WO2020047919A1 - 无人售货方法、装置、系统、服务器及计算机可读存储介质 - Google Patents

无人售货方法、装置、系统、服务器及计算机可读存储介质 Download PDF

Info

Publication number
WO2020047919A1
WO2020047919A1 PCT/CN2018/108182 CN2018108182W WO2020047919A1 WO 2020047919 A1 WO2020047919 A1 WO 2020047919A1 CN 2018108182 W CN2018108182 W CN 2018108182W WO 2020047919 A1 WO2020047919 A1 WO 2020047919A1
Authority
WO
WIPO (PCT)
Prior art keywords
vending machine
information
product
unmanned vending
target unmanned
Prior art date
Application number
PCT/CN2018/108182
Other languages
English (en)
French (fr)
Inventor
黄鼎隆
斯科特·马修·罗伯特
傅恺
张弛
Original Assignee
深圳码隆科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳码隆科技有限公司 filed Critical 深圳码隆科技有限公司
Publication of WO2020047919A1 publication Critical patent/WO2020047919A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07F—COIN-FREED OR LIKE APPARATUS
    • G07F11/00—Coin-freed apparatus for dispensing, or the like, discrete articles
    • G07F11/02—Coin-freed apparatus for dispensing, or the like, discrete articles from non-movable magazines

