KR20220116831A - Intelligent Untact Pick-up Box System - Google Patents

Abstract

Disclosed is an intelligent non-face-to-face pickup box system including at least one non-face-to-face pickup box and a pickup box control device. The pickup box system of the present invention performs the steps of: allowing a controlling unit of a pickup box control device to store an order number and order information in a memory unit; predicting the time when a customer arrives at a store, in which a pickup box is located, based on a location of a customer terminal; analyzing an image taken by a camera installed in the pickup box to determine the order number of the product brought into the pickup box; and in a situation in which it is determined that a plurality of customers are scheduled to arrive at the store for the pickup box or any pickup box among the pickup boxes, generating an instruction, when it is predicted that a customer, other than the customer who ordered the product brought into the pickup box, will arrive at the store first, to withdraw the product brought into the pickup box and bring a product corresponding to the order number of the customer predicted to arrive at the store first into the pickup box. Since the intelligent non-face-to-face pickup box system according to the present invention analyzes movement of an orderer and adaptively supports an order packaging processing sequence according to the expected arrival time, there is no disruption in customer response while using a small number of non-face-to-face pickup boxes such that efficiency is excellent. In addition, by automatically inspecting and handling errors for ordered menu and menu packaged and stored in the pickup box, a workload of a worker can be reduced.

Description

Intelligent Untact Pick-up Box System

The present invention relates to an intelligent non-face-to-face pickup box system, and more particularly, by analyzing the movement of the orderer and adaptively supporting the order packaging processing sequence according to the expected arrival time, thereby serving customers while using a small number of non-face-to-face pickup boxes. It is related to an intelligent non-face-to-face pickup box system that has excellent efficiency because there is no setback, and reduces the workload of workers by inspecting and handling errors on ordered menus and menus packaged and stored in pickup boxes.

With respect to a store such as a restaurant, a customer may visit the store after placing an order by phone or online, pay a price for the ordered product, and retrieve the ordered product. In this case, in the case of an online order, the online ordering platform may be used, and in such case, the price for the product may be paid at the time of the online order. An online ordering platform is a server of a business that intermediaries customers and food businesses, and is also commonly called a delivery app. Examples of online ordering platforms or delivery apps include Baedal Minjok and Yogiyo in Korea. Customers can order food such as hamburgers through the online ordering platform.

On the other hand, recently, due to the spread of Corona Virus Disease-19 (COVID-19), as wearing masks and social distancing have become commonplace, even when customers visit stores and find products ordered online, they are not face-to-face. this is being requested

Patent Publication No. 10-2008-0018392 (published on February 28, 2008) Patent Publication No. 10-2012-0020777 (published on March 08, 2012) Patent Registration No. 10-1450042 (Registered on Oct. 06, 2014) Patent Registration No. 10-1797153 (Registered on Nov. 07, 2017)

Accordingly, it is an object of the present invention to analyze the movement of the orderer and adaptively support the order packaging processing sequence according to the expected arrival time, so that there is no disruption in customer response while using a small number of non-face-to-face pickup boxes, so the efficiency is excellent, It is to provide an intelligent non-face-to-face pickup box system that reduces the workload of workers by inspecting and handling errors on menus and menus packaged and stored in pickup boxes.

The intelligent non-face-to-face pickup box system according to the present invention for achieving the above object is a box body composed of an upper surface, a lower surface, a front surface, a rear surface, a left surface and a right surface and is installed on the rear surface of the box body, , a product entry door that the business operator can open to bring in the product provided to the customer, a product exit door that is installed on the front side of the box body and can be opened when the customer enters a password, is installed inside the box body At least one non-face-to-face pickup box including at least one camera that captures the appearance of a product accommodated in the box body to obtain an image or a video, and a control unit and a program to be executed by the control unit are stored in the program execution and a pickup box control device including a memory unit for storing necessary and data generated as a result of the execution of the program.

When the control unit of the pickup box control device receives an order from the order receiving server, storing the order number and order information in the memory unit, the customer based on the location of the customer terminal receiving the order to the order receiving server Predicting the time when a customer who owns a terminal will arrive at the store in which the pickup box is located, analyzing the image captured by the camera installed in the pickup box to determine the order number of the product brought into the pickup box, and In a situation where it is determined that a plurality of customers are scheduled to arrive at the store with respect to the pickup box or any of the pickup boxes, a customer other than the customer who ordered the product delivered in the pickup box arrives at the store first In this case or when it is predicted to arrive earlier, instruct the customer to withdraw the product received in the pickup box and load the product corresponding to the customer's order number that arrives at the store first or is expected to arrive earlier in the pickup box. Steps to create

The control unit of the pickup box control device further executes the step of determining the type and quantity of products introduced into the pickup box by analyzing the image or video taken by the camera installed in the pickup box, and determining by the step When the type and quantity of the manufactured product are different from the type and quantity of the product according to the order number stored in the memory unit, an error alarm may be issued.

