WO2019057168A1 - 一种货物订单处理方法、装置、服务器、购物终端及系统 - Google Patents

一种货物订单处理方法、装置、服务器、购物终端及系统 Download PDF

Info

Publication number
WO2019057168A1
WO2019057168A1 PCT/CN2018/107028 CN2018107028W WO2019057168A1 WO 2019057168 A1 WO2019057168 A1 WO 2019057168A1 CN 2018107028 W CN2018107028 W CN 2018107028W WO 2019057168 A1 WO2019057168 A1 WO 2019057168A1
Authority
WO
WIPO (PCT)
Prior art keywords
goods
image
information
storage container
order
Prior art date
Application number
PCT/CN2018/107028
Other languages
English (en)
French (fr)
Inventor
张鸿
曾晓东
Original Assignee
阿里巴巴集团控股有限公司
张鸿
曾晓东
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 阿里巴巴集团控股有限公司, 张鸿, 曾晓东 filed Critical 阿里巴巴集团控股有限公司
Priority to EP18857658.1A priority Critical patent/EP3605431A4/en
Priority to SG11201910119V priority patent/SG11201910119VA/en
Publication of WO2019057168A1 publication Critical patent/WO2019057168A1/zh
Priority to US16/692,316 priority patent/US10791199B2/en
Priority to US16/945,650 priority patent/US11019180B2/en

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/06Buying, selling or leasing transactions
    • G06Q30/0601Electronic shopping [e-shopping]
    • G06Q30/0633Lists, e.g. purchase orders, compilation or processing
    • G06Q30/0635Processing of requisition or of purchase orders
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/08Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/087Inventory or stock management, e.g. order filling, procurement or balancing against orders
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/28Constructional details of speech recognition systems
    • G10L15/30Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/02Protocols based on web technology, e.g. hypertext transfer protocol [HTTP]
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/10Protocols in which an application is distributed across nodes in the network
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/30Definitions, standards or architectural aspects of layered protocol stacks
    • H04L69/32Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level
    • H04L69/322Intralayer communication protocols among peer entities or protocol data unit [PDU] definitions
    • H04L69/329Intralayer communication protocols among peer entities or protocol data unit [PDU] definitions in the application layer [OSI layer 7]

Definitions

  • the embodiment of the present specification belongs to the technical field of computer data processing, and in particular, to a method, device, server, shopping terminal and system for processing a goods order.
  • Unmanned supermarkets are a market development hotspot that is currently focused on by major commodity shopping platform operators. It involves technical issues on how to conveniently and reliably process user order data.
  • a self-service order settlement scheme can be adopted in an unmanned supermarket, mainly by arranging a plurality of sensing devices, such as a camera, a pressure sensor, an infrared sensor, a wireless network, etc., on a store shelf or an aisle to monitor the user from the shelf.
  • the goods taken on, and then combined with self-service scan code payment, shopping cart weighing and other equipment to complete the order settlement.
  • This approach requires multiple types of ancillary equipment, and requires a large number of devices to be deployed in multiple locations in the mall, requiring high hardware costs for service implementation.
  • this method requires the data of each sensing device to cooperate with each other to count the goods, and the stability is difficult to guarantee, which is not conducive to large-scale popularization and use. Therefore, there is still a need for a cheaper, simpler, and more reliable order processing solution in an unmanned supermarket.
  • the embodiment of the present specification aims to provide a method, a device, a server, a shopping terminal and a system for processing a goods order, which can provide an order settlement method that is simple, reliable, and has a lower implementation cost, and can greatly reduce the automatic processing of unmanned supermarket orders.
  • a method for processing a goods order comprising:
  • the client detects whether the goods in the storage container have changed based on image recognition
  • the client sends an associated image of the change in the goods in the storage container to the server;
  • the server identifies difference information of the goods in the storage container based on the associated image
  • the server generates the corresponding goods order change information by using the difference information
  • the server updates the user order information corresponding to the client according to the goods order change information
  • the server sends the updated user order information to the client
  • the client displays the updated user order information.
  • a method for processing a goods order comprising:
  • the updated user order information is sent to the client.
  • a method for processing a goods order comprising:
  • a goods order processing device comprising:
  • An image receiving module configured to receive an associated image uploaded by the client, where the associated image includes an image that is detected by the client when the goods in the storage container change according to an image recognition manner;
  • An image recognition module configured to identify difference information of the goods in the storage container based on the associated image
  • An order generating module configured to generate corresponding goods order change information by using the difference information
  • An order update module configured to update user order information corresponding to the client according to the goods order change information
  • the information feedback module is configured to send the updated user order information to the client.
  • a goods order processing device comprising:
  • a cargo change detection module for detecting whether a change in the goods in the storage container is based on an image recognition method
  • An image sending module configured to send an associated image of the change in the storage container to the server when the change of the goods is detected
  • An information receiving module configured to receive updated user order information returned by the server
  • a display module is configured to display the updated user order information.
  • a server comprising at least one processor and a memory for storing processor-executable instructions, the processor implementing the instructions to:
  • the updated user order information is sent to the client.
  • a shopping terminal comprising at least one processor and a memory for storing processor-executable instructions, the processor implementing the instructions to:
  • a goods order processing system comprising: a shopping terminal, a server,
  • the server includes the apparatus described in any one of the embodiments of the present specification,
  • the shopping terminal includes the shopping terminal described in any one of the embodiments of the present specification;
  • the server implements the steps of the method described in any one of the embodiments of the present specification,
  • the shopping terminal implements the steps of the method described in any one of the embodiments of the present specification.
  • a cargo order processing method, device, server, shopping terminal and system provided by one or more embodiments of the present specification can install a photographing device in a storage container such as a shopping basket or a shopping cart, and the remote server can monitor the storage in real time. Changes in the goods in the container to identify and confirm changes in the user's order, and can generate or update the user's shopping order.
  • the terminal equipment of the smart shopping cart and the server for the cargo identification and order processing can be used to realize the unmanned supermarket order processing order data, thereby saving a large number of self-service cash register devices and sensing devices.
  • the image data of the goods detection can be sent to the server through the client image or the video capturing device, and the server performs the processing of the goods identification, the order settlement, etc.
  • the overall hardware implementation is lower, and the data interaction can be reduced.
  • the implementation of the program is simpler and more convenient.
  • the implementation cost of the technical solution for automatic settlement of unmanned supermarket goods orders can be greatly reduced, which is easy to implement and has better stability.
  • the implementation environment and factors of unmanned supermarket area, passenger flow and consumer behavior characteristics are less affected. It is conducive to large-scale promotion and use.
  • FIG. 2 is a schematic diagram of an implementation scenario of a client taking a shopping cart cargo in the present specification
  • Figure 4 is a schematic diagram of the boundary of a cargo identifying the change of the goods in the present specification
  • Figure 5 is an image in the boundary of the goods identified in an implementation scenario of the present specification
  • Figure 6 is a schematic view showing a scene of stacking another piece of goods on the first piece of the shopping cart shown in Figure 3;
  • Figure 8 is a schematic flow chart of another embodiment of the method provided by the present specification.
  • FIG. 9 is a schematic diagram of an embodiment of a method for processing a goods order provided by the present specification.
  • 11 is a flow chart showing an embodiment of the method that can be used on the client side provided by the present specification.
  • FIG. 12 is a flow chart showing another embodiment of the method that can be used on the client side provided by the present specification.
  • FIG. 13 is a schematic structural diagram of a module of an embodiment of a cargo order processing apparatus that can be used on a server side provided by the present specification;
  • FIG. 14 is a block diagram showing the structure of an embodiment of an image recognition module in the apparatus provided by the present specification.
  • Figure 16 is a block diagram showing the structure of another embodiment of the apparatus provided in the present specification.
  • Figure 17 is a block diagram showing the structure of another embodiment of the apparatus provided in the present specification.
  • FIG. 18 is a block diagram showing the structure of an embodiment of a cargo order processing apparatus according to one embodiment of the present specification
  • Figure 19 is a block diagram showing the structure of an embodiment of a cargo change detecting module in the device of the present specification.
  • Figure 20 is a block diagram showing the structure of another embodiment of the cargo change detecting module in the apparatus of the present specification.
  • 21 is a block diagram showing the structure of another embodiment of the apparatus of the present specification.
  • 22 is a schematic structural diagram of an embodiment of the apparatus implemented in conjunction with computer hardware provided by the present specification
  • the embodiments provided in this specification can be used in an application scenario including, but not limited to, an unmanned supermarket, and a system architecture of C/S (Client/Server, Client/Server) can be adopted.
  • C/S Customer/Server, Client/Server
  • various types of goods loading equipment such as a hand-held shopping basket, a hand-drawn shopping basket, a shopping cart, a cargo handling vehicle, etc., which can be provided in an unmanned supermarket (for convenience of description, may be collectively referred to in some embodiments of the present specification).
  • a camera or video recording device such as a smartphone, a camera, a monitor, or a camera is installed in the storage container.
  • These camera or video recording devices may contain image change detection logic that can detect changes in the contents of the storage container based on the captured image or video image.
  • the client can send the image of the change of the goods to the server side for processing, and the server side can identify the goods added by the user to the storage container or the goods taken from the storage container according to the image, and correspondingly generate the goods.
  • Change information Further, the total order information of the user may be generated or updated correspondingly based on the information of the goods change in the storage container. The total order information of these updated goods can be instantly displayed on the client display device for the user to view, confirm, and the like.
  • the C/S described above may be a logically divided client/server architecture, that is, in some embodiments of the present specification, the client may include collecting images of goods in the storage container.
  • the terminal device, the server may include a terminal device that acquires an image collected by the client and performs cargo identification and order processing based on the image.
  • the implementation of the present specification may also be used in an implementation scenario where the client and the server are the same physical device, for example, the device on the smart shopping cart may include both an image change detection module and an image.
  • the cargo identification module and/or the order data processing module each smart shopping cart can perform self-service cargo change detection, cargo identification, order processing, etc., and all smart shopping cart order data can be aggregated into a designated order business server.
  • One or more embodiments of the present specification will be described in detail below with an application scenario in which an order processing is performed on a customer's shopping cart in an unmanned supermarket. It should be noted that the implementation of the present specification is not limited to the implementation application of the unmanned supermarket scene. In other implementation scenarios, for example, the goods handling vehicle, the container, the truck compartment, the express package, and the like identify the goods in the storage container and perform corresponding changes. The implementation scenario of the cargo order data processing can also use the embodiments provided by the embodiments of the present specification.
  • the unmanned supermarket described in this embodiment may include a cashier who has no or a few on-site explanations for the goods in the normal shopping cart, and the supermarket may include a convenience store, a large shopping mall, a self-service station, etc., and the user may be in the supermarket.
  • the shopping cart collects the purchased goods, and the photographing or camera equipment on the shopping cart can detect the change of the goods in the shopping cart, and then the server can identify the goods and generate corresponding orders. After the user completes the shopping, the user's shopping order can be generated automatically, more quickly, more reliably and stably. The user can settle the order by himself. If the payment is settled online, after the order is successfully settled, the exit or channel can be directly used to avoid the queue of commodity settlement, which can greatly improve the user's shopping experience.
  • FIG. 1 is a schematic flow chart of an embodiment of a method for processing a goods order provided by the present specification.