Definitions

  • the present application relates to the technical field of self-service sales, and in particular, to an unmanned sales method, device, system, server, and computer-readable storage medium.
  • the existing self-service shopping method is usually implemented by a self-service vending machine. As long as the user selects his favorite items displayed in the self-service vending machine and inputs corresponding coins from the coin port, he can buy the selected items.
  • this kind of self-service vending machine provides a limited variety of items, usually only beverages such as Cola and Sprite, and requires users to perform relatively complicated operations, making users shopping inefficient.
  • the purpose of this application includes providing an unmanned vending method, device, system, server, and computer-readable storage medium, which can achieve fast and efficient sales through the cooperation of a commodity recognition server and an unmanned vending machine. Goods service, greatly improving the shopping experience of users.
  • an embodiment of the present application provides an unmanned sales method, which is applied to a commodity identification server.
  • the method includes:
  • the start signal is sent by the controller of the unmanned vending machine whose door is opened; the start signal carries the identification of the target unmanned vending machine;
  • the product images include: from the opening of the cabinet door of the target unmanned vending machine to the closing time, a plurality of product images in the box collected by the camera device of the target unmanned vending machine;
  • a pre-shopping list corresponding to the target unmanned vending machine is generated; the pre-shopping list includes: product name, product quantity, product unit price and total price.
  • the embodiment of the present application provides a first possible implementation manner of the first aspect, where the foregoing method further includes:
  • the commodity image sample data is trained based on the neural network model to generate a commodity recognition model.
  • the embodiment of the present application provides a second possible implementation manner of the first aspect, in which the commodity image includes: a period of time from the opening of the cabinet door of the target unmanned vending machine to the time of closing, every preset time , Multiple commodity images in the box collected by the camera device of the target unmanned vending machine;
  • the product image includes: during the period from when the door of the target unmanned vending machine is opened to closed, when the door is opened, when the gravity sensor of the target unmanned vending machine changes, and when the door of the cabinet is closed, the target unmanned vending machine Multiple commodity images in the box collected by the camera of the cargo aircraft.
  • the embodiment of the present application provides a third possible implementation manner of the first aspect, where a plurality of product information is identified according to a plurality of product images and a product recognition model, and specifically includes:
  • the embodiment of the present application provides a fourth possible implementation manner of the first aspect, where the pre-shopping list corresponding to the target unmanned vending machine is generated based on the identification identifier and multiple product information, and specifically includes:
  • the multiple pre-purchased product information is summarized, and based on the identification, a pre-shopping list corresponding to the target unmanned vending machine is generated.
  • the embodiment of the present application provides a fifth possible implementation manner of the first aspect, where after the pre-shopping list corresponding to the target unmanned vending machine is generated based on the identification identifier and multiple product information, the foregoing The method also includes:
  • the embodiment of the present application provides a sixth possible implementation manner of the first aspect, wherein the target unmanned vending machine is generated based on the identification identifier and the plurality of product information. After the corresponding pre-shopping list, it also includes:
  • the embodiment of the present application provides a seventh possible implementation manner of the first aspect, where after the pre-shopping list corresponding to the target unmanned vending machine is generated based on the identification identifier and multiple product information, the above The method also includes:
  • the embodiment of the present application provides an eighth possible implementation manner of the first aspect, further including:
  • an embodiment of the present application further provides an unmanned sales device, including:
  • the start signal receiving module is used to receive the start signal of the target unmanned vending machine; the start signal is sent by the controller of the unmanned vending machine whose door is opened; the start signal carries the identification of the target unmanned vending machine;
  • the image receiving module is used to receive the product image of the target unmanned vending machine; the product image includes: from the opening of the cabinet door of the target unmanned vending machine to the closing time, Multiple product images;
  • An information identification module configured to identify multiple product information based on multiple product images and product identification models
  • the list generation module generates a pre-shopping list corresponding to the target unmanned vending machine based on the identification identifier and multiple product information; the pre-shopping list includes: product name, product quantity, product unit price and total price.
  • the embodiment of the present application provides a first possible implementation manner of the second aspect, where the device further includes:
  • a sample data acquisition module configured to acquire sample image data of a product
  • the model generation model is configured to train the commodity image sample data based on a neural network model to generate the commodity recognition model.
  • the embodiment of the present application provides a second possible implementation manner of the second aspect, wherein the information recognition module is further configured to: respectively input the plurality of product images into the product recognition model; Based on multiple output results of the product recognition model, the multiple product information is obtained.
  • the embodiment of the present application provides a third possible implementation manner of the second aspect, wherein the list generating module is further configured to: compare product information of any two adjacent shooting times; judge Whether the commodity information adjacent to any two shooting times are the same; if yes, it is determined that the user is not currently shopping; if not, according to different commodity information in the commodity information adjacent to any two shooting times, Generate multiple pieces of pre-purchased product information; aggregate multiple pieces of the pre-purchased product information, and generate a pre-shopping list corresponding to the target unmanned vending machine based on the identification identifier.
  • the embodiment of the present application provides a fourth possible implementation manner of the second aspect, where the device further includes:
  • a first list sending module configured to send the pre-shopping list to a controller of the target unmanned vending machine, so that the controller generates payment information according to the pre-shopping list;
  • the first information receiving module is configured to receive payment success information generated by the controller after the user completes a payment operation according to the payment information.
  • the embodiment of the present application provides a fifth possible implementation manner of the second aspect, where the device further includes:
  • a user information receiving module configured to receive user information sent by the controller
  • a second list sending module configured to send the pre-shopping list to the first mobile terminal of the user according to the user information, so that the user makes payment according to the pre-shopping list
  • the second information receiving module is configured to receive payment success information sent by the first mobile terminal after the user completes payment.
  • the embodiment of the present application provides a sixth possible implementation manner of the second aspect, where the device further includes:
  • an embodiment of the present application further provides an unmanned vending system, including: a plurality of unmanned vending machines, a commodity identification server, and a mobile terminal;
  • An unmanned sales device as described in the second aspect is installed on the commodity identification server;
  • Commodity identification server multiple unmanned vending machines and mobile terminals are connected in pairs;
  • the mobile terminal includes: a first mobile terminal and a second mobile terminal;
  • the unmanned vending machine includes a vending machine body and a controller, a camera device and a gravity sensor installed in the vending machine body; the vending machine body includes a cabinet door and a box body.
  • an embodiment of the present application further provides a server, including a memory and a processor, where the memory is configured to store a program that supports the processor to execute the method according to any one of the first aspect, and the processing The processor is configured to execute a program stored in the memory.
  • an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a program code executable by a processor, and the program code is configured to enable the processing
  • the processor performs the method according to any one of the first aspects.
  • the unmanned vending method provided in the embodiment of the present application is applied to a commodity recognition server.
  • the method includes: receiving a start signal of a target unmanned vending machine; the start signal is sent by a controller of the unmanned vending machine whose door is opened; The signal carries the identification of the target unmanned vending machine; the product image of the target unmanned vending machine is received; the product image includes: the door of the target unmanned vending machine is opened from the closed time, and the video of the target unmanned vending machine is taken by the camera Multiple product images in the box collected by the device; multiple product information is identified based on multiple product images and product identification models; based on the identification and multiple product information, a pre-shopping list corresponding to the target unmanned vending machine is generated ; Pre-shopping list includes: product name, product quantity, product unit price and total price.
  • FIG. 1 is a flowchart of an unmanned sales method according to an embodiment of the present application
  • FIG. 3 is a flowchart of another unmanned sales method according to an embodiment of the present application.
  • FIG. 6 is a schematic diagram of an unmanned sales device according to an embodiment of the present application.
  • FIG. 7 is a schematic diagram of an unmanned sales system according to an embodiment of the present application.
  • the existing self-service vending machines provide limited items, usually only beverages such as Coke and Sprite, and require relatively complicated operations by the user, making the user's shopping inefficient.
  • the embodiments of the present application provide an unmanned sales method, device, system, server, and computer-readable storage medium, which can realize a fast and efficient sales service through the cooperation of a commodity identification server and an unmanned vending machine. Greatly improve the user's physical shopping experience.
  • An embodiment of the present application provides an unmanned vending method.
  • the method is applied to a commodity recognition server.
  • the commodity recognition server may be communicatively connected to at least one unmanned vending machine.
  • a plurality of unmanned vending machines may be arranged in different locations in advance. Such as shopping malls, playgrounds, streets, etc. are not limited here.
  • the method includes the following steps:
  • S11 Receive the start signal of the target unmanned vending machine; the start signal is sent from the unmanned vending machine whose cabinet door is opened to the product identification server through the controller; the start signal carries the identification mark of the target unmanned vending machine.
  • the target unmanned vending machine that is, the unmanned vending machine whose cabinet door is opened.
  • the door condition monitoring device such as an infrared sensor in the unmanned vending machine can sense that the door is opened.
  • the controller in the unmanned vending machine Will receive the door opening signal sensed by the cabinet door condition monitoring device, at the same time, the controller sends an activation signal to the product identification server, the activation signal carries the identification mark of the target unmanned vending machine, and the identification mark may be The number, code, or other identity information of the vending machine can be used to distinguish it from other unmanned vending machines.
  • S12 receiving the product image of the target unmanned vending machine; the product image may include: during the period from when the cabinet door of the target unmanned vending machine is opened to closed, a large number of boxes in the box collected by the camera device of the target unmanned vending machine Product images.
  • the opening or closing of the cabinet door of the target unmanned vending machine is determined by the cabinet door condition monitoring device in the unmanned vending machine, and the commodity images collected by the camera device are in the period from the cabinet door opening to closing.
  • the multiple commodity images may be a group of commodity images with relatively continuous shooting times.
  • the product image may include the following two cases:
  • a plurality of commodity images in the box are collected by the camera device of the target unmanned vending machine every preset time.
  • the estimated duration can be adjusted according to the actual situation; for example, the preset duration can be set to 2s.
  • the moment when the gravity sensor changes may include multiple moments. For example, when a user purchases multiple products, the gravity sensor may sense the change in gravity multiple times. Correspondingly, there are multiple commodity images collected by the camera device.
  • the commodity recognition server may store multiple commodity images uploaded by the target unmanned vending machine according to the time series for subsequent recognition and analysis.
  • a plurality of camera devices may be provided and installed in a plurality of positions in the cabinet, respectively, so as to improve the accuracy of subsequent image recognition.
  • multiple camera devices can be installed on the top of the cabinet, under the partition of each shelf in the cabinet (can be used to collect images of products on the lower shelf), and between adjacent shelves on the upper and lower floors.
  • the upper side wall can be specifically set according to the internal structure of the unmanned vending machine and the layout of the product, which is not limited here. It can be understood that the camera device in the unmanned vending machine provided in this embodiment can collect images of all the commodities located in the cabinet,
  • S13 Identify a plurality of product information based on a plurality of product images and a product recognition model.