A scale is further installed in the pickup box, and the control unit of the pickup box control device analyzes the image captured by the camera installed in the pickup box to determine the order number of the product introduced into the pickup box. When the weight of the product corresponding to the order number determined by When it does not match the weight of the product, it is desirable to issue an error alarm.

As a method of analyzing the weight of the packaging menu, the image including the external characteristics of the packaging container and the weight of the container are initially stored in memory, and information such as menu information and the image and weight of the packaged order are stored in the memory when an order is placed. For each order, the menu and weight information are identified and accumulated, and the weight characteristic information for the menu is advanced. Whenever a menu order is made for a menu whose weight information has not yet been identified, the weight of the known menu is calculated from the total weight of the package. It calculates the difference and learns the weight characteristics of individual menus to continuously optimize and upgrade the weight information for the menu.

If a customer other than the customer who ordered the product delivered in the pickup box arrives at the store first or is expected to arrive first, the estimated arrival time of the customer who arrives at the store first or is expected to arrive earlier is 5 minutes Within, the estimated arrival time of the customer who ordered the product brought into the pickup box may be the case where it is determined that the customer arrives at the store first or is 3 minutes or more later than the expected arrival time of the customer that is expected to arrive earlier.

The control unit of the pickup box control device withdraws the product introduced into the pickup box because the other customer, other than the customer who ordered the product brought into the pickup box, arrives at the store first or is expected to arrive earlier When an instruction is generated to bring a product corresponding to the order number of the customer who arrives first or is expected to arrive earlier in the pickup box, the customer of the customer who arrives at the store first or is expected to arrive earlier A notification may be sent to the terminal to notify that the product has been delivered to the pickup box.

On the other hand, if the customer terminal of the customer who ordered the product that was drawn into the pickup box and ordered the product was previously notified that the product has been drawn into the pickup box, the product is withdrawn from the pickup box to the customer terminal and is in a waiting state A notice may be given to inform you that

When the control unit of the intelligent non-face-to-face pickup box system according to the present invention receives an order from the order receiving server, storing the order number and order information in the memory unit, the customer terminal receiving the order to the order receiving server Predicting the time when the customer who owns the customer terminal will arrive at the store where the pickup box is located based on the location of , and displaying the expected arrival time of the customer at the store may be executed.

In a situation in which it is determined that a plurality of customers are scheduled to arrive at the store with respect to the pickup box or a pickup box among the pickup boxes, the control unit has the shortest expected store arrival time among the products of the customers whose packaging has been completed. The step of displaying an instruction to bring the customer's product into the pickup box may be executed.

The intelligent non-face-to-face pickup box system according to the present invention analyzes the movement of the orderer and adaptively supports the order packing processing sequence according to the estimated arrival time, so that there is no disruption in customer service while using a small number of non-face-to-face pickup boxes. It is excellent, and it can reduce the workload of workers by automatically inspecting and handling errors on the menu ordered and packaged and stored in the pickup box.

1 and 2 are diagrams illustrating an example of a network configuration including an intelligent non-face-to-face pickup box system according to the present invention.
3 is a diagram schematically showing the external appearance of a non-face-to-face pickup box applied to the intelligent non-face-to-face pickup box system according to the present invention.
4 is a view showing a state in which a camera is installed on the inner surface of the upper surface of the non-face-to-face pickup box applied to the intelligent non-face-to-face pickup box system according to the present invention.
5 is a diagram showing a schematic configuration of an intelligent non-face-to-face pickup box system according to the present invention.
6 is a network configuration diagram for explaining the operation of the intelligent non-face-to-face pickup box system according to the present invention.
7 is a view showing an exemplary display for explaining the operation of the intelligent non-face-to-face pickup box system according to the present invention.
8 is a schematic diagram illustrating the creation of a dynamic self-learning reference data set and integration with a reference data set initially carried out at the factory to form an integrated reference data set in the intelligent non-face-to-face pickup box system according to the present invention.
9 is a flowchart for analyzing the weight characteristics of the packaging menu in the intelligent non-face-to-face pickup box system according to the present invention.

Hereinafter, the present invention will be described in detail with reference to the drawings.

As shown in FIG. 1 , the intelligent non-face-to-face pickup box system 10 according to the present invention includes a pickup box control device 100 and one or more non-face-to-face pickup boxes 200 . The pickup box control device 100 may be connected to the order reception server 300 by wire or wirelessly. The order receiving server 300 may be connected to the customer terminal 400 through the Internet.