  • the present specification provides method operation steps or device structures as shown in the following embodiments or figures, there may be more or partial merged fewer operational steps in the method or device based on conventional or no inventive labor. Or module unit.
  • the execution order of the steps or the module structure of the device is not limited to the execution order or the module structure shown in the embodiment or the drawings.
  • the device, server or terminal product of the method or module structure When the device, server or terminal product of the method or module structure is applied, it may be executed sequentially or in parallel according to the method or module structure shown in the embodiment or the drawing (for example, parallel processor or multi-thread processing). Environment, even including distributed processing, server cluster implementation environment).
  • the method may include:
  • S0 The client detects whether the goods in the storage container have changed based on image recognition.
  • the client described in this embodiment may include an electronic device having an image change detecting function installed on the shopping cart, such as a smart phone, a camera, a monitor, and the like.
  • the client can also have communication capabilities to send captured images or videos to the server.
  • the client may be a separate device such as a terminal device for image capturing installed on a storage container and a terminal device for image change detection, for example, the client may include intelligence capable of shooting and video uploading. Mobile phones and dedicated devices that detect image changes.
  • the client may also include a terminal device composed of a plurality of devices such as an image capturing device, an image change detecting device, and a communication module, and may even be used as a whole together with a storage container such as a shopping cart.
  • the storage container may include various types of storage devices for storing customer shopping items, such as a hand-held shopping basket, a hand-push shopping cart, and the like.
  • FIG. 2 is a schematic diagram of an implementation scenario of a client capturing a shopping cart cargo in the present specification.
  • the storage container may be a shopping cart used in a supermarket, such as a conventional shopping cart of various containers of 60 liters, 80 liters, 100 liters, and the like.
  • a smartphone can be used as the client, and the smartphone can include a high definition camera and a screen.
  • a smartphone stand can be placed in the cart near the armrest of the user or the holder of the grip, which can be used to place smartphones of various sizes.
  • the smart phone can be placed at the stand so that the camera of the smart phone faces the shopping cart and the shopping cart space can be photographed.
  • the screen of the smartphone faces the user, and the screen can display the user order information that has been produced (the goods that the user puts into the shopping cart) or display other information such as shopping tips, errors, navigation, and the like.
  • the client on the shopping may be considered to be bound to the user. For example, if the user unlocks the shopping cart A by scanning, the shopping cart A passes the verification and the ID of the user. Binding to one.
  • the smartphone on the shopping cart can run the camera app.
  • the smart phone may include an image change detection module, such as an image change detection algorithm or a calculation model set in the photographing application, to enable recording of a difference image or a video containing the image of the servant before and after the image change time.
  • the smart phone may continuously take a photo of a predetermined frequency of the shopping space of the shopping cart after starting the shooting application, or may take a picture all the time.
  • the image obtained by the shooting can be detected, and the image in the shopping cart can be judged by the image recognition method, for example, the user puts in new goods or puts back the goods.
  • the manner of image recognition may include acquiring a direct difference portion of the image, and determining a change of the image by identifying the difference portion.
  • the newly updated image frame is sent to the server at a certain time interval. If the image remains static, the goods can be considered to have not changed and may not be sent.
  • the detecting the change of the goods in the storage container based on the image recognition manner may include:
  • the imaging device may include the aforementioned camera, and may also include a camera, a camera, a monitor, and the like. After obtaining the image of the goods in the storage container, the image of the goods can be detected in the chronological order of the frames to confirm whether the goods in the goods storage container have changed.
  • the image of the goods can be detected in the chronological order of the frames to confirm whether the goods in the goods storage container have changed.
  • FIG. 3 is a customer in the present specification. Take a picture of the implementation scenario of the new cargo in the shopping cart.
  • the client can detect the image change and can obtain the difference data of the image change.
  • the difference data of the image change is greater than the preset change threshold, it may be determined that the goods in the storage container have changed, thereby avoiding some image changes caused by the non-goods themselves being put in or taken out.
  • Interference information such as a small displacement of the goods in the shopping cart caused by the shaking of the shopping cart.
  • the range of values of the change threshold of the specific judgment difference data can be set according to experience or experiment.
  • the visor can be placed around the bottom of the shopping cart and the bottom surface, so that in the area photographed by the imaging device When goods are placed, it will cause significant changes in the image of the goods being photographed, and can reduce interference information such as roads, shelves, and people's shadows outside the space of the shopping cart.
  • the client may photograph the goods in the storage container, and may include image data acquired by photographing at a predetermined frequency, and may also include image data of the recorded video goods.
  • a video may be regarded as a collection of consecutive multiple images, or as one type of image data. Therefore, the image data of the cargo image, the associated image, and the like described in the present specification may include an image obtained by photographing alone, and may also include an image in the video.
  • the client detects a change in the goods in the shopping cart, such as a new item in the shopping cart, the client can detect that the captured image has changed.
  • the client may acquire an associated image in which the goods in the shopping cart change, and may send the related image to a predetermined server, such as a supermarket order processing.
  • the cloud server processes it.
  • the associated image may include an image sent by the client to the server for detecting and identifying the change of the goods, and may be a single image or multiple images.
  • the associated image may include a separate non-contiguous image taken by the client side, and may also include video data recorded by the client that detects a change in the goods. The video data at this time may be considered as continuous.
  • Associated image When determining the associated image, an image or a video having a difference before and after the detected image change timing may be employed as the associated image once the client determines that the goods in the storage container have changed. Of course, other embodiments may also use an image of a period of time before the image change time (e.g., 3 seconds) as the associated image.
  • the associated image may include at least one of the following:
  • the T0 time corresponding to the difference data data0 and the cargo images P0 and P1 at the time T1 may be used as the A correlated image of the change in the goods in the storage container detected.
  • the cargo image P2 corresponding to the time T2 and the ten cargo images before and after are selected as the related images generated to the server.
  • image data recorded by the photographing video within 3 seconds before and after the image change time T3 is selected as the related image.
  • the manner of determining the associated image may include one or more components of the foregoing manner.
  • the cargo image corresponding to the difference data may be selected and 10 times before and after the change time are selected.
  • the image (the overlapping image may not need to be repeatedly selected) is used together as the associated image.
  • Fig. 4 is a schematic diagram of the boundary of the goods identifying the change of the goods in the present specification
  • L is the bounding box of the identified pizza package, which can be found by the image subtraction and the threshold communication area processing, as shown in Fig. 4
  • the client can detect the boundary of the goods in the image in which the goods change, and then upload and identify when uploading the associated image to the server for image identification, order processing, and the like in the image. Images in the boundaries of the cargo can reduce image data transfer, save network traffic and reduce server load. Therefore, in another embodiment of the method described in the present specification, the method may further include:
  • the sending, by the client, the associated image to the server may include: sending the identified image in the cargo boundary to the server.
  • the image of the goods shown in FIG. 5 in the boundary of the goods in FIG. 4 may be uploaded to the server, and FIG. 5 is the goods identified in an implementation scenario of the present specification.
  • the image in the border As can be seen from FIG. 4 and FIG. 5, the image information in the boundary of the cargo identified by the uploading client can effectively save network traffic, reduce redundant information of transmission, reduce load of server image data processing, and provide image recognition. Processing efficiency, etc.
  • FIG. 6 and FIG. 7 are schematic diagrams of a scene in which another piece of goods is stacked on the first piece of the shopping cart shown in FIG. 3, and FIG. 7 is a new view in FIG. An image placed in the boundary of the goods in the goods.
  • the client sends the associated image to a server.
  • the client can send the acquired related image to the server through a network communication connection such as a near field communication such as a mobile phone WIFI or a local area network, or a carrier communication network or a dedicated cable.
  • the server can store associated images uploaded by the client.
  • the client may directly send the associated image to the server, for example, through an operator communication network.
  • a unified interface device of multiple clients may be set, and multiple clients in the supermarket may be used.
  • the associated image can be sent to a unified interface device, such as a supermarket server host, by near field communication, and then the associated image is sent by the supermarket server host to the server that performs the identification of the goods in the image.
  • the server identifies the difference information of the goods in the storage container based on the related image.
  • the server After the server obtains the associated image uploaded by the client, the related image may be analyzed and processed, for example, the difference of the associated image is compared in time series, and the information of the newly added or removed goods in the associated image is identified.
  • the image processed by the server may include data of the image of the image taken by the client or the image of the video of the video type, and may also include image data of the boundary of the goods recognized by the client according to the difference of the image.
  • the cargo difference information may include image data, and may also include identification information indicating a change in the storage container. For example, as shown in FIG. 3 and FIG. 4, it is recognized that a new item is placed in the shopping cart.
  • the image recognition determines that the goods is G01 and the goods identification code is good_001.
  • the difference information of the goods in the storage container can be: ADD: good_001, wherein ADD can indicate that the goods change is a new one. Therefore, in another embodiment of the method provided by the present specification, the identifying, by the server, the difference information of the goods in the storage container based on the associated image may include:
  • S40 Detect a cargo identification identifier in the associated image, and use the detected cargo identification identifier as the cargo difference information.
  • the server may be provided with a barcode detection and recognition module, and the barcode in the associated image may be identified, and the barcode may include multiple attribute information of the goods, such as a cargo name, a price, and the like.
  • the barcode detection module detects the associated image identifying the input and searches for a regular barcode. If the identification is successful, the difference information of the goods can be determined.
  • server side can also set other types of ways to detect and identify the goods in the associated image.
  • a machine learning algorithm can be utilized to detect goods in a tube-associated image. Specifically, any one of the following embodiments is adopted:
  • S42 After acquiring the associated image, using a cargo detection model obtained by a machine learning algorithm to detect cargo information in the associated image, and determining difference information of the cargo;
  • the goods detection model obtained by the machine learning algorithm detects the goods information in the related image, and determines the difference information of the goods.
  • the cargo inspection model can select SVM (Support Vector Machine), RCNN (Regions with CNN, regional convolutional neural network, a target detection algorithm), and other machine learning algorithms and variants.
  • the selected machine learning algorithm can be pre-utilized with the sample data for training construction, and the algorithm for detecting the goods in the associated image can be detected in the embodiment in accordance with the actual scene or the design requirement condition.
  • the related image may be acquired and input into the cargo detection model for cargo detection.
  • the identification of the goods identification in the associated image may be performed first, such as detecting and identifying the barcode or the two-dimensional code of the goods, and then using a higher-level machine learning algorithm to detect the associated image when the identification of the identification of the goods fails. Goods, determine the cargo information.
  • S6 The server generates the corresponding goods order change information by using the difference information.
  • the corresponding order change information may be generated based on the difference information. For example, when the corresponding goods are detected and the corresponding unique price is found, the goods order change information can be generated.
  • the goods order information may indicate that the user newly adds a cargo and can be used in the user's total order settlement process.
  • S8 The server updates the user order information corresponding to the client according to the goods order change information.
  • a total order information of the user can be recorded on the server side.
  • the client when the user unlocks the client to use the storage container for the purchase of goods, the client can bind to the identification of the user, for example, a user binds a shopping cart.
  • the server recognizes that the user newly joins or removes the goods in the shopping cart, the corresponding goods order change information may be generated, and then the user order information of the user recorded on the server side may be updated based on the goods order change information, for example, adding one A purchase order for an item or a purchase order to delete an item.
  • S10 The server sends the updated user order information to the client.