  • the commodity recognition server stores a commodity recognition model in advance, and the model is obtained by training the commodity data through a neural network.
  • the specific process is as follows:
  • the commodity image sample data may include a preset number of commodity images and commodity information, where the commodity image includes: images of various commodities at different angles, and the commodity information includes: the commodity corresponding to the image Various product information such as name and product weight.
  • the commodity image sample data is trained based on the neural network model to generate a commodity recognition model.
  • the product data sample is divided into training data samples and verification data samples according to a preset ratio.
  • the preset ratio can be expressed in the form of a percentage, for example, 70% of the data is used to train the model, and 30% of the data is used to verify the accuracy of the model.
  • the training data samples are then input into a neural network model for training, thereby generating a product recognition model.
  • the verification data samples can also be input into the product recognition model to verify the accuracy of the product recognition model. When the verification result is not ideal, adjust the model parameters and continuously optimize the model to improve the accuracy of the product identification model.
  • the trained commodity recognition model is stored in a commodity recognition server in advance, and can accurately recognize the subsequently acquired commodity images.
  • a plurality of product images are respectively input into a product recognition model; based on a plurality of output results of the product recognition model, a plurality of product information is obtained.
  • S14 Generate a pre-shopping list corresponding to the target unmanned vending machine based on the identification mark and multiple product information.
  • the pre-shopping list may include: product name, product quantity, product unit price, total price, and the like. Based on the identification and multiple product information, the pre-shopping list corresponding to the target unmanned vending machine is generated, which specifically includes the following steps, as shown in Figure 2:
  • step S23 it is determined that the user is not currently making a purchase.
  • step S24 is performed: generating multiple pre-purchased product information based on different product information in any two neighboring product information at the time of shooting; summarizing the multiple pre-purchased product information and generating a target without Pre-shopping list corresponding to the person vending machine.
  • any two adjacent product information at the time of shooting to determine whether the two are the same. If they are the same, it means that the user has not taken away the item, that is, he has not done any shopping. If they are different, it indicates that the user has taken away.
  • Products, and the product information corresponding to the product image at the next time is usually less than the product information corresponding to the product image at the previous time.
  • multiple pre-ordered product information can be generated based on different product information in the two.
  • multiple pre-purchased product information is summarized and superimposed to obtain a pre-shopping list. At the same time, the pre-shopping list is matched with the target unmanned vending machine through the identification of the target unmanned vending machine, that is, the target unmanned vending machine is generated. Pre-shopping list for cargo machine.
  • S31 Send the pre-shopping list to the controller of the target unmanned vending machine, so that the controller generates payment information according to the pre-shopping list.
  • the product identification server After the pre-shopping list is generated, the product identification server sends the list to the controller in the target unmanned vending machine, further causing the controller to generate payment information according to the pre-shopping list, so that the user makes payment according to the payment information.
  • the product identification server can also generate payment information according to the pre-shopping list, and then directly send the payment information to the controller of the target unmanned vending machine.
  • S32 Receive payment success information generated by the user after the user completes the payment operation according to the payment information.
  • the controller After the user completes the payment according to the payment information, the controller sends the user payment success information to the product identification server, so that the server performs subsequent recording and analysis.
  • the method further includes the following steps, as shown in FIG. 4:
  • the controller Before the user opens the cabinet door, first perform a code scanning operation through a mobile terminal such as a mobile phone to open the cabinet door of the unmanned vending machine through this operation. During this process, the controller will automatically receive the user's identity information, such as The mobile phone number of the user, that is, the above user information. After the user completes the retrieval operation, the user information of the user is sent to the product identification server.
  • S42 Send the pre-shopping list to the first mobile terminal of the user according to the user information, so that the user makes payment according to the pre-shopping list.
  • the product identification server sends the pre-shopping list to the user's mobile terminal according to the above-mentioned user information.
  • the pre-shopping list can be sent to the user's mobile terminal via SMS or MMS, or the pre-shopping list can be sent to the user's mobile terminal.
  • the preset APP interface allows users to complete payment operations.
  • S43 Receive payment success information sent by the first mobile terminal after the user completes the payment.
  • the user After the user completes the payment according to the above pre-shopping list, the user receives the payment success information for subsequent recording and analysis.
  • the first mobile terminal here is used to distinguish it from the second mobile terminal described below.
  • the first mobile terminal refers to the mobile terminal of the user, and the second mobile terminal refers to the mobile terminal of the manager.
  • Mobile terminals include: smartphones, tablets, etc.
  • the product identification server After the product identification server receives the user's payment success information, it further includes the following steps, as shown in FIG. 5:
  • S51 Record the sales information of the target unmanned vending machine based on the payment success information.
  • the commodity identification server After receiving the user's payment success information, the commodity identification server records the commodity information in the contract to make statistics on the sales of the target unmanned vending machine.
  • S53 Send the replenishment information to the second mobile terminal of the manager.
  • the unmanned vending method provided in the embodiment of the present application is applied to a commodity recognition server.
  • the method includes: receiving a start signal of a target unmanned vending machine; the start signal is sent by a controller of the unmanned vending machine whose door is opened; The signal carries the identification of the target unmanned vending machine; the product image of the target unmanned vending machine is received; the product image includes: the door of the target unmanned vending machine is opened from the closed time, and the video of the target unmanned vending machine is taken by the camera Multiple product images in the box collected by the device; multiple product information is identified based on multiple product images and product identification models; based on the identification and multiple product information, a pre-shopping list corresponding to the target unmanned vending machine is generated ; Pre-shopping list includes: product name, product quantity, product unit price and total price.
  • the embodiment of the present application can realize a fast and efficient sales service through the cooperation of a product identification server and an unmanned vending machine, which greatly improves the physical shopping experience of
  • an embodiment of the present application further provides an unmanned sales device, as shown in FIG. 6.
  • the device includes: a start signal receiving module 61, an image receiving module 62, an information identification module 63, and a list. Generating module 64.
  • the activation signal receiving module 61 is configured to receive the activation signal of the target unmanned vending machine; the activation signal is sent by the controller from the unmanned vending machine whose door is opened; the activation signal carries the identification of the target unmanned vending machine Identification; an image receiving module 62 configured to receive a product image of the target unmanned vending machine; the product image includes: the cabinet door of the target unmanned vending machine is collected by the camera device of the target unmanned vending machine from opening to closing time; Multiple commodity images in the box; an information recognition module 63 configured to identify multiple commodity information based on the multiple commodity images and a commodity recognition model; a list generation module 64, based on the identification identifier and the multiple commodity information, to generate a target unmanned
  • the above device further includes:
  • a sample data acquisition module configured to acquire sample image data of a product
  • the model generation model is configured to train commodity image sample data based on a neural network model to generate a commodity recognition model.
  • the above-mentioned information recognition module is further configured to: input multiple commodity images into the commodity recognition model respectively; and obtain multiple commodity information based on multiple output results of the commodity recognition model.
  • the above-mentioned list generation module is further configured to: compare product information adjacent to any two shooting times; determine whether the product information adjacent to any two shooting times is the same; if yes, determine that the user is not currently performing Shopping; if not, generate multiple pre-purchased product information based on different product information in any two adjacent product information at the time of shooting; aggregate multiple pre-purchased product information and generate a target unsold based on the identification Pre-shopping list for cargo machine.
  • the above device further includes:
  • the first list sending module is configured to send the pre-shopping list to the controller of the target unmanned vending machine, so that the controller generates payment information according to the pre-shopping list;
  • the first information receiving module is configured to receive payment success information generated by the user after the user completes a payment operation according to the payment information.
  • the above device further includes:
  • a user information receiving module configured to receive user information sent by a controller
  • the second list sending module is configured to send the pre-shopping list to the first mobile terminal of the user according to the user information, so that the user makes payment according to the pre-shopping list;
  • the second information receiving module is configured to receive payment success information sent by the first mobile terminal after the user completes payment.
  • the above device further includes:
  • each module has the same technical characteristics as the aforementioned unmanned vending method, and therefore, the above functions can also be implemented.
  • An embodiment of the present application further provides an unmanned vending system. As shown in FIG.
  • the system includes: a plurality of unmanned vending machines 71 (only one is shown in the figure), a commodity identification server 72 and a mobile terminal 73;
  • An unmanned vending device 721 as described in the second embodiment is installed on the identification server 72;
  • the product identification server 72, a plurality of unmanned vending machines 71, and a mobile terminal 73 are connected in pairs;
  • the mobile terminal 73 includes: a first mobile terminal and The second mobile terminal;
  • the unmanned vending machine 71 includes: a vending machine body 711 and a controller 712, a camera 713, and a gravity sensor 714 installed in the vending machine body;
  • the vending machine body 711 includes a cabinet door and a cabinet.
  • each module has the same technical characteristics as the aforementioned unmanned vending method, and therefore, the above functions can also be implemented.
  • the above functions can also be implemented.
  • An embodiment of the present application further provides a server including a memory and a processor, where the memory is configured to store a program that supports the processor to execute any one of the foregoing unmanned sales methods, and the processor is configured to execute the method Program stored in memory.
  • the product identification server that executes the unmanned sales method may be implemented by using the foregoing server provided in the embodiments of the present application.
  • An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a program code executable by a processor, and the program code is configured to cause the processor to perform any of the foregoing tasks.
  • An unmanned sales method is configured to cause the processor to perform any of the foregoing tasks.
  • the computer program product of the unmanned sales method provided in the embodiments of the present application includes a computer-readable storage medium storing a non-volatile program code executable by a processor, and the program code includes instructions that can be used to execute the foregoing method.
  • the program code includes instructions that can be used to execute the foregoing method.
  • each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, which contains one or more components for implementing a specified logical function Executable instructions.
  • the functions noted in the blocks may also occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved.
  • each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts can be implemented in a dedicated hardware-based system that performs the specified function or action. , Or it can be implemented with a combination of dedicated hardware and computer instructions.
  • the disclosed systems, devices, and methods may be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the unit is only a logical function division.
  • multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some communication interfaces, devices or units, which may be electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objective of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each of the units may exist separately physically, or two or more units may be integrated into one unit.
  • the functions are implemented in the form of software functional units and sold or used as independent products, they may be stored in a non-volatile computer-readable storage medium executable by a processor.
  • the technical solution of this application is essentially a part that contributes to the existing technology or a part of the technical solution can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present application.
  • the foregoing storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks and other media that can store program codes .