The customer can access the order reception server 100 through the customer terminal 400 to order and pay for products online. In this case, the product may be a hamburger product sold at a fast food restaurant or a set product including a hamburger. In addition, the product may be cosmetics or makeup tools sold in a cosmetics shop, and may be various other products.

In FIG. 1 , when the pickup box control device 100 and the order reception server 300 are managed by the same store entity, the order reception server 300 receives an order for product purchase from the customer terminal 400 , A password required to open the non-face-to-face pickup box 200 in order to pick up the product ordered by the customer may be transmitted to the customer terminal 400 together with the order number and order details. At this time, the order reception server 300 may generate an order number.

Meanwhile, as shown in FIG. 2 , the pickup box control device 100 and the order reception server 300 may be managed by different entities. In this case, the order reception server 300 and the pickup box control device 100 may be connected to each other through the Internet. In this case, the order reception server 300 may be an online ordering platform. In this case, when an order for product purchase is received from the customer terminal 400, the order reception server 300 notifies the customer terminal 400 that the order reception for such order details has been completed, and the pickup box 200 for The password and order number are not notified to the customer terminal 400 . The order number, order details, and a password for the pickup box 200 may be notified to the customer terminal 400 by the pickup box control device 100 described below.

The order received by the order reception server 300 in FIGS. 1 and 2 is transmitted to the pickup box control device 100 . At this time, when the order reception server 300 generates an order number and password, the order number and password generated together with customer information and order details are also transmitted to the pickup box control device 100, whereas the order reception server 300 In the case in which the order number and password have not been generated, only customer information and order contents are transmitted to the pickup box control device 100 . The customer's order received by the pickup box control device 100 is delivered to the kitchen of the store. The chef working in the kitchen then prepares the food product by cooking according to those orders.

When the food product ordered by the customer is completed, the store operator puts the food product in the non-face-to-face pickup box 200 . Then, the customer can open the pickup box 200 by inputting a password into the pickup box 200 containing the food product ordered by the customer and go to his or her own food product. Accordingly, the customer can have the food product he ordered without having to face the business operator. In other words, customers can practice perfect social distancing for business operators, thereby preventing the transmission of diseases such as coronavirus infection-19 (COVID-19). This is also true for business owners.

In this process, when the number of pickup boxes 200 is small compared to the number of orders from the customer, the business efficiency of the store may be affected depending on how the customer's order is processed, and customer satisfaction may also be affected. Specifically, for example, if an order is first processed and placed in the pickup box 200, no problem occurs when the customer visits the store in the order in which the order is placed. However, if the customer who ordered first arrives at the store late, while the customer who ordered later arrives at the store early, the customer who ordered later but arrived early before the customer who ordered first visits the store and finds his or her order product You will not be able to find your ordered products. In addition, such customers can only feel dissatisfied because they have to wait in a situation where they are not even informed about when their order will be delivered.

Accordingly, the present invention has been devised to solve the above problems. Therefore, the present invention analyzes the movement of the orderer and adaptively supports the order packaging processing sequence according to the expected arrival time, so that there is no disruption in customer service while using a small number of non-face-to-face pickup boxes, and there is no problem in customer service. to provide a system.

To this end, the intelligent non-face-to-face pickup box system 10 of the present invention includes a pickup box control device 100 and one or more non-face-to-face pickup boxes 200 .

The pickup box control apparatus 100 controls one or more non-face-to-face pickup boxes 200 , and includes a control unit 110 , a memory unit 120 , and a communication unit 130 as shown in FIG. 5 . At this time, the memory unit 120 stores the program executed by the control unit 110 and also stores data necessary for the execution of the program and generated as a result of the program execution. The communication unit 130 provides a communication function with the order reception server 300 and the non-face-to-face pickup box 200 .

The non-face-to-face pickup box 200 includes a box body 210 as shown in FIG. 3 . The box body 210 has an upper surface 211 , a lower surface 212 , a front surface 213 , a rear surface 214 , a left surface 215 , and a right surface 216 , and is empty inside. That is, the box body 210 has a rectangular parallelepiped shape with an empty interior.

A product inlet door (not shown) is installed on the rear surface 214 of the box body 210 . The product entry door is for introducing the product provided by the business operator to the customer into the box body 210 . If access to the product entry door is permitted only to the operator, the product entry door and the rear 214 may not be present. authority must be granted. That is, only the operator should be able to lock and unlock the product entry door. The product entrance door may have a structure that can lock or unlock the product entrance door by means of a manual key, and is automatically locked when the product entrance door is closed It may have a structure that is released.