  • the client side may be provided with order information displayed to the user to be identified in the shopping cart, and the user can immediately obtain the order information.
  • the updated user's order information of the user may be generated to the client, so that the client can display or synchronize the user order information to the user.
  • the client can receive the updated user order information after the user newly puts in or takes out the item, and then can display the updated user order information, for example, in the display.
  • the smartphone on the shopping cart or on other display devices that are placed on the storage container.
  • One or more embodiments provided by the above embodiments provide a method for processing a goods order, which can be installed in a storage container such as a shopping basket or a shopping cart, and the remote server can monitor the change of the goods in the storage container in real time. To identify and confirm changes to a user's order, and to generate or update a purchase order for that user.
  • the terminal equipment of the smart shopping cart and the server for the cargo identification and order processing can be used to realize the unmanned supermarket order processing order data, thereby saving a large number of self-service cash register devices and sensing devices.
  • the image data of the goods detection can be sent to the server through the client image or the video capturing device, and the server performs the processing of the goods identification, the order settlement, etc.
  • the overall hardware implementation is lower, and the data interaction can be reduced.
  • the implementation of the program is simpler and more convenient.
  • the implementation cost of the technical solution for automatic settlement of unmanned supermarket goods orders can be greatly reduced, which is easy to implement and has better stability.
  • the implementation environment and factors of unmanned supermarket area, passenger flow and consumer behavior characteristics are less affected. It is conducive to large-scale promotion and use.
  • the present specification also provides another embodiment of the method. If the server side does not recognize the goods in the associated image during the process of identifying and detecting the associated image uploaded by the client, for example, the barcode cannot be scanned or the product detection model using the machine learning cannot identify the associated image. In this case, the information of the associated image or the user information, the storage container information, and the like can be transmitted to the manually recognized terminal device. For example, the PC end of the manual checkout counter, the dedicated cash register device, or other terminal devices that are set to handle the detection of the goods detected and failed. In the present embodiment, the above-described terminal device including the recognition failure image for manual recognition processing may be collectively referred to as a cashier node.
  • the image can be sent to the cashier node for manual identification processing, which can effectively ensure timely, stable, and continuous processing of the goods order, and reduce the user's failure to identify the goods on the server side.
  • the phenomenon of goods purchase experience is
  • a cashier node may be disposed, or multiple cashier nodes may be disposed.
  • a plurality of cashier nodes may be set in an unmanned supermarket, and each cashier node may correspond to a network cashier, and the network cash register may operate on the device of the cashier node to identify or discard the received cargo image. Or prompt the user, etc.
  • the embodiment of the present application may be configured with a task scheduling queue and a network cashier category, which can uniformly perform more reasonable task scheduling on multiple unrecognized cargo images.
  • the method may further include:
  • S48 Push the job task in the task scheduling queue to the cashier node that meets the job condition according to the obtained network cashier list, where the network cashier list records the job status of the cashier node processing task.
  • the job task may include data information to be processed formed based on the received image information of the recognition failure and other information such as a user identification, a client identification, and the like.
  • the job task in the task scheduling queue can be assigned to the cashier node, and the cashier node can perform corresponding processing according to the task information.
  • a network cashier list may also be configured.
  • the network cashier list may record the current job status of each cashier node, such as whether the cashier node is online, and how many unprocessed tasks are currently present at a cashier node.
  • the tasks in the task scheduling queue may be assigned to the cashier nodes that meet the job conditions according to the job status of each cashier node in the network cashier list.
  • the working conditions may be set according to a scene implementation scenario or a design requirement. For example, it may be specified that one or more or all job tasks may be assigned to a designated cashier node for processing when using a certain channel for person recognition, or The job task is assigned to the cashier node with the
  • the cash register node is not limited to the terminal device for manual identification processing, and may also include a terminal device that stores or records or forwards or alarms.
  • the user when the user selects the shopping cart or the server side fails to identify the goods, the user binds the identification of the user to the network cashier. Or, when there is no online cashier to process a user's job task, you can bind a network cash register to the user's job task in the task scheduling queue. In this way, when assigning a job task in the task scheduling list, it may first query whether there is a network cashier bound to the task, and if so, assign the job task to the cashier node where the bound cashier is located. .
  • the bound cashier may include a cashier who is processing a job task belonging to the same user as the job task to be assigned, or a cashier designated by the server or specified by the user.
  • the binding may also include an implementation in which the client is bound to the network cash register.
  • the method may further include:
  • the corresponding job task is sent to the cashier node where the corresponding network cashier is located for processing.
  • the method may further include:
  • the job task is sent to a cashier node in the network cashier list where the job status is non-work volume saturation.
  • a task queue may be saved, and each item of the task queue includes an associated image of the user id+identification failure or an order corresponding to the video content+user id.
  • the user id is searched to confirm whether the network cashier has been bound. If it is already, it is judged whether the workload is saturated. If it is not saturated, the job task can be preferentially pushed to the network cashier. If the workload is saturated, it is pushed to the new network cashier for processing and the user id is bound to the new cashier id.
  • This new network cashier can cash in on designated networks that are not in a job-saturated state or randomly selected cashiers or network cashiers selected in a predetermined order.
  • the job task (usually including the associated image of the goods identification failure) can be pushed to a plurality of cashiers for processing, and then the results between the plurality of cashiers can be checked, or referenced, or merged.
  • the pushing the job task in the task scheduling queue to the cash register node that meets the working condition according to the obtained network cashier list includes:
  • the identifying the difference information of the goods in the storage container based on the associated image comprises: determining difference information of the goods in the storage container according to the goods identification result of the at least two cashier nodes.
  • a user id can be pushed to multiple cashiers for processing, and the results between multiple cashiers can be checked and audited. This can further improve the accuracy of cargo identification.
  • the promotion prompt message may be sent to the client.
  • the prompt message may include multiple types of information, such as informing the user that the order cannot be generated, or returning the result of the goods identification failure to the user, and the promotion is unrecognizable.
  • the prompt message may also include reminding the user that the goods just placed cannot generate an order, please re-place the request, or directly identify the job task as an unidentified order of the goods.
  • the method may further include: if the cargo identification result of the cashier node includes a cargo identification failure, performing one of the following:
  • S691 The server returns a prompt message to the client, where the prompt message includes at least information content that cannot be generated or relocated by the goods order;
  • valve opening and closing of the channel device can be controlled based on a result of querying whether there is an unidentified goods order in the storage container when the storage container needs to pass through the passage device.
  • the channel device valve can be controlled to not open or close when the storage container is to exit through the outlet.
  • the passage may include a control passage valve used by a supermarket, a mall, or a subway to open and close the valve to control the passage of a control person or a storage container.
  • the user ID can be bound to the ID of the shopping cart used.
  • the self-checkout can identify the user ID or the ID of the shopping cart through the identification code scanning or infrared light near field communication.
  • the method for processing one or more goods orders can provide an order settlement method that is convenient for users to purchase and has a lower implementation cost of the merchant.
  • the shopping items can be monitored in real time.
  • the transformation (by image recognition processing) identifies and confirms the user's purchase list, and the server can remotely generate the user's total order in real time.
  • the order settlement is automatically performed.
  • the unmanned supermarket can be directly left, the data interaction node is less, and there is no need to configure a large number of expensive cash register equipment, light curtain and other sensing devices, and the implementation cost is lower and the order data processing is performed. More efficient and more stable.
  • Positioning devices can be used on the storage container where the client or client is located, such as GPS positioning, or UWB (Ultra Wideband, a carrierless communication technology that transmits data using non-sinusoidal narrow pulses of nanoseconds to picoseconds) It can be used for precise positioning and can be used for indoor positioning. Then, it is possible to set an identification mark of the area within a certain distribution range of the goods, and the goods in the area are uniformly attributed to one section, and each section can correspond to the cargo information in the section. Thus, when the goods enter different cargo attribution intervals, the goods selected by the user are usually goods distributed within the ownership zone of the goods.
  • the present embodiment combines the positioning of the storage container and the spatial partition distribution of the cargo to determine the cargo belonging interval in which the selected cargo is selected when the user selects the cargo, so that when the cargo cannot be directly identified by the associated image, the cargo can be
  • the identification within the cargo information range of the attribution interval can greatly reduce the cargo search interval, reduce the image recognition search interval, and improve the recognition speed and the recognition success rate while reducing the load of the recognition processing. Therefore, in another embodiment of the method provided by the present specification, the method may further include:
  • S14 The client acquires the location information of the storage container, and sends the location information to the server.
  • the identifying, by the server, the difference information of the goods in the storage container based on the associated image comprises: identifying the cargo information in the associated image within a range of the cargo attribution interval, and determining based on the identified cargo information The difference information of the change in the goods in the storage container.
  • the server described in the foregoing embodiment may include a single server in some application scenarios, or may be divided into different logical processing units.
  • it may include a module for receiving and storing images uploaded by the client, a module for image analysis and identification of goods, a module for performing manual task scheduling when the automatic identification of goods fails, and a plurality of logical terminals such as a network cashier management module, and the plurality of logics.
  • the terminal can be regarded as the server side in a unified manner, or can be independently used as a device connected or coupled with the server.
  • the client may include a photographing device and a communication device installed on a shopping cart, and other application scenarios may also use a photographing device, a communication device, a shopping cart, etc.
  • the positioning device acts as a client.
  • the device 1 may be used to detect whether the goods in the storage container have changed or not, and the image recognition and the data of the device 2 of the order, processing, etc., different processing sides/different processing stages are used to divide the client.
  • the server can specifically define various functional modules on the client, server, and server side according to the actual application environment.
  • FIG. 9 is a schematic diagram of an embodiment of a method for processing a goods order provided by the present specification.
  • a bracket can be installed on the shopping cart for placing a mobile phone or a webcam.
  • the server side can be divided into a number of different processing units. specific,
  • the shopping cart can be a general shopping cart of a shopping mall or a supermarket, and has various capacities such as 60 liters, 80 liters, 100 liters, and 125 liters.
  • the stand can be an external smart phone stand that can be attached to the front side of the shopping cart pusher. When a user makes a purchase, the item can be considered to be placed in a shopping cart.
  • the phone can include a high-definition rear camera and a screen that can be placed on the 101 stand.
  • the bracket and the phone can be adjusted to make the screen tilt toward the user, and the camera can capture the shopping cart space.
  • the camera application running in the mobile phone may include an image change detection module, and the difference image and video upload cloud server are recorded before and after the image change time, and the changed image area is uploaded to the cloud, so that the server side can automatically or manually analyze the image.
  • it can be set as a mobile phone and a user to bind one by one.
  • the mobile phone on the shopping cart can be bound with the ID (identification, identification) of the user.
  • the cloud server or the background service, and the mobile phone communicates with the mobile phone through the wifi or the carrier network, and can store the shopping cart pictures and videos taken by the camera.
  • the cloud server can be responsible for storing user orders and delivering analysis content to 105 and 103.
  • the video automatic analysis unit may include a barcode detection and recognition module and an alarm module.
  • Other embodiments may include a machine learning commodity detection module.
  • Machine learning detection can be performed based on methods such as SVM, RCNN, and the like.
  • the barcode detection recognition module inputs the difference image and searches for a regular barcode. If the identification is successful, the user's shopping order is added. If the barcode detection fails, the image and video information can be sent to the machine learning commodity detection module. In the machine learning commodity detection module, if a corresponding item is detected and a corresponding unique price is found, a user shopping order can be added. If the test fails, this image can be sent to the alarm module.