Abstract

一种无人售货方法、装置及系统，涉及自助售货技术领域，无人售货方法应用于商品识别服务器，该方法包括：接收目标无人售货机的启动信号；启动信号由柜门被打开的无人售货机通过控制器进行发送；启动信号中携带有目标无人售货机的识别标识（S11）；接收目标无人售货机的商品图像；商品图像包括：目标无人售货机的柜门从打开到关闭时间内，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像（S12）；根据多个商品图像及商品识别模型，识别出多个商品信息（S13）；基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单（S14），能够通过商品识别服务器及无人售货机的配合，实现快速、高效的售货服务，大大提高用户的购物体验。

Description

无人售货方法、装置、系统、服务器及计算机可读存储介质
相关申请的交叉引用
本申请要求于2018年9月6日提交中国专利局的申请号为2018110403628，名称为“无人售货方法、装置及系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及自助售货技术领域，尤其是涉及一种无人售货方法、装置、系统、服务器及计算机可读存储介质。
背景技术
随着国家实力的增强以及国民经济的快速发展，人们的购物方式变得越来越多样化，其中，自助购物成为一种越来越流行的购物方式，极大地方便了人们的生活。
现有的自助购物方式通常是采用自助售货机来实现的，用户只要选中自助售货机内展示的自己心仪的物品，并从钱币口投入相应的钱币，就可以买到所选的物品。但是这种自助售货机提供的物品种类有限，通常只是可乐、雪碧等饮料类物品，且需要用户进行相对复杂的操作，使得用户购物效率低下。
申请内容
有鉴于此，本申请的目的包括，提供一种无人售货方法、装置、系统、服务器及计算机可读存储介质，能够通过商品识别服务器及无人售货机的配合，实现快速、高效的售货服务，大大提高用户的购物体验。
为了实现上述目的至少之一，本申请采用的技术方案如下：
第一方面，本申请实施例提供了一种无人售货方法，应用于商品识别服务器，该方法包括：
接收目标无人售货机的启动信号；启动信号由柜门被打开的无人售货机通过控制器进行发送；启动信号中携带有目标无人售货机的识别标识；
接收目标无人售货机的商品图像；商品图像包括：目标无人售货机的柜门从打开到关闭时间内，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像；
根据多个商品图像及商品识别模型，识别出多个商品信息；
基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单；预购物清单包括：商品名称、商品数量、商品单价及总价。
结合第一方面，本申请实施例提供了第一方面的第一种可能的实施方式，其中，上述方法还包括：
获取商品图像样本数据；
基于神经网络模型对商品图像样本数据进行训练，生成商品识别模型。
结合第一方面，本申请实施例提供了第一方面的第二种可能的实施方式，其中，商品图像包括：目标无人售货机的柜门从打开到关闭的期间内，每隔预设时间，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像；
或者，
商品图像包括：目标无人售货机的柜门从打开到关闭的期间内，柜门的打开时刻、目标无人售货机的重力传感器发生变化的时刻、柜门的关闭时刻，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像。
结合第一方面，本申请实施例提供了第一方面的第三种可能的实施方式，其中，根据多个商品图像及商品识别模型，识别出多个商品信息，具体包括：
将多个商品图像分别输入商品识别模型中；
基于商品识别模型的多个输出结果，得到多个商品信息。
结合第一方面，本申请实施例提供了第一方面的第四种可能的实施方式，其中，基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单，具体包括：
将任意两个拍摄时间相邻的商品信息进行比对；
判断任意两个拍摄时间相邻的商品信息是否相同；
如果是，则确定用户当前没有进行购物；
如果否，则根据任意两个拍摄时间相邻的商品信息中的不同的商品信息，生成多个预购商品信息；
将多个预购商品信息进行汇总，并基于识别标识，生成目标无人售货机所对应的预购物清单。
结合第一方面，本申请实施例提供了第一方面的第五种可能的实施方式，其中，在基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单之后，上述方法还包括：
将预购物清单向目标无人售货机的控制器发送，以使控制器根据预购物清单生成支付信息；
接收控制器所发送的用户根据支付信息完成支付操作后所生成的支付成功信息。
结合第一方面，本申请实施例提供了第一方面的第六种可能的实施方式，其中，在所述基于所述识别标识及所述多个商品信息，生成所述目标无人售货机所对应的预购物清单之后，还包括：
根据所述预购物清单生成支付信息，将所述支付信息发送给所述目标无人售货机的控制器；
接收所述控制器所发送的用户根据所述支付信息完成支付操作后所生成的支付成功信息。
结合第一方面，本申请实施例提供了第一方面的第七种可能的实施方式，其中，在基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单之后，上述方法还包括：
接收控制器所发送的用户信息；
根据用户信息，将预购物清单发送至用户的第一移动终端，以使用户根据预购物清单进行付款；
接收用户完成付款后第一移动终端所发送的支付成功信息。
结合第一方面，本申请实施例提供了第一方面的第八种可能的实施方式，其中，还包括：
根据支付成功信息，对目标无人售货机的销售情况信息进行记录；
根据销售情况信息，确定当前存货情况及补货信息；
将补货信息向管理者的第二移动终端发送。
第二方面，本申请实施例还提供一种无人售货装置，包括：
启动信号接收模块，用于接收目标无人售货机的启动信号；启动信号由柜门被打开的无人售货机通过控制器进行发送；启动信号中携带有目标无人售货机的识别标识；
图像接收模块，用于接收目标无人售货机的商品图像；商品图像包括：目标无人售货机的柜门从打开到关闭时间内，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像；
信息识别模块，用于根据多个商品图像及商品识别模型，识别出多个商品信息；
清单生成模块，基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单；预购物清单包括：商品名称、商品数量、商品单价及总价。
结合第二方面，本申请实施例提供了第二方面的第一种可能的实施方式，其中，所述装置还包括：
样本数据获取模块，配置成获取商品图像样本数据；
模型生成模型，配置成基于神经网络模型对所述商品图像样本数据进行训练，生成所述商品识别模型。
结合第二方面，本申请实施例提供了第二方面的第二种可能的实施方式，其中，所述信息识别模块进一步配置成：将所述多个商品图像分别输入所述商品识别模型中；基于所述商品识别模型的多个输出结果，得到所述多个商品信息。
结合第二方面，本申请实施例提供了第二方面的第三种可能的实施方式，其中，所述清单生成模块进一步配置成：将任意两个拍摄时间相邻的商品信息进行比对；判断所述任意两个拍摄时间相邻的商品信息是否相同；如果是，则确定用户当前没有进行购物；如果否，则根据所述任意两个拍摄时间相邻的商品信息中的不同的商品信息，生成多个预购商品信息；将多个所述预购商品信息进行汇总，并基于所述识别标识，生成所述目标无人售货机所对应的预购物清单。
结合第二方面，本申请实施例提供了第二方面的第四种可能的实施方式，其中，所述装置还包括：
第一清单发送模块，配置成将所述预购物清单向所述目标无人售货机的控制器发送，以使所述控制器根据所述预购物清单生成支付信息；
第一信息接收模块，配置成接收所述控制器所发送的用户根据所述支付信息完成支付操作后所生成的支付成功信息。
结合第二方面，本申请实施例提供了第二方面的第五种可能的实施方式，其中，所述装置还包括：
用户信息接收模块，配置成接收所述控制器所发送的用户信息；
第二清单发送模块，配置成根据所述用户信息，将所述预购物清单发送至所述用户的第一移动终端，以使所述用户根据所述预购物清单进行付款；
第二信息接收模块，配置成接收用户完成付款后所述第一移动终端所发送的支付成功信息。
结合第二方面，本申请实施例提供了第二方面的第六种可能的实施方式，其中，所述装置还包括：
根据所述支付成功信息，对所述目标无人售货机的销售情况信息进行记录；
根据所述销售情况信息，确定当前存货情况及补货信息；
将所述补货信息向管理者的第二移动终端发送。
第三方面，本申请实施例还提供一种无人售货系统，包括：多个无人售货机、商品识别服务器及移动终端；
商品识别服务器上安装有如第二方面所述的无人售货装置；
商品识别服务器、多个无人售货机及移动终端两两通信连接；
移动终端包括：第一移动终端和第二移动终端；
无人售货机包括：售货机本体及安装于售货机本体中的控制器、摄像装置和重力传感器；售货机本体包括柜门、箱体。
