200 is for displaying the products on sale. A certain number of the product shelves 200 are formed, and their forms and quantities may vary according to the types and quantities of the products.
[0045] The price indicator 300 is located on one side of the shelf 200 and is linked with the store management server 700 to display the price information sent from the store management server 700 or the price discount information sent from the store management server 700.
[0046] As Figure 2 shown, when the price guide 300 sends price information from the store management server 700, it may display the price information as shown in Figure 2 (a), and if price discount information is sent, it may also display the price discount information as shown in Figure 2 (b).
[0047] The two-dimensional code part 400 is located on one side of the shelf 200 and is used to input menu information.
[0048] The information of the above menu is information about the storage method, cooking method, etc. of the menu, and it can provide information about any product to the consumer.
[0049] The unmanned store booth shooting part 500 is for shooting videos inside the unmanned store booth 100 and transmitting them to the store management server 700.
[0050] The unmanned store shooting part 500 is located on the inner side of the unmanned store booth 100 and allows the whole inside of the unmanned store 100 to be shot.
[0051] The in-direction part 600 is linked with the store management server 700, receives product position information from the product position information providing part 740 of the store management server 700, and displays direction indicator lights according to the sent product position information.
[0052] Figure 3It is to illustrate that according to the sent product position information, the display mode of the direction indicator is different. Divide the direction into intervals 600a, 600b, and 600c within 600. If the product shelf 600 is located at a position corresponding to each interval, such as Figure 3 as shown in (a) of
[0053] Figure 3 If the product shelf 200 near the product position displays the block of 600b, the direction indication in the 600b interval may point to the product shelf 200 close to the 600b interval.
[0054] In the case of (b) of
[0055] If the product position is on the product shelf 200 closer to the block marked 600c, the direction mark in the 600c interval may point to the product shelf 200 closer to the 600c interval. Figure 4 The store management server 700 is to transmit the price information displayed on the price indicator 300, receive the captured video from the unmanned store booth shooting unit 500, identify people from the transmitted video, separate them from the internal background of the unmanned store booth, and estimate the actions of the extracted people.
[0056] Specifically, each component is described as follows:
[0057] The shelf life management department 710 of each product is to receive the shelf life information of each product containing the shelf life information of each product displayed in the product shelf 200 from the franchise store management server for managing franchise stores, judge whether each product has expired, and if the remaining date of the shelf life of each product is within the set range, judge that the shelf life is approaching, and send the judged product shelf life approaching information to the guiding price management department 720.
[0058] The price indicator management department 720 is to receive price information from the franchise store management server for managing franchise stores and send the transmitted price information to the price indicator 300. If it receives the approaching information of the circulation period of the product from the circulation period management department 710 of each product, it sends the approaching information of the circulation period to the price indicator 300 instead of the price information.
[0059] The price indicator management department 720 receives the consumer information of consumers visiting the unmanned store booth 100, and considering the customer satisfaction of the product and the remaining date of the shelf life, can calculate the price discount information of the product through Mathematical Formula 1.
[0060]
Mathematical Formula 1
[0061]
[0062] (Where K refers to the product price discount information, Z refers to the product selling price information, H refers to the remaining date of the shelf life, w refers to the average value of the product evaluation scores in the consumer information, and v refers to the consumer level score).
[0063] (Where K refers to the product price discount information, Z refers to the product selling price information, H refers to the remaining date of the shelf life, w refers to the average value of the product evaluation scores in the consumer information, and v refers to the consumer level score).
[0064] Among them, the average of the product evaluation scores in the consumer information means that consumers who purchase the selected product input the evaluation scores of the product into the server in the cash register 120 of the unmanned store booth 100, and the average value of the input product evaluation scores can be a score between 1 and 10 points. The consumer level score divides the level into 1 to 5 points according to the frequency set by the product purchase frequency. The higher the frequency of repurchasing the selected product, the higher the score.