A product take-out door 220 is installed on the front side 213 of the box body 210 . The product take-out door 220 is for a customer to take out a product inside the box body 210 . A handle 221 may be installed on the product take-out door 220 . In order for the customer to open the product take-out door 220 , the password notified to the customer terminal 400 must be input. The password may be notified by the order reception server 300 to the customer terminal 400 , or the pickup box control device 100 may notify the customer terminal 400 .

In order for the customer to input a password, a password input unit 251 may be installed in the product take-out door 220 . The password input unit 251 may be installed virtually by, for example, the touch screen 250 . The touch screen 250 may include, for example, a password input unit 251 and an advertisement display unit 252 installed therein. In this case, the password input unit 251 may be provided with, for example, a number pad and an enter key. The advertisement display unit 252 may display an image or a moving picture advertising the products of the business operator. On the other hand, one part of the advertisement display unit 252 may be provided with a password input unit appearance switching button 253. The password display unit 251 may disappear when the password input unit appearance switching button 253 is pressed and a predetermined time, for example, 10 seconds passes after being displayed on the touch screen 250, or when the customer presses the password and presses the enter key. Then, the entire screen of the touch screen 250 is switched to the advertisement display unit 252 .

Meanwhile, the password may be a barcode or a QR code. In this case, a barcode or QR code reader is provided on the front 213 of the product take-out door 220 or the box body 210 .

A camera 230 may be installed inside the box body 210 . The camera 230 is for obtaining an image or a moving picture by photographing the appearance of the product accommodated in the box body 210 . A mirror 217 may be installed on one or both of the front 213 , the back 214 , the left side 215 , and the right side 216 of the box body 210 . The camera 230 may be installed inside the upper surface 211 of the box body 210 as shown in FIG. 4 . If necessary, the camera 230 may be installed on any one or all of the inner surfaces of the front surface 213 , the rear surface 214 , the left surface 215 , and the right surface 216 in addition to the upper surface 211 . By analyzing the image or video captured by the camera 230 , the order number of the product introduced into the pickup box 200 may be grasped, and the type and quantity of the product may also be identified. In order to prevent a product other than the ordered product from being introduced into the pickup box 200 , a scale 240 may be installed inside the box body 210 , particularly on the lower surface 212 of the box body 210 .

The above functions are performed by the pickup box control device 100 . To this end, the control unit 110 of the pickup box control device 100, as shown in FIG. 5, the product take-out door 220, the camera 230, the scale 240, and the front touch installed in the pickup box 200, Control screen 250 or receive data from them.

When the control unit 110 of the pickup box control device 100 receives a customer's order from the order reception server 300 , the control unit 110 stores such order information in the memory unit 120 . In this case, the order information includes a mobile communication phone number and order details as customer information. The order contents may include the type and quantity of food products. When the order reception server 300 does not generate an order number and a password, the control unit 110 generates an order number and a password and transmits it to the customer terminal 400 . The generated order number and password are stored in the memory unit 120 together with the order information. The control unit 110 also notifies the customer terminal 400 who ordered the food product if the food product is introduced into a certain pickup box 200 along with the identification number of the corresponding pickup box 200 .

As shown in FIG. 6 , the control unit 110 controls the customer who owns the customer terminal 400 based on the location of the customer terminal 400 that separately received the order from the order reception server 300 as shown in FIG. ) to predict the time it will arrive at the store where it is located. To this end, the control unit 110 of the pickup box control device 100 may directly perform a procedure for periodically tracking the location of the customer terminal 400, and as shown in FIG. 6, a dedicated server performs such a procedure. and the control unit 110 may receive such result data periodically.

Specifically, the location information service server 600 may periodically determine the locations of the customer terminal #1 (401), the customer terminal #2 (402), and the customer terminal #3 (403). In this case, the period may be 30 seconds or 1 minute. The positions of the identified customer terminals 400 may be displayed on a map. The location information service server 600 may also recognize the location of the store. The location information service server 600 may calculate the distance between the store and the customer terminals 400 and also calculate the time required when the customer visits the store on foot or by car. At this time, the location information service server 600 analyzes the movement of the customer terminal according to time to determine whether the customer moves on foot or by car, thereby more accurately grasping the expected arrival time of the store. Thereafter, the location information service server 600 may transmit map information in which the locations of the customer terminals 400 are displayed on the map and information about the expected arrival time of the store to the pickup box control device 100 .