第四方面，本申请实施例还提供一种服务器，其特征在于，包括存储器以及处理器，所述存储器用于存储支持处理器执行如第一方面任一项所述方法的程序，所述处理器被配置为用于执行所述存储器中存储的程序。
第三方面，本申请实施例还提供一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有处理器可执行的程序代码，所述程序代码被配置成使所述处理器执行如第一方面任一项所述方法。
本申请实施例带来了以下有益效果：
本申请实施例提供的无人售货方法应用于商品识别服务器，该方法包括：接收目标无人售货机的启动信号；启动信号由柜门被打开的无人售货机通过控制器进行发送；启动信号中携带有目标无人售货机的识别标识；接收目标无人售货机的商品图像；商品图像包括：目标无人售货机的柜门从打开到关闭时间内，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像；根据多个商品图像及商品识别模型，识别出多个商品信息；基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单；预购物清单包括：商品名称、商品数量、商品单价及总价。本申请能够通过商品识别服务器及无人售货机的配合，实现快速、高效的售货服务，大大提高用户的体购物验。
本申请的其他特征和优点将在随后的说明书中阐述，并且，部分地从说明书中变得显而易见，或者通过实施本申请而了解。本申请的目的和其他优点在说明书、权利要求书以及附图中所特别指出的结构来实现和获得。
为使本申请的上述目的、特征和优点能更明显易懂，下文特举较佳实施例，并配合所附附图，作详细说明如下。
附图说明
为了更清楚地说明本申请具体实施方式或现有技术中的技术方案，下面将对具体实施方式或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施方式，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的一种无人售货方法的流程图；
图2为本申请实施例提供的另一种无人售货方法的流程图；
图3为本申请实施例提供的另一种无人售货方法的流程图；
图4为本申请实施例提供的另一种无人售货方法的流程图；
图5为本申请实施例提供的另一种无人售货方法的流程图；
图6为本申请实施例提供的一种无人售货装置的示意图；
图7为本申请实施例提供的一种无人售货系统的示意图。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合附图对本申请的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
现有的自助售货机提供的物品有限，通常只是可乐、雪碧等饮料类物品，且需要用户进行相对复杂的操作，使得用户购物效率低下。基于此，本申请实施例提供一种无人售货方法、装置、系统、服务器及计算机可读存储介质，能够通过商品识别服务器及无人售货机的配合，实现快速、高效的售货服务，大大提高用户的体购物验。
为便于对本实施例进行理解，首先对本申请实施例所公开的一种无人售货方法进行详细介绍。
本申请实施例提供了一种无人售货方法，该方法应用于商品识别服务器，该商品识别服务器可以与至少一个无人售货机通信连接，多个无人售货机可以预先布设在不同位置，诸如商场、游乐场所、街头等，在此不进行限定。参见图1所示，该方法包括以下步骤：
S11：接收目标无人售货机的启动信号；该启动信号由柜门被打开的无人售货机通过控制器发送给商品识别服务器；启动信号中携带有目标无人售货机的识别标识。在本实施例中，目标无人售货机也即柜门被打开的无人售货机。
具体实现的时候，当用户打开无人售货机的柜门时，无人售货机中诸如红外传感器等柜门状态监测器件可以感应到柜门被打开，这时，无人售货机中的控制器会接收到柜门状态监测器件所感应到的开门信号，与此同时，控制器向商品识别服务器发送启动信号，该启动信号中携带有目标无人售货机的识别标识，该识别标识可以是无人售货机的编号、代号或者其它可以用于区别于其它无人售货机的身份信息。
S12：接收目标无人售货机的商品图像；该商品图像可以包括：目标无人售货机的柜门从打开到关闭的期间内，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像。
其中，目标无人售货机的柜门的打开或者关闭均由无人售货机中的柜门状态监测器件来判断，而摄像装置所采集的商品图像均是从柜门打开到关闭的期间内的。多个商品图像可以为拍摄时间相对连续的一组商品图像。
在一些实施方式中，商品图像可以包括以下两种情况：
第一种：目标无人售货机的柜门从打开到关闭的期间内，每隔预设时长，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像。其中，预计时长可以根据实际情况进行调整；诸如，预设时长可以设定为2s等。
第二种：目标无人售货机的柜门从打开到关闭的期间内，柜门的打开时刻、目标无人售货机的重力传感器发生变化的时刻、柜门的关闭时刻，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像。其中，重力传感器发生变化的时刻可以包括多个时刻，比如用户购买多个商品的情况下，重力传感器会多次感应到重力变化，相应的，摄像装置所采集的商品图像也是多个。
需要说明的是，上述采集的多个商品图像是按拍摄时间顺序排列的，商品识别服务器可按照时间序列对目标无人售货机上传的多个商品图像进行存储，以便后续进行识别和分析。
本申请实施例中，摄像装置可以设置多个，分别安装于柜体内的多个位置，以便提高后续对图像识别的准确性。具体实现时，多个摄像装置可安装于柜体顶部、安装于柜体内每层置物架的隔板下侧(可用于采集下层置物架上的商品图像)、上下两层相邻置物架之间的侧壁上等，具体可根据无人售货机的内部结构以及商品布局进行设置，在此不进行限制。可以理解的是，本实施例提供的无人售货机内的摄像装置能够采集位于柜体内的所有商品的图像，
S13：根据多个商品图像及商品识别模型，识别出多个商品信息。
商品识别服务器中预先存储有商品识别模型，该模型通过神经网络对商品数据进行训练得到，具体的过程如下：
首先，获取商品图像样本数据；该商品图像样本数据可以包括：预设数量的商品图像及商品信息，其中，商品图像包括：不同角度的各种商品的图像，商品信息包括：图像所对应的商品名称、商品重量等多种商品信息。
然后，基于神经网络模型对商品图像样本数据进行训练，生成商品识别模型。
在一种优选的实施方式中，在获取到大量的商品图像样本数据后，会将商品数据样本按照预设比例划分为训练数据样本和验证数据样本。预设比例可以采用百分比的形式体现，诸如，70％的数据用来训练模型，30％数据用来进行模型准确性的验证。然后将训练数据样本输入神经网络模型中进行训练，从而生成商品识别模型。此外，在训练好模型之后，还可以将验证数据样本输入商品识别模型中，对商品识别模型的准确性进行验证。当验证结果不够理想时，对模型参数进行调整，不断对模型进行优化，提高商品识别模型的准确性。
上述训练好的商品识别模型预先存储于商品识别服务器中，可以对后续获取到的商品图像进行准确识别。
具体的，将多个商品图像分别输入商品识别模型中；基于商品识别模型的多个输出结果，得到多个商品信息。
S14：基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单。
其中，预购物清单可以包括：商品名称、商品数量、商品单价及总价等。基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单，具体包括以下步骤，参见图2所示：
S21：将任意两个拍摄时间相邻的商品信息进行比对。
S22：判断任意两个拍摄时间相邻的商品信息是否相同。
如果是，则执行步骤S23：确定用户当前没有进行购物。
如果否，则执行步骤S24：根据任意两个拍摄时间相邻的商品信息中的不同的商品信息，生成多个预购商品信息；将多个预购商品信息进行汇总，并基于识别标识，生成目标无人售货机所对应的预购物清单。
具体的，将任意两个拍摄时间相邻的商品信息进行比对，判断二者是否相同，如果相同，则表明用户当前没有取走物品，也就是没有进行购物，如果不同，则表明用户取走了物品，并且下一时刻的商品图像所对应的商品信息通常会比上一时刻的商品图像对应的商品信息少，进一步可根据二者中不同的商品信息生成多个预购商品信息，在此基础上，对多个预购商品信息进行汇总叠加，得到预购物清单，同时通过上述目标无人售货机的识别标识，将该预购物清单与目标无人售货机进行对应，也就是生成目标无人售货机所对应的预购物清单。
在上述基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单之后，还包括以下步骤，参见图3所示：
S31：将预购物清单向目标无人售货机的控制器发送，以使控制器根据预购物清单生成支付信息。
在生成预购物清单后，商品识别服务器将该清单发送给目标无人售货机中的控制器，进一步使控制器根据预购物清单生成支付信息，以使用户根据该支付信息进行支付。当然，商品识别服务器也可以根据预购物清单生成支付信息，然后将支付信息直接下发给目标无人售货机的控制器。
S32：接收控制器所发送的用户根据支付信息完成支付操作后所生成的支付成功信息。
在用户根据上述支付信息完成付款后，控制器将用户支付成功信息发送给商品识别服务器，以使服务器进行后续的记录和分析。