  • the alarm module can encapsulate this information into a job task and send it to the human task scheduling unit.
  • the job task may include information such as user id+alarm image or video content+current order corresponding to user id.
  • the human task scheduling unit may save a task queue, and each item of the task queue may include an order corresponding to the user id+alarm image or video content+user id. You can also save a list of network cashiers and a list of tasks for each network cashier.
  • the user id can be searched to confirm whether the network cashier has been bound. If it is already, it is judged whether the workload is saturated. If it is not saturated, it is preferentially pushed to the cashier. If the workload is saturated, it can be pushed to the new cashier for processing and the user id is bound to the new cashier id.
  • the user id is not processed by the network cashier at the current time, it is pushed to the new cashier for processing and the user id is bound to the new cashier id.
  • one user id may be pushed to multiple cashiers for processing, and the results between multiple cashiers may be checked and audited.
  • the network cashier manual control unit can communicate with the human task scheduling unit through an Ethernet or carrier network. You can set up a screen to display an image or video of an alarm. The network cashier can be a real operator. The network cashier can visually observe the image of the alarm, find the barcode area, and identify it. After the identification is successful, the user order is added. If the barcode recognition fails (such as occlusion or other reasons), you can find out whether there is a completely corresponding item from the store's goods list. If the search is successful, join the user order. If the search fails, the product can be marked and the result returned to 103. 103 The result can be fed back to the user screen 100, reminding the user that the product just placed cannot generate an order, please re-place it as required, or directly put the product into the user order. Unrecognized item.
  • the shopping cart of 101 can be installed with a UWB-like positioning method, which can help the customer to navigate and the passenger flow statistics, and can also know the product SKU range of the surrounding shelves by positioning (SKU, Stock Keeping Unit
  • SKU Stock Keeping Unit
  • the stock unit can be in units, boxes, trays, etc. SKUs are usually referred to as the abbreviation of the product uniform number, each product has a unique SKU number), reducing the image recognition search space and improving the commodity machine/manpower Identify the success rate.
  • one or more embodiments of the present specification provide a method for processing a goods order, which can be installed in a storage container such as a shopping basket or a shopping cart, and the remote server can be monitored in real time. Changes in the goods in the storage container to identify and confirm changes in the user's order, and can generate or update the user's shopping order.
  • the terminal equipment of the smart shopping cart and the server for the cargo identification and order processing can be used to realize the unmanned supermarket order processing order data, thereby saving a large number of self-service cash register devices and sensing devices.
  • the image data of the goods detection can be sent to the server through the client image or the video capturing device, and the server performs the processing of the goods identification, the order settlement, etc.
  • the overall hardware implementation is lower, and the data interaction can be reduced.
  • the implementation of the program is simpler and more convenient.
  • the implementation cost of the technical solution for automatic settlement of unmanned supermarket goods orders can be greatly reduced, which is easy to implement and has better stability.
  • the implementation environment and factors of unmanned supermarket area, passenger flow and consumer behavior characteristics are less affected. It is conducive to large-scale promotion and use.
  • the above embodiment describes an embodiment of one or more goods order processing methods provided by the present specification from a client and a server, or can be understood as a multi-party interaction between a terminal device on the client side and a terminal device on the server side.
  • the present specification further provides a cargo order processing method that can be used on the server side, and the server may include a separate server operating system, and may also include image storage, image analysis, job scheduling, and the like.
  • the present disclosure provides a method for processing a goods order that can be used on the server side.
  • FIG. 10 is a schematic flow chart of an embodiment of the method applicable to the server side provided by the present specification, as shown in FIG. Can include:
  • S100 Receive an associated image uploaded by the client, where the associated image includes an image that is detected by the client when the goods in the storage container change according to an image recognition manner;
  • S101 Identify difference information of the goods in the storage container based on the associated image
  • S104 Send the updated user order information to the client.
  • the associated image received by the server may include an image or video image taken by the client when the change in the goods in the storage container is detected (the video is considered a continuous image herein).
  • the client can detect and identify the image captured by the storage container when detecting whether the goods in the storage container change, for example, the image of the goods in the front and rear frames can be subtracted, and the absolute value of the image difference in the delineated area. If the threshold is exceeded, the goods in the storage container can be considered to have changed.
  • the detecting, by the image recognition manner, whether the goods in the storage container change may include:
  • S1001 capturing an image of the goods in the storage container by using a client camera device
  • the associated image may include an image determined in at least one of the following ways:
  • the identifying the difference information of the goods in the storage container based on the associated image may include:
  • S1011 Detect a cargo identification identifier in the associated image, and use the detected cargo identification identifier as the cargo difference information.
  • any one of the following embodiments may be employed:
  • S1012 After acquiring the associated image, using a cargo detection model obtained by a machine learning algorithm to detect cargo information in the associated image, and determining difference information of the cargo;
  • S1013 Detect a cargo identification identifier in the associated image, and use the detected cargo identification identifier as the cargo difference information;
  • the goods detection model obtained by the machine learning algorithm detects the goods information in the associated image, and determines the difference information of the goods.
  • the method may further include:
  • S1015 Push the job task in the task scheduling queue to the cashier node that meets the job condition according to the obtained network cashier list, where the network cashier list records the job status of the cashier node processing task.
  • the method may further include:
  • S1016 Query whether there is a network cashier bound to the job task in the network cashier list;
  • the corresponding job task is sent to the cashier node where the corresponding network cashier is located for processing.
  • the method may further include:
  • S1017 After determining a network cashier bound to the job task, querying a job status of the bound cash register node;
  • the job task is sent to a cashier node in the network cashier list where the job status is non-work volume saturation.
  • the pushing, by the obtained network cashier list, the job task in the task scheduling queue to the cash register node that meets the working condition comprises:
  • the identifying the difference information of the goods in the storage container based on the associated image comprises: determining difference information of the goods in the storage container according to the goods identification result of the at least two cashier nodes.
  • the method may further include:
  • the prompt message includes at least information content that the goods order cannot generate and relocate one of the goods;
  • the method may further include:
  • S10110 Receive positioning information of the storage container uploaded by the client, and determine, according to the positioning information, a cargo belonging interval in which the goods in the storage container change;
  • the identifying the difference information of the goods in the storage container based on the associated image comprises: identifying the cargo information in the associated image within a range of the cargo attribution interval, and determining the location based on the identified cargo information The difference information of the change of the goods in the storage container.
  • the present specification also provides a cargo order processing method that can be used on the client side, and the client can include a shooting and installation on, for example, a shopping cart.
  • the terminal device of the data communication capability the client may also include a module for image change detection, and the change of the goods in the storage container may be detected based on the image acquired by the camera.
  • one or more of the shopping cart, the cradle device, the display screen, or the locating device may also be included as a client together with the data communication device of the imaging device and the image transmission.
  • the method can be applied to the client side, and can detect the change of the goods in the shopping, and send the picture or video when the goods change to the server for identification and processing, and generate the order information of the user.
  • the present specification provides a method for processing a goods order that can be used on the client side.
  • FIG. 11 is a schematic flow chart of an embodiment of the method that can be used on the client side provided by the present specification, as shown in FIG. 11 . Show, can include:
  • S202 Receive updated user order information returned by the server.
  • the present specification further provides another method for processing a goods order that can be used on the client side.
  • the image recognition-based manner of detecting whether the goods in the storage container are changed may include:
  • the associated image may include a plurality of determinations.
  • the associated image may include an image determined in at least one of the following ways:
  • S2006 an image of the goods within a predetermined time range of the image of the goods corresponding to the change in the time of the goods when it is determined that the goods in the storage container have changed.
  • Figure 12 is a flow diagram of another embodiment of the method provided by the present specification that can be used on the client side.
  • the method in the process of uploading the associated image by the client, only the image content of the goods image transmission change may be sent to the server for processing, which can save network bandwidth browsing and improve processing speed.
  • the method after acquiring the associated image, the method may further include:
  • the sending the associated image to the server may include: transmitting the identified image in the cargo boundary to the server.
  • the method may further include:
  • S2008 Acquire location information of the storage container, and send the positioning information to the server, so that the server determines a cargo attribution interval of the goods in the storage container based on the positioning information.
  • the cargo order processing method provided by the present specification can be used for consumer self-service shopping and platform automatic identification of unmanned supermarket types.
  • a photographing device can be installed in a storage container such as a shopping basket or a shopping cart, and the remote server can identify and confirm the change of the user's order by monitoring the change of the goods in the storage container in real time, and can generate Or update the user's purchase order.
  • the terminal equipment of the smart shopping cart and the server for the cargo identification and order processing can be used to realize the unmanned supermarket order processing order data, thereby saving a large number of self-service cash register devices and sensing devices.
  • the image data of the goods detection can be sent to the server through the client image or the video capturing device, and the server performs the processing of the goods identification, the order settlement, etc.
  • the overall hardware implementation is lower, and the data interaction can be reduced.
  • the implementation of the program is simpler and more convenient.
  • the implementation cost of the technical solution for automatic settlement of unmanned supermarket goods orders can be greatly reduced, which is easy to implement and has better stability.
  • the implementation environment and factors of unmanned supermarket area, passenger flow and consumer behavior characteristics are less affected. It is conducive to large-scale promotion and use.
  • one or more embodiments of the present specification further provide a goods order processing apparatus.
  • the apparatus may include a system (including a distributed system), software (applications), modules, components, servers, clients, etc., using the methods described in the embodiments of the present specification in conjunction with necessary implementation hardware.
  • the apparatus in one or more embodiments provided by the embodiments of the present specification is as described in the following embodiments.
  • the term "unit” or "module” may implement a combination of software and/or hardware of a predetermined function.
  • FIG. 13 is a schematic diagram of a module structure of an embodiment of a cargo order processing apparatus that can be used on the server side provided by the present specification. As shown in FIG. 13, the apparatus may include:
  • the image receiving module 101 may be configured to receive an associated image uploaded by the client, where the associated image includes an image that is detected by the client when the goods in the storage container change according to an image recognition manner;
  • the image recognition module 102 is configured to identify difference information of the goods in the storage container based on the associated image
  • the order generating module 103 can be configured to generate corresponding goods order change information by using the difference information
  • the order update module 104 may be configured to update user order information corresponding to the client according to the goods order change information;
  • the information feedback module 105 can be configured to send the updated user order information to the client.
  • the detecting, by the image recognition, whether the goods in the storage container have changed may include:
  • the difference data is greater than a preset change threshold, it is determined that the goods in the storage container have changed.
  • the associated image may comprise an image determined in at least one of the following ways:
  • FIG. 14 is a block diagram showing the structure of an embodiment of an image recognition module in the apparatus provided by the present specification.
  • the image recognition module 102 can include:
  • the identification code detecting module 1021 may be configured to detect a cargo identification identifier in the associated image, and use the detected cargo identification identifier as the cargo difference information.
  • the image recognition module 102 in one embodiment of the apparatus may include:
  • the cargo detecting module 1022 includes a cargo detecting model obtained after the sample training based on the selected machine learning algorithm, and may be used to acquire the associated image, detect the cargo information in the associated image, and determine the cargo.