The controller 110 may display the estimated arrival time of the store determined as described above on the management unit of the store and the display device 700 provided in the kitchen. Then, the chef of the kitchen can increase the work efficiency of the store by determining the cooking order for the ordered food products with reference to the customer's expected arrival time at the store. In addition, the store management unit not only increases the work efficiency of the store, but also reduces or eliminates the customer's waiting time by determining the order in which the finished food products are brought into the assigned pickup box 200 by referring to the customer's expected arrival time at the store. It can improve customer satisfaction.

In addition, the pickup box control device 100 instructs the pickup box 200 to introduce the food product of the order number ordered first or the food product of the customer who is expected to arrive first, but after that the situation changes and the pickup box ( When a customer other than the customer who ordered the product brought in 200) arrives at the store first or is predicted to arrive first, the order of entering the pickup box 200 may be changed.

First, the control unit 110 of the pick-up box control device 100 needs to know what the product introduced into the pick-up box 200 is. To this end, the controller 110 may receive and analyze the image captured by the camera 230 installed in the pickup box 200 to determine the order number of the product introduced into the pickup box 200 . The control unit 110 may know which product is brought into the pickup box 200 by referring to the type and quantity of the product corresponding to the corresponding order number stored in the memory unit 120 .

The control unit 110 may determine whether the product introduced into the pickup box 200 matches the product ordered by the customer, otherwise may issue an error alert. Specifically, the control unit 110 controls the weight of the product introduced into the pickup box 200 measured by the scale 240 installed in the pickup box 200, that is, the actually measured weight and the camera installed in the pickup box 200 ( If the weight calculated based on the type and quantity of the product stored in the memory unit 120, which corresponds to the order number of the product received in the pickup box 200 identified by analyzing the image taken by the 230, matches the pickup box ( While it is considered that the correct product has been delivered to 200), if it does not match, it may be considered that the wrong product has been delivered to the pickup box 200 . That is, it may be considered that a product different from the product corresponding to the order number stored in the memory unit 120 is drawn into the pickup box 200 . When such an error occurs, the controller 110 may issue an error alert. These errors can be caused by mistakenly cooking other food products in the kitchen without cooking them according to the order, or omitting some of them.

On the other hand, in addition to or separate from the weight measurement by the scale 240, the control unit 110 analyzes the image or video taken by the camera 230 installed in the pickup box 200, The type and quantity can be determined. The determined product type and quantity may be compared with the product type and quantity corresponding to the order number stored in the memory unit 120 . As a result of such comparison, if they match, it is considered that the correct product has been delivered to the pickup box 200 , whereas if they do not match, it may be considered that the wrong product has been delivered into the pickup box 200 . When such an error occurs, the controller 110 may issue an error alert.

A method of analyzing the weight of the packaging menu will be described with reference to FIG. 9 .

The packaging menu is divided into packaging containers and ordering menus. Initially, an image including the external characteristics of the packaging container and the weight of the container are stored in the memory. As shown in FIG. 9 , if the packaging container is not stored in the memory, the packaging container is first registered in the memory, and the weight of the packaging container is also registered in the memory.

When an order is placed, menu information and information such as the image and weight of the packaged order are stored in the memory. The menu information may be stored in memory before an order is placed. The weight of a single menu can be determined by measuring the weight when an order for a single menu is placed. That is, when a specific single menu order is packaged in a container with a known container weight and loaded into the pickup box, the weight of the single menu can be calculated by subtracting the weight of the container from the total weight of the single menu order. The weight of the single menu thus calculated is stored in the memory.

On the other hand, the weight of the menu for which the weight value has not yet been grasped can also be grasped by the following method. When a menu whose weight value has not yet been determined is ordered together with a menu whose weight value has been determined, that is, when a complex menu is ordered, the complex menu is packaged in a container with known weight and put into the pickup box. By subtracting the tare weight and the menu weight for which the weight value is determined from the total weight of the menu order, the menu weight for which the weight value has not yet been determined may be determined. The detected menu weight is stored in the memory, and the menu information is updated and advanced.

The weight of a single menu of known weight is then verified when a single menu order or a complex menu order including that single menu occurs, thereby optimizing the weight value for the individual menu. In addition, the menu information stored in the memory is updated and advanced by understanding and storing the weight of the single menu by the analysis of the single menu described above and the weight of the single menu by the analysis of the complex menu and storage.

On the other hand, as described above, it is determined whether the corresponding menu weight characteristic is normal, and when it is determined that it is not, an error alarm is issued.