或者，在另一种实施方式中，在上述基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单之后，还包括以下步骤，参见图4所示：
S41：接收控制器所发送的用户信息。
在用户打开柜门前，首先通过诸如手机等移动终端进行扫码操作，以通过此操作使无人售货机的柜门打开，在此过程中，控制器会自动接收到用户的身份信息，比如用户的手机号，即上述用户信息。在用户完成取物操作后，将该用户的用户信息发送给商品识别服务器。
S42：根据用户信息，将预购物清单发送至用户的第一移动终端，以使用户根据预购物清单进行付款。
进一步，商品识别服务器根据上述用户信息，将预购物清单发送给用户的移动终端，诸如可通过短信、彩信形式给用户的移动终端发送预购物清单，也可以将预购物清单发送给用户移动终端上的预设APP界面，从而使用户完成支付操作。
S43：接收用户完成付款后第一移动终端所发送的支付成功信息。
在用户根据上述预购物清单完成付款后，接收用户支付成功信息，以进行后续的记录和分析。
这里的第一移动终端用于与下述第二移动终端进行区分，第一移动终端指代用户的移动终端，而第二移动终端指代管理者的移动终端。移动终端包括：智能手机、平板电脑等。
在商品识别服务器接收到用户的支付成功信息后，还包括以下步骤，参见图5所示：
S51：根据支付成功信息，对目标无人售货机的销售情况信息进行记录。
商品识别服务器在接收到用户的支付成功信息后，对该成交单中的商品信息进行记录，以对目标无人售货机的销售情况进行统计。
S52：根据销售情况信息，确定当前存货情况及补货信息。
进一步，通过上述销售情况信息，可以知道目标无人售货机的当前商品剩余量，及确定该无人售货机需要补充的商品名称及数量，即上述存货情况和补货信息。
S53：将补货信息向管理者的第二移动终端发送。
在确定无人售货机的补货信息后，将其发送给管理者的移动终端，以使管理人员进行及时的补货操作，进而满足用户的购物需求。
本申请实施例提供的无人售货方法应用于商品识别服务器，该方法包括：接收目标无人售货机的启动信号；启动信号由柜门被打开的无人售货机通过控制器进行发送；启动信号中携带有目标无人售货机的识别标识；接收目标无人售货机的商品图像；商品图像包括：目标无人售货机的柜门从打开到关闭时间内，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像；根据多个商品图像及商品识别模型，识别出多个商品信息；基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单；预购物清单包括：商品名称、商品数量、商品单价及总价。本申请实施例能够通过商品识别服务器及无人售货机的配合，实现快速、高效的售货服务，大大提高用户的体购物验。
对应于前述无人售货方法，本申请实施例还提供一种无人售货装置，参见图6所示，该装置包括：启动信号接收模块61、图像接收模块62、信息识别模块63及清单生成模块64。
其中，启动信号接收模块61，配置成接收目标无人售货机的启动信号；启动信号由柜门被打开的无人售货机通过控制器进行发送；启动信号中携带有目标无人售货机的识别标识；图像接收模块62，配置成接收目标无人售货机的商品图像；商品图像包括：目标无人售货机的柜门从打开到关闭时间内，由目标无人售货机的摄像装置所采集的箱体内的多个商品图像；信息识别模块63，配置成根据多个商品图像及商品识别模型，识别出多个商品信息；清单生成模块64，基于识别标识及多个商品信息，生成目标无人售货机所对应的预购物清单；预购物清单包括：商品名称、商品数量、商品单价及总价。
可选的，上述装置还包括：
样本数据获取模块，配置成获取商品图像样本数据；
模型生成模型，配置成基于神经网络模型对商品图像样本数据进行训练，生成商品识别模型。
可选的，上述信息识别模块进一步配置成：将多个商品图像分别输入商品识别模型中；基于商品识别模型的多个输出结果，得到多个商品信息。
可选的，上述清单生成模块进一步配置成：将任意两个拍摄时间相邻的商品信息进行比对；判断任意两个拍摄时间相邻的商品信息是否相同；如果是，则确定用户当前没有进行购物；如果否，则根据任意两个拍摄时间相邻的商品信息中的不同的商品信息，生成多 个预购商品信息；将多个预购商品信息进行汇总，并基于识别标识，生成目标无人售货机所对应的预购物清单。
可选的，上述装置还包括：
第一清单发送模块，配置成将预购物清单向目标无人售货机的控制器发送，以使控制器根据预购物清单生成支付信息；
第一信息接收模块，配置成接收控制器所发送的用户根据支付信息完成支付操作后所生成的支付成功信息。
可选的，上述装置还包括：
用户信息接收模块，配置成接收控制器所发送的用户信息；
第二清单发送模块，配置成根据用户信息，将预购物清单发送至用户的第一移动终端，以使用户根据预购物清单进行付款；
第二信息接收模块，配置成接收用户完成付款后第一移动终端所发送的支付成功信息。
可选的，上述装置还包括：
根据支付成功信息，对目标无人售货机的销售情况信息进行记录；
根据销售情况信息，确定当前存货情况及补货信息；
将补货信息向管理者的第二移动终端发送。
本申请实施例所提供的无人售货装置中，各个模块与前述无人售货方法具有相同的技术特征，因此，同样可以实现上述功能。本装置中各个模块的具体工作过程参见上述方法实施例，在此不再赘述。本申请实施例还提供一种无人售货系统，参见图7所示，该系统包括：多个无人售货机71(图中仅示出一个)、商品识别服务器72及移动终端73；商品识别服务器72上安装有如实施例二所述的无人售货装置721；商品识别服务器72、多个无人售货机71及移动终端73两两通信连接；移动终端73包括：第一移动终端和第二移动终端；无人售货机71包括：售货机本体711及安装于售货机本体中的控制器712、摄像装置713和重力传感器714；售货机本体711包括柜门、箱体。
本申请实施例所提供的无人售货系统中，各个模块与前述无人售货方法具有相同的技术特征，因此，同样可以实现上述功能。本系统中各个模块的具体工作过程参见上述方法实施例，在此不再赘述。
本申请实施例还提供一种服务器，包括存储器以及处理器，所述存储器用于存储支持处理器执行前述任一项无人售货方法的程序，所述处理器被配置为用于执行所述存储器中存储的程序。执行无人售货方法的商品识别服务器即可采用本申请实施例还提供的上述服务器实现。本申请实施例还提供一种计算机可读存储介质，其特征在于，所述计算机 可读存储介质存储有处理器可执行的程序代码，所述程序代码被配置成使所述处理器执行前述任一项无人售货方法。本申请实施例所提供的无人售货方法的计算机程序产品，包括存储了处理器可执行的非易失的程序代码的计算机可读存储介质，所述程序代码包括的指令可用于执行前面方法实施例中所述的方法，具体实现可参见方法实施例，在此不再赘述。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的装置及电子设备的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
附图中的流程图和框图显示了根据本申请的多个实施例方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或代码的一部分，所述模块、程序段或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
在本申请的描述中，需要说明的是，术语“中心”、“上”、“下”、“左”、“右”、“竖直”、“水平”、“内”、“外”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本申请和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本申请的限制。此外，术语“第一”、“第二”、“第三”仅用于描述目的，而不能理解为指示或暗示相对重要性。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，又例如，多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些通信接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个处理器可执行的非易失的计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是：以上所述实施例，仅为本申请的具体实施方式，用以说明本申请的技术方案，而非对其限制，本申请的保护范围并不局限于此，尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，其依然可以对前述实施例所记载的技术方案进行修改或可轻易想到变化，或者对其中部分技术特征进行等同替换；而这些修改、变化或者替换，并不使相应技术方案的本质脱离本申请实施例技术方案的精神和范围，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应所述以权利要求的保护范围为准。
工业实用性：
通过应用本申请的技术方案，能够通过商品识别服务器及无人售货机的配合，实现快速、高效的售货服务，大大提高用户的体购物验。