[0065] In addition, in order to prevent the consumer level score and the average value of the product evaluation scores from being reflected as relatively large values compared to the shelf life or the change range of the discount price being too large, the change range can be adjusted through the inverse function of sinh.
[0066] For example, if the selling price of product A is 5,000 won, the remaining date of the shelf life of product A is 3 days, the average value of the product evaluation scores of product A in the consumer information is 7 points, and the consumer level score is 3 points, then the product price discount information is as follows.
[0067]
[0068] Based on the price discount information calculated in this way, a reasonable discount amount is calculated, which is beneficial for consumers to purchase discounted products at a reasonable price.
[0069] The motion recognition information receiving and judging unit 730 receives the captured video from the unmanned store booth shooting unit 500, recognizes people in the transmitted video, separates them from the internal background of the unmanned store booth, and when it is judged that a specific product is being searched for based on the extracted motion of the people, it speculates what product is being searched for based on the motion.
[0070] The motion recognition information receiving and judging unit 730 recognizes people through a pre-learned object recognition model and divides them into bounding boxes. The deep learning model learned through a large amount of learning data can recognize people in the unmanned store booth and separate them from the background. In addition, people can also be separated from the background and only people can be recognized through the existing person-background separation recognition technology.
[0071] In addition, the extracted human movements are used to collect the movements and movement paths of people entering and leaving the unmanned store booth through a pre-learned human movement recognition model. The deep learning model learned through a large amount of learning data can infer the movement paths of the identified people in the unmanned store booth, which is used as an indicator for judging what products to look for.
[0072] That is to say, if the actions of people who purchase product A are mainly actions of K and move along the actions of L, then if the extracted human actions move along the actions within the range of action K and the actions within the range of action L, it can be speculated that product A is being sought.
[0073] The product position information providing unit 740 is to send product position information including the speculated product position to the direction interior 600 through the movement recognition information receiving and judging unit 730 to guide the position of the speculated product.
[0074] The store management server 700 may also include a product warehousing management unit 750, which is to compare the number of products displayed on the product shelf 200 with the number of products sold to judge whether the products in the product shelf 200 are exhausted.
[0075] That is to say, the number of products displayed on the product shelf 200 is the value input by the management staff at the time of product warehousing, and the number of products sold receives information about the number of products from the cash register 120 of the unmanned store booth 100. Among the products displayed on the received product shelf 200, it is possible to judge whether the products are exhausted by comparing the number of a specific product selected to understand the remaining quantity with the number of a specific product selected from the products calculated at the cash register 120 to understand the remaining quantity.
[0076] In addition, the product warehousing management unit 750 can master the product exhaustion cycle index to judge which products are exhausted relatively quickly. The product exhaustion cycle index can be calculated by Mathematical Formula 2. In this way, for products that are exhausted relatively quickly, additional quantities can be prepared in advance so that consumers can use the products without inconvenience. For products with a relatively small product exhaustion cycle index, the number of products sold can be reduced, thereby minimizing the products that have not been sold for a long time and maintaining the freshness of the products.
[0077]
Mathematical Formula 2
[0078]
[0079] (Where Ni is the product depletion index of product i, Ti is the number of people staying within the set radius of the location where the product shelf of product i is located within a specified time period, Si is the sales quantity of product i within the set time period, and Mi is the number of people visiting the location where the product shelf is located after product i is depleted within the set time period).
[0080] The set time period can be in hours or in hours.
[0081] In addition, within the set radius of the location where the product shelf of product i is located, the number of people staying within the specified time is the number of people interested in product i, including those who are considering whether to purchase or are purchasing product i. The number can be obtained by analyzing the movement of people extracted by the motion recognition information receiving and judging unit 730. The number of people staying within the set radius of the location where product i is located within the set time is the optimal number.
[0082] Within the set time period, after product i is depleted, the number of people visiting the location where the product shelf is located can be obtained by analyzing the actions of people extracted by the motion recognition information receiving and judging unit 730, or it can be the number of people visiting in front of the product shelf after product i is depleted.