Next, for convenience of explanation, only one pickup box 200 is connected to the pickup box control device 100 , and as shown in FIG. 6 , three customers order simultaneously or sequentially and are shown in FIG. 7 . The present invention will be described with reference to an example progressed according to a given scenario. The status shown in FIG. 7 may be displayed by the display device 700 . This example is a situation in which it is determined that a plurality of customers are scheduled to arrive at a store with respect to one pickup box or a pickup box from among a plurality of pickup boxes.

As shown in FIG. 7 , at some point in time when an order has just been placed, customer 1 using customer terminal #1 401 placed an order with order number 21, and the expected time to arrive at the store from customer 1's location is 20 minutes. was predicted to Similarly, customer 2 using customer terminal #2 (402) ordered with order number 22, and the expected time to arrive at the store from customer 2's location was predicted to be 25 minutes, and customer 2 using customer terminal #3 (403) Customer 3 placed an order with order number 23 and the expected time to arrive at the store from Customer 3's location was predicted to be 30 minutes (top part of FIG. 7 ). Accordingly, the chef in the kitchen started cooking in the order of the predicted time.

After at least 22 minutes, the dishes of customers 1, 2, and 3 were all finished and only the packaging remained, but customer 1 (customer number 21) did not move at all, and the estimated time of arrival at the store was still predicted to be 20 minutes. On the other hand, Customer 2 (Customer Order No. 22) moved toward the store at a slower-than-expected speed and the estimated time to arrive at the store was predicted to be 10 minutes, and Customer 3 (Customer Order No. 23) moved toward the store at the expected speed and arrived at the store. The expected time was predicted to be 8 minutes (interrupted portion of FIG. 7 ). At this time, since customer 3's dish is complete and the expected arrival time at the store is the shortest, the control unit 110 can expect that customer 3 will arrive at the store first, so the control unit 110 sends Yes to the display device 700 . For example, an instruction such as "Put order number 23 into the pickup box first" may be displayed (interrupted portion of FIG. 7 ). Then, the store employee may put the order number 23 product into the pickup box 200 according to the instruction of the control unit 110 . When the product of customer 3, ie, product of order number 23, is received into the pickup box 200, the control unit 110 may recognize that the product of order number 23 has been received by the above procedure, so that the customer terminal of customer 3 #3 (403) may be notified that the product has been drawn into the pickup box (200). On the other hand, as described below, even if the corresponding product is introduced into the pickup box 200 , since it may be changed later, the control unit 110 controls the customer terminal 400 when the corresponding product enters the pickup box 200 . ), instead of notifying the customer, wait and notice that the product has been delivered to the pickup box 200 have.

Next, after at least 7 minutes has elapsed again, customer 2 moves towards the store at the expected speed, with an estimated time of arrival at the store of 3 minutes, while customer 3, for personal reasons or other reasons, moves towards the store at a slower speed than expected. Due to the slow movement, the estimated time of arrival at the store was predicted to be 7 minutes. On the other hand, the estimated time of arrival at the store of customer 1 was predicted to be 15 minutes (lower part of FIG. 7 ). This case corresponds to a case where it is predicted that a customer (customer 2) other than the customer (customer 3) who ordered the product brought into the pickup box 200 will arrive at the store first. In this case, the control unit 100 may display, for example, an instruction such as "Please change the product to order number 22 and put it in the pickup box" on the display device 700 (lower part of FIG. 7 ). The store's staff can then implement those instructions. At this time, in order to put the order number 22 product into the pickup box 200, it is natural that the first order number 23 product must be withdrawn.

In the above, when it is predicted that a customer (customer 2) other than the customer (customer 3) who ordered the product delivered in the pickup box 200 will arrive at the store first, the customer (customer 2) who is expected to arrive at the store first Estimated arrival time is within 5 minutes, and the estimated arrival time of the customer (customer 3) who ordered the product delivered in the pickup box 200 is 3 minutes or longer than the estimated arrival time of the customer (customer 2) who is expected to arrive at the store first It may be set to a case where it is judged to be late. However, these settings are exemplary, and it would be desirable to allow more suitable time settings to be changed while operating according to the circumstances of each store.

On the other hand, as described above, when the product introduced into the pickup box 200 is changed from customer 3's product (order number 23 product) to customer 2's product (order number 22 product), the control unit 110 performs the same procedure as above can recognize that the product of order number 22 has been received, so that the customer terminal #2 (402) of the customer 2 can be notified that the product has been received in the pickup box (200). In addition, the control unit 110 to the customer terminal #3 (403) of customer 3 who ordered the product (order number 23 product) that was drawn in and withdrawn into the pickup box 200 before that the product was introduced into the pickup box 200 . In the case of notification, a notification may be made informing that the product is withdrawn from the pickup box 200 and changed to a standby state.