Claims (19)

  1. 一种无人售货方法，其特征在于，应用于商品识别服务器，所述方法包括：
    接收目标无人售货机的启动信号；所述启动信号由柜门被打开的无人售货机通过控制器进行发送；所述启动信号中携带有所述目标无人售货机的识别标识；
    接收所述目标无人售货机的商品图像；所述商品图像包括：所述目标无人售货机的柜门从打开到关闭时间内，由所述目标无人售货机的摄像装置所采集的箱体内的多个商品图像；
    根据所述多个商品图像及商品识别模型，识别出多个商品信息；
    基于所述识别标识及所述多个商品信息，生成所述目标无人售货机所对应的预购物清单；所述预购物清单包括：商品名称、商品数量、商品单价及总价。
  2. 根据权利要求1所述的方法，其特征在于，还包括：
    获取商品图像样本数据；
    基于神经网络模型对所述商品图像样本数据进行训练，生成所述商品识别模型。
  3. 根据权利要求1或2所述的方法，其特征在于，所述商品图像包括：所述目标无人售货机的柜门从打开到关闭的期间内，每隔预设时间，由所述目标无人售货机的摄像装置所采集的箱体内的多个商品图像；
    或者，
    所述商品图像包括：所述目标无人售货机的柜门从打开到关闭的期间内，所述柜门的打开时刻、所述目标无人售货机的重力传感器发生变化的时刻、所述柜门的关闭时刻，由所述目标无人售货机的摄像装置所采集的箱体内的多个商品图像。
  4. 根据权利要求1至3任一项所述的方法，其特征在于，所述根据所述多个商品图像及商品识别模型，识别出多个商品信息，具体包括：
    将所述多个商品图像分别输入所述商品识别模型中；
    基于所述商品识别模型的多个输出结果，得到所述多个商品信息。
  5. 根据权利要求1至4任一项所述的方法，其特征在于，所述基于所述识别标识及所述多个商品信息，生成所述目标无人售货机所对应的预购物清单，具体包括：
    将任意两个拍摄时间相邻的商品信息进行比对；
    判断所述任意两个拍摄时间相邻的商品信息是否相同；
    如果是，则确定用户当前没有进行购物；
    如果否，则根据所述任意两个拍摄时间相邻的商品信息中的不同的商品信息，生成多个预购商品信息；
    将多个所述预购商品信息进行汇总，并基于所述识别标识，生成所述目标无人售货机所对应的预购物清单。
  6. 根据权利要求1至5任一项所述的方法，其特征在于，在所述基于所述识别标识及所述多个商品信息，生成所述目标无人售货机所对应的预购物清单之后，还包括：
    将所述预购物清单向所述目标无人售货机的控制器发送，以使所述控制器根据所述预购物清单生成支付信息；
    接收所述控制器所发送的用户根据所述支付信息完成支付操作后所生成的支付成功信息。
  7. 根据权利要求1至5任一项所述的方法，其特征在于，在所述基于所述识别标识及所述多个商品信息，生成所述目标无人售货机所对应的预购物清单之后，还包括：
    根据所述预购物清单生成支付信息，将所述支付信息发送给所述目标无人售货机的控制器；
    接收所述控制器所发送的用户根据所述支付信息完成支付操作后所生成的支付成功信息。
  8. 根据权利要求1至5任一项所述的方法，其特征在于，在所述基于所述识别标识及所述多个商品信息，生成所述目标无人售货机所对应的预购物清单之后，还包括：
    接收所述控制器所发送的用户信息；
    根据所述用户信息，将所述预购物清单发送至所述用户的第一移动终端，以使所述用户根据所述预购物清单进行付款；
    接收用户完成付款后所述第一移动终端所发送的支付成功信息。
  9. 根据权利要求6至8任一项所述的方法，其特征在于，还包括：
    根据所述支付成功信息，对所述目标无人售货机的销售情况信息进行记录；
    根据所述销售情况信息，确定当前存货情况及补货信息；
    将所述补货信息向管理者的第二移动终端发送。
  10. 一种无人售货装置，其特征在于，包括：
    启动信号接收模块，配置成接收目标无人售货机的启动信号；所述启动信号由柜门被打开的无人售货机通过控制器进行发送；所述启动信号中携带有所述目标无人售货机的识别标识；
    图像接收模块，配置成接收所述目标无人售货机的商品图像；所述商品图像包括：所述目标无人售货机的柜门从打开到关闭时间内，由所述目标无人售货机的摄像装置所采集的箱体内的多个商品图像；
    信息识别模块，配置成根据所述多个商品图像及商品识别模型，识别出多个商品信息；
    清单生成模块，配置成基于所述识别标识及所述多个商品信息，生成所述目标无人售货机所对应的预购物清单；所述预购物清单包括：商品名称、商品数量、商品单价及总价。
  11. 根据权利要求10所述的装置，其特征在于，所述装置还包括：
    样本数据获取模块，配置成获取商品图像样本数据；
    模型生成模型，配置成基于神经网络模型对所述商品图像样本数据进行训练，生成所述商品识别模型。
  12. 根据权利要求10或11所述的装置，其特征在于，所述信息识别模块进一步配置成：将所述多个商品图像分别输入所述商品识别模型中；基于所述商品识别模型的多个输出结果，得到所述多个商品信息。
  13. 根据权利要求10至12任一项所述的装置，其特征在于，所述清单生成模块进一步配置成：将任意两个拍摄时间相邻的商品信息进行比对；判断所述任意两个拍摄时间相邻的商品信息是否相同；如果是，则确定用户当前没有进行购物；如果否，则根据所述任意两个拍摄时间相邻的商品信息中的不同的商品信息，生成多个预购商品信息；将多个所述预购商品信息进行汇总，并基于所述识别标识，生成所述目标无人售货机所对应的预购物清单。
  14. 根据权利要求10至13任一项所述的装置，其特征在于，所述装置还包括：
    第一清单发送模块，配置成将所述预购物清单向所述目标无人售货机的控制器发送，以使所述控制器根据所述预购物清单生成支付信息；
    第一信息接收模块，配置成接收所述控制器所发送的用户根据所述支付信息完成支付操作后所生成的支付成功信息。
  15. 根据权利要求10至13任一项所述的装置，其特征在于，所述装置还包括：
    用户信息接收模块，配置成接收所述控制器所发送的用户信息；
    第二清单发送模块，配置成根据所述用户信息，将所述预购物清单发送至所述用户的第一移动终端，以使所述用户根据所述预购物清单进行付款；
    第二信息接收模块，配置成接收用户完成付款后所述第一移动终端所发送的支付成功信息。
  16. 根据权利要求14或15所述的装置，其特征在于，所述装置还包括：
    根据所述支付成功信息，对所述目标无人售货机的销售情况信息进行记录；
    根据所述销售情况信息，确定当前存货情况及补货信息；
    将所述补货信息向管理者的第二移动终端发送。
  17. 一种无人售货系统，其特征在于，包括：多个无人售货机、商品识别服务器及移动终端；
    所述商品识别服务器上安装有如权利要求9所述的无人售货装置；
    所述商品识别服务器、多个所述无人售货机及所述移动终端两两通信连接；
    所述移动终端包括：第一移动终端和第二移动终端；
    所述无人售货机包括：售货机本体及安装于所述售货机本体中的控制器、摄像装置和重力传感器；所述售货机本体包括柜门、箱体。
  18. 一种服务器，其特征在于，包括存储器以及处理器，所述存储器用于存储支持处理器执行权利要求1至9任一项所述方法的程序，所述处理器被配置为用于执行所述存储器中存储的程序。
  19. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有处理器可执行的程序代码，所述程序代码被配置成使所述处理器执行所述权利要求1-9任一项所述方法。
PCT/CN2018/108182 2018-09-06 2018-09-28 无人售货方法、装置、系统、服务器及计算机可读存储介质 WO2020047919A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811040362.8A CN109190705A (zh) 2018-09-06 2018-09-06 无人售货方法、装置及系统
CN201811040362.8 2018-09-06

Publications (1)

Publication Number Publication Date
WO2020047919A1 true WO2020047919A1 (zh) 2020-03-12

Family

ID=64914980

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/108182 WO2020047919A1 (zh) 2018-09-06 2018-09-28 无人售货方法、装置、系统、服务器及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN109190705A (zh)
WO (1) WO2020047919A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111784445A (zh) * 2020-06-18 2020-10-16 杭州海康威视系统技术有限公司 一种商品购物清单的确定方法、装置、系统及电子设备
CN113515543A (zh) * 2021-03-23 2021-10-19 广东便捷神科技股份有限公司 一种无人售货机自动提货方法及系统
CN114529317A (zh) * 2020-10-30 2022-05-24 广东飞企互联科技股份有限公司 一种店铺商品的浏览量和成交量监控方法和系统
CN116684626A (zh) * 2023-08-04 2023-09-01 广东星云开物科技股份有限公司 视频压缩方法和共享售卖柜