  • the difference information may be used to detect the cargo information in the associated image when the identification of the identification of the goods in the associated image fails, and determine the difference information of the goods.
  • Figure 15 is a block diagram showing the structure of another embodiment of the apparatus provided in the present specification. As shown in FIG. 15, the apparatus may further include:
  • the manual scheduling module 106 may be configured to detect the identification of the cargo identification identifier in the associated image or the detection of the cargo detection model, generate a job task that fails the associated image detection, and place the job task in the task scheduling queue; as well as,
  • the job task in the task scheduling queue is pushed to the cashier node in accordance with the job condition according to the obtained network cashier list, and the network cashier list records the job status of the cashier node processing task.
  • Figure 16 is a block diagram showing the structure of another embodiment of the apparatus provided in the present specification. As shown in FIG. 16, in another embodiment of the device, the device may further include:
  • the cashier control module 107 can be configured to query whether there is a network cashier bound to the job task in the network cashier list;
  • the corresponding job task is sent to the cashier node where the corresponding network cashier is located for processing.
  • the cashier control module 107 may also be used to query the job status of the bound cashier node;
  • the job task is sent to a cashier node in the network cashier list where the job status is non-work volume saturation.
  • the manual scheduling module 106 according to the obtained network cashier list, pushing the job task in the task scheduling queue to the cash register node that meets the working condition includes:
  • the image recognition module 102 based on the associated image, identifying difference information of the goods in the storage container, comprising: determining difference information of goods in the storage container according to the cargo identification result of the at least two cashier nodes .
  • the information feedback module 105 may perform one of the following when the cargo identification result of the cashier node includes a cargo identification failure:
  • the prompt message includes at least information content that the goods order cannot generate and relocate one of the goods;
  • Figure 17 is a block diagram showing the structure of another embodiment of the apparatus provided in the present specification. As shown in FIG. 17, in another embodiment of the device, the device may further include:
  • the positioning processing module 108 is configured to receive positioning information of the storage container uploaded by the client, and determine, according to the positioning information, a cargo belonging interval in which the goods in the storage container change;
  • the image recognition module 102 identifying the difference information of the goods in the storage container based on the associated image comprises: identifying the cargo information in the associated image within the range of the cargo attribution interval, based on the identified The cargo information determines the difference information of the changes in the goods in the storage container.
  • the foregoing device may further include other implementation manners according to the embodiment of the client-server interaction or the server-side method embodiment.
  • the specific implementation manner may refer to the description of the related method embodiment. Do not repeat them one by one.
  • FIG. 18 is a block diagram showing an embodiment of a cargo order processing apparatus according to an embodiment of the present invention. As shown in FIG. 18, the apparatus may include:
  • the cargo change detecting module 201 can be configured to detect whether the goods in the storage container change based on an image recognition manner
  • the image sending module 202 is configured to send an associated image that changes the goods in the storage container to the server when the change of the goods is detected;
  • the information receiving module 203 is configured to receive updated user order information returned by the server;
  • the display module 204 can be configured to display the updated user order information.
  • FIG. 19 is a block diagram showing an embodiment of a cargo change detecting module in the device of the present specification.
  • the cargo change detecting module 201 may include:
  • the photographing unit 2011 can be used to capture an image of the goods in the storage container
  • the image difference processing unit 2012 may be configured to detect difference data between the images of the goods in chronological order.
  • the goods change determining module 2013 may be configured to determine that the goods in the storage container change when the difference data is greater than a preset change threshold.
  • the associated image in the image sending module 202 may include an image determined by at least one of the following methods:
  • FIG 20 is a block diagram showing another embodiment of the cargo change detecting module in the device of the present specification.
  • the cargo change detecting module 201 may include:
  • a change boundary identification module 2014 which can be used to identify a cargo boundary in which the goods in the associated image change
  • the sending, by the image sending module 202, the associated image to the server may include: sending the identified image in the cargo boundary to the server.
  • Figure 21 is a block diagram showing the structure of another embodiment of the device in the present specification. As shown in Figure 21, the device may further include:
  • the location locating module 205 may be configured to acquire positioning information of the storage container, and send the positioning information to the server, so that the server determines, according to the positioning information, attribution of goods in the storage container. Interval.
  • the foregoing device may further include other implementation manners according to the embodiment of the client-server interaction or the client-side method embodiment.
  • the specific implementation manner may refer to the description of the related method embodiment. This will not be repeated.
  • a cargo order processing device provided by one or more embodiments of the present specification may be installed with a photographing device installed in a storage container such as a shopping basket or a shopping cart, and the remote server may identify and change the load in the storage container in real time. Confirm changes to the user's order and can generate or update the user's purchase order.
  • the terminal equipment of the smart shopping cart and the server for the cargo identification and order processing can be used to realize the unmanned supermarket order processing order data, thereby saving a large number of self-service cash register devices and sensing devices.
  • the image data of the goods detection can be sent to the server through the client image or the video capturing device, and the server performs the processing of the goods identification, the order settlement, etc.
  • the overall hardware implementation is lower, and the data interaction can be reduced.
  • the implementation of the program is simpler and more convenient.
  • the implementation cost of the technical solution for automatic settlement of unmanned supermarket goods orders can be greatly reduced, which is easy to implement and has better stability.
  • the implementation environment and factors of unmanned supermarket area, passenger flow and consumer behavior characteristics are less affected. It is conducive to large-scale promotion and use.
  • the above-mentioned goods order processing method or apparatus provided by the embodiments of the present specification may be implemented by a processor executing a corresponding program instruction in a computer, such as using a C++ language of a Windows operating system on a PC side, a Linux system implementation, or the like, for example, using android.
  • the iOS system programming language is implemented in intelligent terminals, and the processing logic based on quantum computers is implemented.
  • the device can be used in various servers of a plurality of order settlement system platforms, a self-service shopping order processing system, a shopping cloud service system, etc., to realize low-cost, fast, efficient, and reliable order processing for the user's shopping items.
  • the present specification provides a server, as shown in FIG. 22, that can include at least one processor and a memory for storing processor-executable instructions that, when executed by the processor, implement:
  • the updated user order information is sent to the client.
  • server may further include other implementation manners according to the description of the method or the device embodiment.
  • specific implementation manner reference may be made to the description of the related method embodiments, and details are not described herein.
  • the server may include a separate server, or may include a server system including multiple servers, such as a cloud server storing user orders, a server for manual job scheduling, a server operated by a network cashier, and automatic analysis of video or images.
  • a server, etc. can also be a framework of a distributed server or server cluster.
  • the foregoing description of the device or the electronic device in the present specification may further include other embodiments according to the description of the related method embodiments.
  • the various embodiments in the specification are described in a progressive manner, and the same or similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments.
  • the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment.
  • the present specification is based on the foregoing embodiment of the client side method or apparatus, and may further provide a shopping terminal device, which may include the aforementioned mobile phone having a function of photographing and image data transmission, or may further include Terminal equipment such as shopping carts, positioning equipment, etc.
  • the client may include an image capturing and detecting unit, which may be implemented by a processor executing a corresponding program instruction, such as implementing a C++ language of a Windows operating system on a PC side, a Linux system implementation, or other using, for example, an android or iOS system.
  • the programming language is implemented in intelligent terminals, as well as processing logic based on quantum computers.
  • the shopping terminal device can capture a storage space image in a storage container such as a shopping cart, and can detect a change of the goods when the consumer puts in or takes out the goods, and uploads the captured image to the server so that the server is in the image.
  • the goods are identified to generate or process the user's order information, enabling low-cost, fast, efficient, and reliable order processing of the user's shopping items.
  • the present specification provides a client that can include at least one processor and a memory for storing processor-executable instructions that are implemented when the processor executes the instructions:
  • the above-mentioned client may further include other implementation manners according to the description of the method or the device embodiment.
  • specific implementation manner reference may be made to the description of the related method embodiments, and details are not described herein.
  • the present specification also provides one or more goods order processing systems based on the description of the method or apparatus or client or server described above.
  • 23 is a schematic structural diagram of an embodiment of the system provided by the present specification, where the system may include a shopping terminal 1 and a server 2,
  • the shopping terminal 2 may include the apparatus described in the embodiment of the shopping terminal side described in any one of the above embodiments or the method of implementing the method described on the side of any one of the shopping terminals.
  • a cargo order processing method, device, server, shopping terminal and system provided by one or more embodiments of the present specification can install a shooting device in a storage container such as a shopping basket or a shopping cart, and the remote server can monitor the storage in real time.
  • the change in the goods in the container identifies and confirms changes in the user's order and can generate or update the user's purchase order.
  • the terminal equipment of the smart shopping cart and the server for the cargo identification and order processing can be used to realize the unmanned supermarket order processing order data, thereby saving a large number of self-service cash register devices and sensing devices.
  • the image data of the goods detection can be sent to the server through the client image or the video capturing device, and the server performs the processing of the goods identification, the order settlement, etc.
  • the overall hardware implementation is lower, and the data interaction can be reduced.
  • the implementation of the program is simpler and more convenient.
  • the implementation cost of the technical solution for automatic settlement of unmanned supermarket goods orders can be greatly reduced, which is easy to implement and has better stability.
  • the implementation environment and factors of unmanned supermarket area, passenger flow and consumer behavior characteristics are less affected. It is conducive to large-scale promotion and use.
  • embodiments of the present specification are not limited to having to comply with industry communication standards, machine learning models, standard computer data processing and data storage rules, or one or more of these specifications. The case described in the embodiment. Certain industry standards, machine learning models, or implementations that have been modified in a manner that uses a custom approach or an embodiment described above may also achieve the same, equivalent, or similar, or post-deformation implementation effects of the above-described embodiments. Embodiments obtained by applying these modified or modified data acquisition, storage, judgment, processing methods, etc., may still fall within the scope of alternative embodiments of the embodiments of the present specification.
  • PLD Programmable Logic Device
  • FPGA Field Programmable Gate Array
  • HDL Hardware Description Language
  • the controller can be implemented in any suitable manner, for example, the controller can take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (eg, software or firmware) executable by the (micro)processor.
  • computer readable program code eg, software or firmware
  • examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, The Microchip PIC18F26K20 and the Silicone Labs C8051F320, the memory controller can also be implemented as part of the memory's control logic.
  • the controller can be logically programmed by means of logic gates, switches, ASICs, programmable logic controllers, and embedding.
  • a controller can be considered a hardware component, and the means included in the alignment for implementing various functions can also be considered as structures within the hardware components.
  • a device for implementing various functions can be considered as a software module that can be both a method of implementation and a structure within a hardware component.
  • the system, device, module or unit illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product having a certain function.
  • a typical implementation device is a computer.
  • the computer can be, for example, a personal computer, a laptop computer, a car-mounted human-machine interaction device, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet.
  • each module may be implemented in the same software or software and/or hardware when implementing one or more of the specification, or the modules implementing the same function may be implemented by a plurality of sub-modules or a combination of sub-units, etc. .
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or integrated. Go to another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the controller can be logically programmed by means of logic gates, switches, ASICs, programmable logic controllers, and embedding.
  • a controller can be considered a hardware component, and the means included within the alignment for implementing various functions can also be considered as structures within the hardware components.
  • a device for implementing various functions can be considered as a software module that can be both a method of implementation and a structure within a hardware component.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.