[0083] For example, assume the set time is one week. The number of people staying within 2 meters of the location where the product shelf displaying "apples" is located for more than 30 seconds within one week is 15, the number of "apples" sold within one week is 5, and the number of people visiting the location where the product shelf is located after "apples" are depleted within one week is 5. Then the product depletion index of "apples" is:
[0084]
[0085] Assume the set time is one week. The number of people staying within 2 meters of the location where the product shelf displaying "oranges" is located for more than 30 seconds within one week is 20, the number of "oranges" sold within one week is 8, and if the number of people visiting the location where the product shelf is located after "oranges" are depleted within one week is 5. Then the product depletion index of "oranges" is:
[0086]
[0087] That is to say, compared with the number of people visiting presumably to purchase the product, the higher the sales quantity of the product and the more people who return after purchasing the product, the lower the product depletion index.
[0088] The franchise store management server 800 is for managing franchise stores and transmitting the shelf life information, price information, and price discount information of each product.
[0089] As described above, the embodiments described in this list and the configurations shown in the drawings are only one of the most ideal embodiments of the present invention and do not represent all the technical ideas of the present invention. Therefore, it should be understood that there may be various homogeneous and deformed examples that can replace them.
Claims
1. A store management system for unmanned store operation, characterized in that: The store management system for unmanned store operation includes: a price bulletin board (300) for displaying price information sent from a store management server (700) or displaying price discount information sent from a store management server (700); and A store management server (700) for sending price information or price discount information to be displayed in the pricing guide (300), The store management server (700) includes: The product shelf life management unit (710) receives shelf life information of each product displayed on the product shelf of the store from the franchise store management server for managing the franchise stores, including the shelf life information of each product, and determines whether the shelf life of each product has expired. If the remaining shelf life of each product is within a set range, it is considered that the shelf life has expired, and the determined product shelf life approaching information is sent to the price guide management unit (720); and The price bulletin board management unit (720) receives price information from the franchise store management server used to manage franchise stores, and sends the sent price information to the price bulletin board (300). When receiving information about the approaching circulation period of a product from the circulation period management unit (710) of each product, it sends the price discount information of the sent product to the price bulletin board (300) in place of the price information.
2. The store management system for unmanned store operation according to claim 1, characterized in that: The store management server (700) includes a product warehousing management unit (750) for determining whether the product is exhausted in the product shelf (200) by comparing the number of products displayed on the product shelf (200) with the number of products sold. The price sign management unit (720) receives consumer information about consumers who visit the unmanned store booth (100), and calculates the price discount information of the product through mathematical expression 1 according to the customer satisfaction and the remaining date of the shelf life of the product. [Mathematical formula 1] is (in, K refers to product price discount information, Z refers to product sales price information, H refers to the remaining shelf life, w refers to the average product evaluation score in consumer information, and v refers to the consumer rating score). In the product storage management unit (750), the number of products displayed on the product shelf (200) is a value input by a manager when the products are stored, the number of products sold is information about the calculated number of products sent from the cashier counter (120) of the unmanned store booth (100), the number of specific products selected on the received product shelf (200) to understand the remaining number of displayed products is compared with the remaining number of products selected from the cashier counter (120) to determine whether the product quantity has been exhausted, The product warehousing management unit (750) learns the product depletion cycle index and determines which products are depleted relatively quickly. The product depletion cycle index is calculated by mathematical formula 2. For products that are depleted relatively quickly, additional quantities should be prepared in advance. For products with relatively low product depletion cycle indexes, the sales quantity of the products should be reduced. [Mathematical formula 2] is (where Ni is the product exhaustion index of product i, Ti refers to the number of people who stay within the set radius of the place where the product shelf where product i is located is located within the set time period, Si refers to the sales volume of product i within the set time period, and Mi refers to the number of people who visit the place where the product shelf is located after product i is exhausted within the set time period).
Citation Information
Patent Citations
System for optimizing cashierless store shelf information and customer shopping cart information
KR1020230080758A