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109886169B (zh) * 2019-02-01 2022-11-22 腾讯科技（深圳）有限公司 应用于无人货柜的物品识别方法、装置、设备及存储介质
CN115830764A (zh) * 2019-03-15 2023-03-21 万维数码智能有限公司 通过智能货架系统对货物进行控制的装置及其方法
CN109948538A (zh) * 2019-03-19 2019-06-28 成都好享你网络科技有限公司 基于行为检测的视觉识别方法、装置和自动售货机
CN112466036A (zh) * 2019-09-09 2021-03-09 阿里健康科技(杭州)有限公司 售货机、售货机出货检测方法及装置
CN110675106A (zh) * 2019-09-12 2020-01-10 创新奇智(合肥)科技有限公司 一种基于动态商品库存信息的无人货柜商品识别方法
CN111126264A (zh) * 2019-12-24 2020-05-08 北京每日优鲜电子商务有限公司 图像处理方法、装置、设备及存储介质
CN111724230B (zh) * 2020-05-18 2021-10-29 深圳易马达科技有限公司 自助售货方法、自助售货机、终端设备、服务器及系统
CN111815852A (zh) * 2020-07-07 2020-10-23 武汉马克到家科技有限公司 一种用于开门自取式售货柜的图像和重力双模自动商品识别系统
CN112950329A (zh) * 2021-03-26 2021-06-11 苏宁易购集团股份有限公司 商品动态信息生成方法、装置、设备及计算机可读介质
CN113111801A (zh) * 2021-04-19 2021-07-13 北京每日优鲜电子商务有限公司 自动售货机的货品校验方法、设备和存储介质
CN113128898A (zh) * 2021-04-30 2021-07-16 上海商汤智能科技有限公司 售货机管理方法及装置、电子设备和存储介质
CN113470250B (zh) * 2021-06-30 2023-02-17 杭州海康机器人股份有限公司 售卖柜、监控方法、服务器、系统及计算机可读存储介质
CN113723383B (zh) * 2021-11-03 2022-06-28 武汉星巡智能科技有限公司 异视角同步识别同区域商品的订单生成方法及智能售货机
CN113727029B (zh) * 2021-11-03 2022-03-18 武汉星巡智能科技有限公司 多视角采集图像后合并的智能生成订单方法及智能售货机
CN115512484B (zh) * 2022-09-14 2023-06-02 国筹(北京)数据科技有限公司 一种智能售货系统及其控制方法
CN116205554B (zh) * 2023-04-26 2024-02-09 浙江天柜科技有限公司 一种移动自助售货设备及售货控制方法
CN116862390B (zh) * 2023-09-01 2023-12-12 常熟市恒仕达电器有限公司 商品售卖柜的管理方法、商品售卖柜及计算机存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106971457A (zh) * 2017-03-07 2017-07-21 深圳市楼通宝实业有限公司 自助售卖方法及系统
CN107146325A (zh) * 2017-07-13 2017-09-08 杭州宗盛智能科技有限公司 自动售货机及其控制方法、控制装置和系统
US20170301173A1 (en) * 2014-09-23 2017-10-19 Bubbly Group As Vending machine
CN107833366A (zh) * 2017-12-12 2018-03-23 成都缤柜科技有限公司 一种基于感应开关的自选式自动售货机
CN107918983A (zh) * 2017-11-15 2018-04-17 深圳码隆科技有限公司 一种基于神经网络的无人售货方法和设备
CN108320379A (zh) * 2018-02-28 2018-07-24 成都果小美网络科技有限公司 基于图像比对的售货方法、装置和无人售货机

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170301173A1 (en) * 2014-09-23 2017-10-19 Bubbly Group As Vending machine
CN106971457A (zh) * 2017-03-07 2017-07-21 深圳市楼通宝实业有限公司 自助售卖方法及系统
CN107146325A (zh) * 2017-07-13 2017-09-08 杭州宗盛智能科技有限公司 自动售货机及其控制方法、控制装置和系统
CN107918983A (zh) * 2017-11-15 2018-04-17 深圳码隆科技有限公司 一种基于神经网络的无人售货方法和设备
CN107833366A (zh) * 2017-12-12 2018-03-23 成都缤柜科技有限公司 一种基于感应开关的自选式自动售货机
CN108320379A (zh) * 2018-02-28 2018-07-24 成都果小美网络科技有限公司 基于图像比对的售货方法、装置和无人售货机

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111784445A (zh) * 2020-06-18 2020-10-16 杭州海康威视系统技术有限公司 一种商品购物清单的确定方法、装置、系统及电子设备
CN111784445B (zh) * 2020-06-18 2024-03-01 杭州海康威视系统技术有限公司 一种商品购物清单的确定方法、装置、系统及电子设备
CN114529317A (zh) * 2020-10-30 2022-05-24 广东飞企互联科技股份有限公司 一种店铺商品的浏览量和成交量监控方法和系统
CN113515543A (zh) * 2021-03-23 2021-10-19 广东便捷神科技股份有限公司 一种无人售货机自动提货方法及系统
CN113515543B (zh) * 2021-03-23 2024-02-13 广东便捷神科技股份有限公司 一种无人售货机自动提货方法及系统
CN116684626A (zh) * 2023-08-04 2023-09-01 广东星云开物科技股份有限公司 视频压缩方法和共享售卖柜
CN116684626B (zh) * 2023-08-04 2023-11-24 广东星云开物科技股份有限公司 视频压缩方法和共享售卖柜

Also Published As

Publication number Publication date
CN109190705A (zh) 2019-01-11

Similar Documents

Publication Publication Date Title
WO2020047919A1 (zh) 无人售货方法、装置、系统、服务器及计算机可读存储介质
CN108922026B (zh) 一种自动售货机的补货管理方法、装置和用户终端
WO2020010827A1 (zh) 一种基于开放式自动售货机的购物结算方法和装置
CN108171879B (zh) 一种自动售货结算的方法、装置及无人售货柜
CN109979130A (zh) 一种商品自动识别和结算售货柜、方法以及系统
CN109215254B (zh) 实体店铺应用系统、用户信息处理方法及装置
JP2020526847A (ja) 価値のある電子装置を遠隔収集するための低コストのコンピュータ化した売店
CN111222870B (zh) 结算方法、装置和系统
JP6821009B2 (ja) 会計方法、装置及びシステム
CN109241883B (zh) 一种基于自动售货机的退货控制方法及其装置
WO2021179137A1 (zh) 结算方法、装置和系统
CN108198315A (zh) 一种身份验证方法以及身份验证装置
US11023728B1 (en) Machine learning algorithm trained to identify algorithmically populated shopping carts as candidates for verification
WO2020047918A1 (zh) 无人售货方法、装置、系统、电子设备及计算机存储介质
CN108805644A (zh) 贩售机的商品售卖方法以及贩售机
CN109934569B (zh) 结算方法、装置和系统
CN110490697A (zh) 无人便利店结算方法、装置、计算机及存储介质
US20220309784A1 (en) System and method for populating a virtual shopping cart based on a verification of algorithmic determinations of items selected during a shopping session in a physical store
CN110689389A (zh) 基于计算机视觉的购物清单自动维护方法及装置、存储介质、终端
CN110555939A (zh) 自动售货机
CN111222899A (zh) 商品信息及补货的处理方法、设备及存储介质
JP2023504871A (ja) 不正検出システムおよび方法
Hsu et al. New generation artificial intelligent vending machine system based on LoRaWan IOT network
CN109858448A (zh) 一种零售场景下的物品识别方法及设备
US11475656B2 (en) System and method for selectively verifying algorithmically populated shopping carts

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18932556

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18932556

Country of ref document: EP

Kind code of ref document: A1