  • a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
  • processors CPUs
  • input/output interfaces network interfaces
  • memory volatile and non-volatile memory
  • the memory may include non-persistent memory, random access memory (RAM), and/or non-volatile memory in a computer readable medium, such as read only memory (ROM) or flash memory.
  • RAM random access memory
  • ROM read only memory
  • Memory is an example of a computer readable medium.
  • Computer readable media includes both permanent and non-persistent, removable and non-removable media.
  • Information storage can be implemented by any method or technology.
  • the information can be computer readable instructions, data structures, modules of programs, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory.
  • PRAM phase change memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • RAM random access memory
  • ROM read only memory
  • EEPROM electrically erasable programmable read only memory
  • flash memory or other memory technology
  • compact disk read only memory CD-ROM
  • DVD digital versatile disk
  • Magnetic cassette tape magnetic tape storage
  • graphene storage or other magnetic storage devices or any other non-transportable media
  • computer readable media does not include temporary storage of computer readable media, such as modulated data signals and carrier waves.
  • one or more embodiments of the present specification can be provided as a method, system, or computer program product.
  • one or more embodiments of the present specification can take the form of an entirely hardware embodiment, an entirely software embodiment or a combination of software and hardware.
  • one or more embodiments of the present specification can employ a computer program embodied on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer usable program code embodied therein. The form of the product.
  • One or more embodiments of the present specification can be described in the general context of computer-executable instructions executed by a computer, such as a program module.
  • program modules include routines, programs, objects, components, data structures, and the like that perform particular tasks or implement particular abstract data types.
  • One or more embodiments of the present specification can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are connected through a communication network.
  • program modules can be located in both local and remote computer storage media including storage devices.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Computational Linguistics (AREA)
  • Multimedia (AREA)
  • Acoustics & Sound (AREA)
  • Human Computer Interaction (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Security & Cryptography (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Human Resources & Organizations (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Information Transfer Between Computers (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Image Analysis (AREA)
  • Cash Registers Or Receiving Machines (AREA)

Abstract

本说明书实施例公开了一种货物订单处理方法、装置、服务器、购物终端及系统。所述方法包括:客户端基于图像识别的方式检测储物容器中的货物是否发生变化;若有变化,则获取所述储物容器中货物发生变化的关联图像,发送至服务器;服务器基于所述关联图像识别所述储物容器中货物的差异信息,生成相应的货物订单变更信息;基于所述货物订单变更信息更新所述客户端对应的订单信息;所述服务器将更新后的订单信息发送给所述客户端;所述客户端展示所述更新后的订单信息,消费者可以查看已放入购物车中的购物清单。

Description

一种货物订单处理方法、装置、服务器、购物终端及系统 技术领域
本说明书实施例属于计算机数据处理技术领域,尤其涉及一种货物订单处理方法、装置、服务器、购物终端及系统。
背景技术
计算机和互联网技术的迅速发展,带动了许多新兴产业的快速崛起,改变和方便了许多消费者的商品购物方式。无人超市是目前各大商品购物平台运营商重点挖掘的一个市场开拓热点,其中就涉及到如何便捷、可靠的对用户的订单数据进行处理的技术问题。
目前无人超市中可以采用一种自助订单结算方案,主要是通过在商场货架上、过道等地方布局多个传感设备,例如摄像头、压力传感器、红外传感器、无线网络等设备,监测用户从货架上拿取的货物,然后结合自助扫码支付、购物车称重等设备完成订单结算。这种方式需要多种类型的辅助设备,并且需要在商场多处布局大量装置,业务实现上需要较高的硬件成本。同时,该方式需要各个传感设备的数据相互配合来统计货物,稳定性难以保障,不利于大规模推广使用。因此,目前在无人超市中还需要一种成本更低、简单快捷、可靠的订单处理方案。
发明内容
本说明书实施例目的在于提供一种货物订单处理方法、装置、服务器、购物终端及系统,可以提供一种便简单、可靠、实施成本更低的订单结算方式,可以大大降低无人超市订单自动处理的方案实施成本和订单数据处理的可靠性。
本说明书实施例提供的一种货物订单处理方法、装置、服务器、购物终端及系统是包括如下的方式实现的:
一种货物订单处理方法,所述方法包括:
客户端基于图像识别的方式检测储物容器中的货物是否发生变化;
若检测到货物发生变化,则所述客户端将所述储物容器中货物发生变化的关联图像发送至服务器;
所述服务器基于所述关联图像识别所述储物容器中货物的差异信息;
所述服务器利用所述差异信息生成相应的货物订单变更信息;
所述服务器根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
所述服务器将更新后的用户订单信息发送给所述客户端;
所述客户端展示所述更新后的用户订单信息。
一种货物订单处理方法,所述方法包括:
接收客户端上传的关联图像,所述关联图像包括客户端基于图像识别的方式检测到储物容器中货物发生变化时获取的图像;
基于所述关联图像识别所述储物容器中货物的差异信息;
利用所述差异信息生成相应的货物订单变更信息;
根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
将更新后的用户订单信息发送给所述客户端。
一种货物订单处理方法,所述方法包括:
基于图像识别的方式检测储物容器中的货物是否发生变化;
若检测到货物发生变化,则将所述储物容器中货物发生变化的关联图像发送至服务器;
接收所述服务器返回的更新后的用户订单信息;
展示所述更新后的用户订单信息。
一种货物订单处理装置,所述装置包括:
图像接收模块,用于接收客户端上传的关联图像,所述关联图像包括客户端基于图像识别的方式检测到储物容器中货物发生变化时获取的图像;
图像识别模块,用于基于所述关联图像识别所述储物容器中货物的差异信息;
订单生成模块,用于利用所述差异信息生成相应的货物订单变更信息;
订单更新模块,用于根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
信息反馈模块,用于将更新后的用户订单信息发送给所述客户端。
一种货物订单处理装置,所述装置包括:
货物变化检测模块,用于基于图像识别的方式检测储物容器中的货物是否发生变化;
图像发送模块,用于在检测到货物发生变化时,将所述储物容器中货物发生变化的关联图像发送至服务器;
信息接收模块,用于接收所述服务器返回的更新后的用户订单信息;
显示模块,用于展示所述更新后的用户订单信息。
一种服务器,包括至少一个处理器以及用于存储处理器可执行指令的存储器,所述处理器执行所述指令时实现:
接收客户端上传的关联图像,所述关联图像包括客户端基于图像识别的方式检测到储物容器中货物发生变化时获取的图像;
基于所述关联图像识别所述储物容器中货物的差异信息;
利用所述差异信息生成相应的货物订单变更信息;
根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
将更新后的用户订单信息发送给所述客户端。
一种购物终端,包括至少一个处理器以及用于存储处理器可执行指令的存储器,所述处理器执行所述指令时实现:
基于图像识别的方式检测储物容器中的货物是否发生变化;
若检测到货物发生变化,则将所述储物容器中货物发生变化的关联图像发送至服务器;
接收所述服务器返回的更新后的用户订单信息;
展示所述更新后的用户订单信息。
一种货物订单处理系统,包括:购物终端、服务器,
所述服务器包括本说明书实施例中任意一个所述的装置,
所述购物终端包括本说明书实施例中任意一个所述的购物终端;
或者,
所述服务器实现本说明书实施例中任意一个所述的方法的步骤,
所述购物终端实现本说明书实施例中任意一个所述的方法的步骤。