Meanwhile, for reference, the controller 110 may perform image analysis by artificial intelligence as shown in FIG. 8 . 8 is a schematic diagram illustrating the creation of a dynamic self-learning reference data set in the intelligent non-face-to-face pickup box system 10 according to the present invention and the integration with a reference data set initially carried out at the factory to form an integrated reference data set. has been

Referring to FIG. 8 , the image/video recognition reference data set includes a reference data set as an initial reference data set 510 , a dynamic self-learning reference data set 520 , and an initial reference data set 510 and dynamic self-learning reference data. Collected according to identification classified into a unified reference data set 530 created by aggregating set 520 . A standard database is established by integrating all reference data sets and the reference data set 540 of the standard database is used in the process of image/video recognition. The reference data set 540 of the standard database grows through the reproduction and feedback process of the dynamic reference data set 520 .

The initial reference data set 510 and the dynamic self-learning reference data set 520 can be identified by segmentation labeling, and the reference data set is collected with identification of an active or inactive (inappropriate) state.

The initial reference data set 510 may be generated by artificial data massaging of artificially collected data. The generation of the initial reference data set 510 is performed by artificially collecting image/video data of a product and performing an artificial learning process, ie, converting the collected data into a reference data set for recognition. When collecting image/video data for the generation of the initial reference data set 510, the camera 230 selects an appropriate data set model to be applied to the internal environment of the non-face-to-face pickup box 200 in various placement directions of the product. can be applied to The initial reference data set 510 may be generated according to the resolution and magnification of the camera 230 .

The dynamic self-learning reference data set 520 may be generated by automatically collecting or generating data with software installed in the non-face-to-face pickup box 200 .

During the process of generating the dynamic self-learning reference data set 520, the state information adapted to the resolution and magnification of the camera 230 and the brightness of the internal lighting of the non-face-to-face pickup box 200 automatically converts product images and video data. It is created by software that generates and learns by itself using the data. By applying the result to the state information database inside the non-face-to-face pickup box 200, the self-learning files are continuously updated without additional generation and learning of artificial data.

The function of image/video recognition for the product is to analyze the state descriptor from the image and video data input from the camera 230 to recognize the input state, to self-lear the state, and to learn the state of the product. can be accurately recognized.

The product image/video recognition software classifies all state descriptors of image/video data into N representative state descriptor types. The product image/video recognition software analyzes each state descriptor of the product image/video data according to N representative state descriptors and classifies it into the most similar representative state descriptor type. A representative state descriptor may be selected by statistically analyzing the similarity of various types of state descriptors. For example, by applying a K-mean clustering algorithm, various types of state descriptors may be classified into N representative types. The state of the product can be recognized by classifying the input image/video data into one of N representative types.

The product image/video recognition software of the non-face-to-face pickup box 200 stores a dynamic self-learning state type, that is, a representative state descriptor and a corresponding product image/video state probability in the state information database. By integrating the results generated by the dynamic self-learning into the reference data set state information database, the non-face-to-face pickup box 200 can provide a service with a high degree of performance without additional artificial data generation and learning. In addition, the intelligence of product status recognition of the non-face-to-face pickup box 200 can continuously grow, so that a high degree of performance can be achieved by continuously updating the self-learning data set.

10: intelligent non-face-to-face pickup box system 100: pickup box control device
110: control unit 120: memory unit
130: communication unit 200: pickup box
210: box body 220: product take-out door
221: handle 230: camera
240: scale 250: touch screen
251: password input unit 252: advertisement display unit
253: password input unit appearance switching button 300: order reception server
400: customer terminal

Claims (9)

A box body consisting of an upper surface, a lower surface, a front surface, a rear surface, a left side and a right side, the inside of which is empty, is installed on the rear side of the box body, and is a product entry door that can be opened by a business operator to introduce a product provided to a customer , a product take-out door installed in the front of the box body, which can be opened when a customer enters a password, installed inside the box body, and capturing the appearance of a product accommodated in the box body to obtain an image or video Pickup comprising at least one non-face-to-face pickup box including at least one camera, and a control unit and a memory unit for storing a program executed by the control unit and storing data necessary for the execution of the program and generated as a result of the execution of the program a box control device;
When the control unit of the pickup box control device receives an order from the order receiving server, storing the order number and order information in the memory unit, the customer based on the location of the customer terminal receiving the order to the order receiving server Predicting the time when a customer who owns a terminal will arrive at the store in which the pickup box is located, analyzing the image captured by the camera installed in the pickup box to determine the order number of the product brought into the pickup box, and In a situation where it is determined that a plurality of customers are scheduled to arrive at the store with respect to the pickup box or any of the pickup boxes, a customer other than the customer who ordered the product delivered in the pickup box arrives at the store first In this case or when it is predicted to arrive earlier, instruct to withdraw the product received in the pickup box and put the product corresponding to the customer's order number that arrives at the store first or is expected to arrive earlier in the pickup box. Intelligent non-face-to-face pickup box system, characterized in that executing the step of generating. According to claim 1,
The control unit of the pickup box control device further executes the step of determining the type and quantity of products introduced into the pickup box by analyzing the image or video taken by the camera installed in the pickup box, and determining by the step Intelligent non-face-to-face pickup box system, characterized in that when the type and quantity of the product is different from the type and quantity of the product according to the order number stored in the memory unit, an error alarm is issued. 3. The method of any one of claims 1 and 2,
A scale is further installed in the pickup box, and the control unit of the pickup box control device analyzes the image captured by the camera installed in the pickup box to determine the order number of the product introduced into the pickup box. When the weight of the product corresponding to the order number determined by Intelligent non-face-to-face pickup box system, characterized in that when it does not match the weight of the product, an error alarm is issued. 4. The method according to any one of claims 1 to 3,
If a customer other than the customer who ordered the product delivered in the pickup box arrives at the store first or is expected to arrive first, the estimated arrival time of the customer who arrives at the store first or is expected to arrive earlier is 5 minutes within, and the estimated arrival time of the customer who ordered the product delivered in the pickup box is determined to be 3 minutes or more later than the expected arrival time of the customer who arrives at the store first or is expected to arrive earlier Non-face-to-face pickup box system. 5. The method of claim 4,
The control unit of the pickup box control device withdraws the product introduced into the pickup box because the other customer, other than the customer who ordered the product brought into the pickup box, arrives at the store first or is expected to arrive earlier, When an instruction is generated to bring a product corresponding to the order number of the customer who arrives first or is expected to arrive earlier in the pickup box, the customer of the customer who arrives at the store first or is expected to arrive earlier An intelligent non-face-to-face pickup box system, characterized in that it notifies the terminal that the corresponding product has entered the pickup box. 6. The method of claim 5,
If the customer terminal of the customer who ordered the product that was drawn into and withdrawn from the pickup box has previously been notified that the product has been drawn into the pickup box, the product is withdrawn from the pickup box to the customer terminal and is in a standby state. Intelligent non-face-to-face pickup box system, characterized in that it notifies that there is. A box body consisting of an upper surface, a lower surface, a front surface, a rear surface, a left side and a right side, the inside of which is empty, is installed on the rear side of the box body, and is a product entry door that can be opened by a business operator to introduce a product provided to a customer , a product take-out door installed in the front of the box body, which can be opened when a customer enters a password, installed inside the box body, and capturing the appearance of a product accommodated in the box body to obtain an image or video Pickup comprising at least one non-face-to-face pickup box including at least one camera, and a control unit and a memory unit for storing a program executed by the control unit and storing data necessary for the execution of the program and generated as a result of the execution of the program a box control device;
When the control unit of the pickup box control device receives an order from the order receiving server, storing the order number and order information in the memory unit, the customer based on the location of the customer terminal receiving the order to the order receiving server An intelligent non-face-to-face pickup box system, comprising: predicting a time when a customer who owns a terminal will arrive at the store in which the pickup box is located; and displaying the expected arrival time of the customer at the store. 8. The method of claim 7,
In a situation in which it is determined that a plurality of customers are scheduled to arrive at the store with respect to the pickup box or a pickup box among the pickup boxes, the controller has the shortest expected arrival time at the store among the products of the customers whose packaging is completed. An intelligent non-face-to-face pickup box system, characterized in that executing the step of displaying an instruction to bring the customer's product into the pickup box. 9. The method according to any one of claims 1 to 8,
The control unit stores the image including the external characteristics of the packaging container and the weight of the packaging container in the memory unit, storing the image and weight information of the packaged order when an order is generated in the memory unit, for a single menu When an order is placed, when the order for the single menu is packaged in a container with a known container weight and loaded into the pickup box, the weight of the single menu is calculated by subtracting the weight of the container from the total weight of the order, Storing the calculated weight in the memory unit, and when an order for a complex menu including a menu whose weight value has not yet been identified and a menu whose weight value has been identified occurs, the complex menu has a tare weight When it is packaged in a container and brought into the pickup box, the weight of the single menu whose weight has not yet been calculated is calculated by subtracting the weight of the container and the weight of the menu for which the weight value has been determined from the total weight for the order, and so calculated. Intelligent non-face-to-face pickup box system, characterized in that executing the step of storing one weight in the memory unit.

Images