本说明书一个或多个实施例提供的一种货物订单处理方法、装置、服务器、购物终端及系统,可以在购物篮、购物车等储物容器中安装拍摄设备,远程服务器可以通过实时监控储物容器中货物的变化来识别和确认用户订单的变化,并可以生成或更新该用户的购物订单。这样,利用智能购物车等终端设备结合货物识别、订单处理的服务器来可以实现无人超市订单处理订单数据,节省大量自助收银设备、传感设备等。利用本说明书提供的实施方案,可以通过客户端图像或视频拍摄设备将货物检测的图像数据发送给服务器,由服务器进行货物识别、订单结算等处理,总体硬件实施成功更低,并且可以减少数据交互节点,方案实施更加简单便捷。在实际应用中可以大大降低无人超市货物订单自动结算处理的技术方案实施成本,便于实施,稳定性更好,无人超市面积、客流量、消费者行为特征等实施环境、因素的影响较小,利于大规模推广使用。
附图说明
为了更清楚地说明本说明书实施例或现有技术中的技术方案,下面将对实施例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,下面描述中的附图仅仅是说明书中记载的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1是本说明书提供的所述一种货物订单处理方法实施例的流程示意图;
图2是本说明书中一个客户端拍摄购物车货物的实施场景示意图;
图3是本说明书中一个客户端拍摄购物车中新加货物的实施场景示意图;
图4是本说明书中一个识别货物发生变化的货物边界示意图;
图5是本说明书一个实施场景中识别出的货物边界中的图像;
图6是在图3所示的购物车中的第一件货物上叠放另一件货物的场景示意图;
图7是根据图像差别识别出的图6中新放入货物的货物边界中的图像;
图8是本说明书提供的所述方法的另一个实施例的流程示意图;
图9是本说明书提供的一种货物订单处理方法的实施例场景示意图;
图10是本说明书提供的可用于服务器一侧的所述方法一个的实施例流程示意图;
图11是本说明书提供的可以用于客户端一侧所述方法的一个实施例流程示意图;
图12是本说明书提供的可以用于客户端一侧所述方法的另一个实施例流程示意图;
图13是本说明书提供的一种可以用于服务器一侧的货物订单处理装置实施例的模块结构示意图;
图14是本说明书提供的所述装置中图像识别模块一个实施例的模块结构示意图;
图15是本说明书提供的所述装置中另一个实施例的模块结构示意图;
图16是本说明书提供的所述装置中另一个实施例的模块结构示意图;
图17是本说明书提供的所述装置中另一个实施例的模块结构示意图;
图18是本说明书一个的一种货物订单处理装置的实施例模块结构示意图;
图19是本说明书所述装置中货物变化检测模块一种实施例的模块结构示意图;
图20是本说明书所述装置中货物变化检测模块另一种实施例的模块结构示 意图;
图21是本说明书所述装置另一种实施例的模块结构示意图;
图22是本说明书提供的结合计算机硬件实现的所述装置一种实施例的结构示意图;
图23是本说明书提供的所述系统一种实施例的架构示意图。
具体实施方式
为了使本技术领域的人员更好地理解本说明书中的技术方案,下面将结合本说明书一个或多个实施例中的附图,对本说明书一个或多个实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是说明书一部分实施例,而不是全部的实施例。基于说明书一个或多个实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都应当属于本说明书实施例方案保护的范围。
本说明书提供的实施方案可以用于包括但不限于无人超市的应用场景,可以采用C/S(Client/Server,客户端/服务器)的系统架构。一般的,可以在无人超市中提供的手提购物篮、手拉购物篮、购物车、货物搬运车等多种类型的装载货物的设备(为便于统一描述,在本说明书一些实施例中可以统称为储物容器)中安装智能手机、摄像仪、监控器、相机等具有照相或录像功能的设备。这些照相或录像设备中可以含有图像变化检测逻辑,可以基于拍照图像或视频图像检测储物容器中货物的变化情况。客户端可以将货物变化的图像发送给服务器一侧进行处理,服务器一侧则可以根据图像识别用户新增加到储物容器中的货物或从储物容器中拿走的货物,并相应的生成货物变更信息。进一步的,可以基于储物容器中货物变更的信息相应的生成或更新用户的总订单信息。这些更新后的货物的总订单信息可以即时展示给客户端显示设备上,供用户查看、确认等。
需要说明的是,上述所述的C/S可以是一种逻辑上划分的客户端/服务器架 构,即在本说明书的一些实施例中,所述的客户端可以包括采集储物容器中货物图像的终端设备,所述的服务器可以包括获取所述客户端采集的图像并基于该图像进行货物识别、订单处理的终端设备。在一些实施产品中,本说明书的实施方案也可以用于所述的客户端和服务器为同一物理设备的实施场景中,例如智能购物车上的设备可以既包括图像变化检测模块,也可以包括图像中货物识别模块和/或订单数据处理模块,每个智能购物车均可以进行自助货物变化检测、货物识别、订单处理等,所有智能购物车的订单数据可以汇总到指定的订单业务服务器中。
下面以一个无人超市中对用户购物车中的货物进行订单处理的应用场景对本说明书的一个或多个实施例方案进行详细说明。需要说明的是,本说明书实施方案不限于无人超市场景的实施应用,在其他的实施场景中,例如货物搬运车、集装箱、货车车厢、快递包装箱等识别储物容器中货物变化并进行相应货物订单数据处理的实施场景同样可以使用本说明书实施例所提供的实施方案。本实施例中所述的无人超市可以包括无或少数现场对正常购物车中货物进行解释的收银人员,所述的超市可以包括便利店、大型商场、自助服务站等,用户可以在超市中使用购物车采集购买的商品货物,购物车上的拍照或摄像设备可以检测到购物车中货物的变化,然后可以由服务器识别货物,生成相应的订单。用户购物完成后,可以实现自动、更加快速、更加可靠稳定的生成用户的购物订单。用户可以自行进行订单结算,如使用支付平台在线结算,在订单结算成功后,可以直接通过出口或通道,免去商品结算的排队,可以大幅提高用户购物体验。
图1是本说明书提供的所述一种货物订单处理方法实施例的流程示意图。虽然本说明书提供了如下述实施例或附图所示的方法操作步骤或装置结构,但基于常规或者无需创造性的劳动在所述方法或装置中可以包括更多或者部分合并后更少的操作步骤或模块单元。在逻辑性上不存在必要因果关系的步骤或结构中,这些步骤的执行顺序或装置的模块结构不限于本说明书实施例或附图所 示的执行顺序或模块结构。所述的方法或模块结构的在实际中的装置、服务器或终端产品应用时,可以按照实施例或者附图所示的方法或模块结构进行顺序执行或者并行执行(例如并行处理器或者多线程处理的环境、甚至包括分布式处理、服务器集群的实施环境)。具体的一个实施例如图1所示,本说明书提供的一种货物订单处理方法的一种实施例中,所述方法可以包括:
S0:客户端基于图像识别的方式检测储物容器中的货物是否发生变化。
本实施例中所述的客户端可以包括安装在所述购物车上的具有图像变化检测功能的电子设备,如智能手机、摄像机、监控仪等。所述的客户端还可以具有通信能力,可以将拍摄的图像或视频发送给服务器。一种实施方式中,所述的客户端可以为安装在储物容器上进行图像拍摄的终端设备以及图像变化检测的终端设备等的单独装置,如客户端可以包括可以进行拍摄和视频上传的智能手机以及检测图像变化的专用设备。其他的实施方式中,所述的客户端也可以包括由图像拍摄装置、图像变化检测装置、通信模块等多个设备组成的一个终端设备的,甚至可以连同购物车等储物容器作为一个整体作为所述的客户端。所述的储物容器可以包括多种类型的用于存储用户购物货物的储物装置,如手提购物篮、手推购物车等。
具体的在本实施场景的一个示例中,如图2所示,图2是本说明书中一个客户端拍摄购物车货物的实施场景示意图。所述的储物容器可以为超市中使用的购物车,如常规的60升、80升、100升等的各种容器的购物车。本示例中可以使用智能手机作为所述的客户端,所述的智能手机可以包括高清摄像头和屏幕。可以在购物车靠近用户扶手或把持部分的支架处设置智能手机支架,可以用于放置多种尺寸类型的智能手机。所述的智能手机可以放置在支架处,使得智能手机的摄像头朝向购物车,可以拍摄到购物车空间。所述智能手机的屏幕朝向用户,所述屏幕上可以展示已生产的用户订单信息(用户放入购物车的货物)或者展示购物提示、出错、导航等其他信息。
用户在购物时,可以认为是将货物一件件放置到购物车中的。在用户确定 使用某辆购物车时,可以认为该购物上的客户端(智能手机)与该用户绑定,例如用户通过扫描解锁购物车A,则购物车A通过验证后与该用户的ID一对一的绑定。此时,购物车上的智能手机可以运行摄像应用。所述智能手机中可以包括图像变化检测模块,如设置在拍摄应用中的图像变化检测算法或计算模型,可以实现在图像变化时刻的前后录制差异的图像或包含该差役的图像的视频。一种实施方式中,所述智能手机可以自启动拍摄应用后可以一直对购物车的购物空间进行预定频率的拍照,或者一直进行拍摄。同时可以对拍摄获得的图像进行检测,通过图像识别的方式判断购物车中的货物是否发生变化,如用户放入新的货物或者放回货物。所述的图像识别的方式可以包括获取图像直接的差异部分,通过识别差异部分来确定图像的变化。具体实现时,可以根据前后两帧相减法,在划定区域内图像差绝对值超过一定阈值,就以一定时间间隔发送新更新的图像帧到服务器。如果图像保持静止不变,则可以认为货物没有发生变化,可以不用发送。
因此,本说明书提供的所述方法的一个实施例中,所述基于图像识别的方式检测储物容器中的货物是否发生变化可以包括:
S02:通过摄像设备拍摄所述储物容器中的货物图像;
S04:按时间顺序检测所述货物图像之间差异数据;
S06:若所述差异数据大于预设的变化阈值,则确定所述储物容器中的货物发生变化。
所述的摄像设备可以包括前述所述的摄像头,也可以包括相机、摄像仪、监控仪等。在获取储物容器中的货物图像后,可以按照帧的时间顺序检测货物图像,确认货物储物容器中货物是否发生变化。一般的,储物容器如购物车中放置新的货物或拿掉货物时,摄像设备拍摄到的储物容器中的空间图像会发生变化,如图3所示,图3是本说明书中一个客户端拍摄购物车中新加货物的实施场景示意图。客户端可以检测该图像变化,并可以获取图像变化的差异数据。在本实施例中,如果图像变化的差异数据大于预设的变化阈值,则可以确定所 述储物容器中的货物发生了变化,这样可以避免一些非货物本身放入或拿出引起的图像变化干扰信息,例如购物车晃动导致的购物车中的货物发生少量位移。当然,具体的判断差异数据的变化阈值的取值范围可以根据经验或实验进行设置。在图2和图3所示的购物车从全空到放入一件商品的示意图中,为了图像检测方便,可以在购物车四周及底面放置遮档物,这样,在摄像设备拍摄的区域内有货物放入时会引起拍摄的货物图像的明显变化,并可以减少拍摄的购物车货物空间之外的干扰信息,如道路、货架、人影等。
需要说明的是,客户端对储物容器中的货物进行拍摄可以包括预定频率的拍照获取的图像数据,也可以包括录制视频货物的图像数据。在本说明书的一个或多个实施例中,视频可以视为连续的多张图像的集合,也可以作为一种图像数据。因此,本说明书中所述的货物图像、关联图像等图像数据可以包括单独拍照获取的图像,也可以包括视频中的图像。
S2:若检测到货物发生变化,则所述客户端将所述储物容器中货物发生变化的关联图像发送至服务器。
如果客户端检测到购物车中的货物发生了变化,如在购物车中新放入一件货物,则客户端可以检测到所拍摄的图像发生了变化。当基于拍摄的货物图像确认购物车中的货物发生了变化,则客户端可以获取所述购物车中货物发生变化的关联图像,并可以将该关联图像发送至预定的服务器,如超市订单处理的云服务器进行处理。
所述的关联图像可以包括客户端发送给服务器的进行检测和识别货物变化的图像,可以为单张图像,也可以为多张图像。一种实施情况下,所述的关联图像可以包括客户端一侧拍摄的单独的非连续图像,也可以包括客户端录制的检测到货物发生变化的视频数据,此时的视频数据可以认为为连续的关联图像。在确定关联图像时,可以采用一旦客户端确定储物容器中的货物发生变化则将检测到的图像变化时刻前后存在差异的图像或一段视频作为所述关联图像。当然,其他的实施方式也可以将图像变化时刻前后一段时间内(如3秒)的图像 作为所述关联图像。本说明书提供的一个或多个实施例中,所述关联图像可以包括下述中的至少一种:
S20:大于所述变化阈值的差异数据对应的货物图像;
S22:当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定窗口范围内的货物图像;
S24:当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定时间范围内的货物图像。
具体的应用场景中,例如当检测到T0时刻和T1时刻货物图像的差异数据data0大于预设的变化阈值时,可以将该差异数据data0对应的T0时刻和T1时刻的货物图像P0和P1作为此次检测到的储物容器中货物发生变化的关联图像。或者,在T2时刻检测到货物发生变化时,选取T2时刻对应的货物图像P2以及其前后10张货物图像(此时的窗口大小为11张图像)作为发生给服务器的关联图像。或者,选取图像变化时刻T3前后3秒内的拍照视频录制的图像数据作为所述关联图像。本说明书提供的实施例中,所述的关联图像的确定方式可以包括上述所述方式的一种或者多种的组成,例如可以选取所述差异数据对应的货物图像的同时选取变化时刻前后10张图像(重叠图像可以不需要重复选择)一同作为所述关联图像。
在本实施例应用场景中,可以在检测货物图像是否发生变化的处理中通过图像相减后的差异数据确定购物车中的货物是否发生变化。如4所示,图4是本说明书中一个识别货物发生变化的货物边界示意图,L是识别出的披萨包装盒的边界框,可以通过图像相减以及阈值联通区域处理,找到如图4中L所示的购物车中新加入的货物的图像边界。本说明书提供的所述方法的另一个实施方式中,客户端可以检测货物发生变化的图像中的货物边界,然后在上传关联图像到服务器进行图像中货物识别、订单处理等时,可以上传识别出的货物边界中的图像,可以减少图像的数据传输,节省网络流量和降低服务器负载。因此,本说明书所述方法的另一个实施例中,所述方法还可以包括:
S3:所述客户端获取所述关联图像后识别所述关联图像中货物发生变化的货物边界;
相应的,所述客户端将所述关联图像发送至服务器可以包括:将识别出的所述货物边界中的图像发送至所述服务器。
具体的一个示例中,在确定图4所示的货物边界后,可以将图4中货物边界中的图5所示的货物图像上传给服务器,图5是本说明书一个实施场景中识别出的货物边界中的图像。由图4、图5可以看出,本实施例中采用上传客户端识别的货物边界中的图像信息可以有效节省网络流量,减少传输的冗余信息,减少服务器图像数据处理负载并可以提供图像识别处理效率等。
当然,也可以采取其他识别货物边界的算法,包括在货物堆叠放到一起依然可以识别出货物变化的货物图像。例如图6、图7所示,图6是在图3所示的购物车中的第一件货物上叠放另一件货物的场景示意图,图7是根据图像差别识别出的图6中新放入货物的货物边界中的图像。
所述客户端将所述关联图像发送至服务器。客户端可以将获取的关联图像通过手机WIFI、局域网等近场通信或者运营商通信网络、专用电缆等网络通信连接发送给服务器。服务器可以存储客户端上传的关联图像。一种实施方式中,所述客户端可以直接将关联图像发送给服务器,例如通过运营商通信网络,其他的实施方式中,也可以设置多个客户端的统一接口设备,超市中的多个客户端可以通过近场通信将关联图像发送给统一接口设备,如超市服务器主机,然后由超市服务器主机将关联图像发送给进行图像中货物识别的服务器。
S4:所述服务器基于所述关联图像识别所述储物容器中的货物差异信息。
服务器获取客户端上传的关联图像后,可以对所述关联图像进行分析处理,例如按照时间顺序分别对比关联图像的差异,识别关联图像中新增加或拿掉的货物的信息。服务器处理的图像可以包括上述客户端拍摄的图像或视频类型的货物图像的数据,也可以包括客户端根据图像差别识别出的货物边界中的图像数据。
经过服务器对关联图像的分析处理,可以获取所述储物容器中货物变化的货物差异信息。所述的货物差异信息可以包括图像数据,也可以包括表示储物容器中发生变化的货物的标识信息,例如图3、图4所示的,识别出购物车中新放入一件货物,通过图像识别确定该货物为G01,货物识别代码为good_001,则此时可以确定所述储物容器中货物差异信息可以为:ADD:good_001,其中ADD可以表示此次货物变化为新增加一件货物。因此,本说明书提供的所述方法的另一个实施例中,所述述服务器基于所述关联图像识别所述储物容器中的货物差异信息可以包括:
S40:检测所述关联图像中的货物识别标识,以检测到的所述货物识别标识作为所述货物差异信息。
具体的一个示例中,例如可以服务器可以设置有条形码检测识别模块,可以识别所述关联图像中的条形码,所述的条形码中可以包括了货物的多重属性信息,例如货物名称、价格等。当购物车中的货物发生变化时,条形码检测模块检测识别输入的关联图像,并搜索有规律的条形码。如果识别成功,则可以确定货物差异信息。
当然,服务器一侧还可以设置其他类型检测和识别关联图像中货物的实现方式。本说明书提供的所述方法的另一种实施例中,可以利用机器学习算法来检测管关联图像中的货物。具体的,采用下述中的任意一种实施方式:
S42:获取所述关联图像后,利用机器学习算法得到的货物检测模型来检测所述关联图像中的货物信息,确定货物的差异信息;
S44:检测所述关联图像中的货物识别标识,以检测到的所述货物识别标识作为所述货物差异信息;
当对所述关联图像中的货物识别标识识别失败时,利用机器学习算法得到的货物检测模型检测所述关联图像中的货物信息,确定货物的差异信息。
所述的货物检测模型可以选择SVM(Support Vector Machine,支持向量机)、RCNN(Regions with CNN,区域卷积神经网络,一种目标检测算法)以及其他 机器学习算法及变形。选取的机器学习算法可以预先利用样本数据进行训练构建,符合实际场景或设计需要条件时可以本实施例中检测关联图像中货物的算法。一种实现方式上,可以获取所述关联图像后即将所述关联图像输入到货物检测模型中进行货物检测。另一种实现方式中,可以先进行关联图像中货物标识的识别,如检测识别货物条形码或二维码等,在货物识别标识识别失败时再利用更高级别的机器学习算法来检测关联图像中的货物,确定货物信息。
S6:所述服务器利用所述差异信息生成相应的货物订单变更信息。
当确定储物容器中货物发生变化的差异信息后,可以基于该差异信息生成相应的订单变更信息。例如检测出对应的货物并且查找到对应的唯一价格时,则可以生成货物订单变更信息。该货物订单信息可以表示用户新增加一个货物,可以用于用户的总订单结算处理中。
S8:所述服务器根据所述货物订单变更信息更新所述客户端对应的用户订单信息。
服务器一侧可以记录有用户的一个总的订单信息。一个实施方式中,当用户解锁客户端使用储物容器进行货物选购时,客户端可以与用户的识别标识进行绑定,例如一个用户绑定一个购物车。服务器识别出用户新加入或者拿掉购物车中的货物时,可以生成相应的货物订单变更信息,然后可以基于该货物订单变更信息相应的更新服务器一侧记录的用户的用户订单信息,例如添加一件物品的购物订单或者删除一件物品的购物订单。
S10:所述服务器将更新后的用户订单信息发送给所述客户端。
在本实施例实施场景中,客户端一侧可以设置有展示给用户已放入购物车中识别出的货物的订单信息,用户可以即时获取订单信息。当服务器一侧记录的用户的用户订单信息出现更新时,可以将更更新后的该用户的用户订单信息发生给客户端,以便客户端可以向用户展示或同步用户订单信息。
S12:所述客户端展示所述更新后的用户订单信息。
用户新放入或拿出物品时,客户端可以接收服务器识别出的用户新放入或 拿出物品后更新后的用户订单信息,然后可以将该更新后的用户订单信息进行展示,例如展示在按照在购物车上的智能手机中,或者展示在设置在储物容器上的其他显示设备。
上述实施例提供的一个或多个实施例提供的一种货物订单处理方法,可以在购物篮、购物车等储物容器中安装设置拍摄设备,远程服务器可以通过实时监控储物容器中货物的变化来识别和确认用户订单的变化,并可以生成或更新该用户的购物订单。这样,利用智能购物车等终端设备结合货物识别、订单处理的服务器来可以实现无人超市订单处理订单数据,节省大量自助收银设备、传感设备等。利用本说明书提供的实施方案,可以通过客户端图像或视频拍摄设备将货物检测的图像数据发送给服务器,由服务器进行货物识别、订单结算等处理,总体硬件实施成功更低,并且可以减少数据交互节点,方案实施更加简单便捷。在实际应用中可以大大降低无人超市货物订单自动结算处理的技术方案实施成本,便于实施,稳定性更好,无人超市面积、客流量、消费者行为特征等实施环境、因素的影响较小,利于大规模推广使用。
进一步的,本说明书还提供所述方法的另一种实施例。所述服务器一侧在对客户端上传的关联图像进行识别检测的过程中如果检测识别,无法识别关联图像中的货物,例如扫描不到条形码或者利用机器学习的商品检测模型也无法识别出关联图像中的货物,则在这种情况下可以将关联图像的信息或者以及用户信息、储物容器信息等发送给人工识别的终端设备。例如人工收银台的PC端、专用收银设备,或者设置的处理货物检测识别失败的其他终端设备。在本实施例中可以将上述包括用于人工识别处理的识别失败图像的终端设备统一称为收银节点。这样,当服务器图像中的货物检测失败时,则可以将图像发送给收银节点进行人工识别处理,可以有效保障货物订单的及时、稳定、持续处理,减少因服务器一侧无法识别货物而造成降低用户货物选购体验的现象。
本实施例中,可以设置有一个收银节点,也可以设置有多个收银节点。一般的,在无人超市中可以设置多个收银节点,每个收银节点可以对应一个网络 收银员,所述网络收银可以对所述收银节点的设备进行操作,对接收的货物图像进行识别或丢弃或提示用户等。对此,本申请实施例可以设置有任务调度队列和网络收银员类别,可以统一对多个无法识别的货物图像进行更加合理的任务调度。具体的,本说明书提供的所述方法的另一个实施例中,所述方法还可以包括:
S46:若所述关联图像中的货物识别标识检测识别或者所述货物检测模型检测失败,生成关联图像检测失败的作业任务,将所述作业任务置于任务调度队列中;
S48:按照获取的网络收银员列表将所述任务调度队列中的作业任务推送给符合作业条件的收银节点,所述网络收银员列表记录有收银节点处理任务的作业状态。
所述作业任务可以包括基于接收到的识别失败的图像信息以及其他例如用户标识、客户端标识等的信息形成的待处理的数据信息。任务调度队列中的作业任务可以分配给收银节点,收银节点可以根据任务信息进行相应的处理。还可以配置有网络收银员列表,所述网络收银员列表中可以记录各个收银节点当前的作业状态,如收银节点是否在线,某个收银节点当前有多少未处理的任务等。可以根据网络收银员列表中各个收银节点的作业状态将所述任务调度队列中的任务分配给符合作业条件的收银节点。所述的作业条件可以根据现场实施场景或设计需要进行设定,例如可以指定使用某个通道进行人识别时可以将一个或多个或者全部作业任务分配给指定的收银节点进行处理,或者,将作业任务分配给当前作业量少的收银节点进行处理。
需要说明的,在其他的实施例中所述的收银节点并不限于人工识别处理的终端设备,也可以包括存储或记录或转发或告警等作用的终端设备。
本说明书还提供另一种实施方式。在一些实施场景中,用户在选取购物车或者服务器一侧货物识别失败时,将用户的识别标识与网络收银员进行绑定。或者,没有网络收银处理某个用户的作业任务时,可以为该用户在任务调度队 列中的作业任务绑定一个网络收银。这样,当对任务调度列表中的作业任务进行分配时,可以先查询是否有与该任务绑定的网络收银员,如果有,则优先将该作业任务分配给绑定的收银员所在的收银节点。所述绑定的收银员可以包括正在处理与待分配的作业任务属于同一个用户的作业任务的收银员,或者服务器指定或者用户指定的某个收银员。当然,所述的绑定也可以包括客户端与网络收银绑定的实施方式。具体的,本申请提供的所述方法的另一个实施例中,所述方法还可以包括:
S47:查询所述网络收银员列表中是否有与作业任务绑定的网络收银员;
若有,则将对应的作业任务发送给相应的绑定的网络收银员所在的收银节点进行处理。
另一种实施方式中,如果绑定的收银网络收银员作业量已经达到饱和状态,如设置的最大处理作业量,则可以将作业任务发送给其他网络收银进行处理,可以均衡作业处理负载,提高处理效率。因此,本说明书所述方法的另一个实施例中,所述方法还可以包括:
S471:确定作业任务所绑定的网络收银员后,查询所述绑定的收银节点的作业状态;
若所述作业状态为作业量饱和,则将所述作业任务发送给所述网络收银员列表中作业状态为非作业量饱和的网络收银员所在的收银节点进行处理。
具体的一个示例中,例如可以保存有一个任务队列,该任务队列每一项包含用户id+识别失败的关联图像或视频内容+用户id对应的订单。同时保存有一个网络收银员的列表,以及每个网络收银员的任务列表。新来一个关联图像识别失败的作业任务时,查找用户id,确认是否已有网络收银员在绑定处理。如果已有,则判断其工作量是否饱和,若不饱和,则可以优先将作业任务推送给该网络收银员。如果其工作量饱和,则推送给新的网络收银员进行处理,并将用户id和新的收银员id绑定。如果用户id下的作业任务当前没有网络收银员处理,则可以推送给新的网络收银员处理,并将用户id和新的网络收银员id绑 定。这个新的网络收银员可以为非作业量饱和状态的指定网络收银或者随机选择的收银员或者按照预定顺序选择的网络收银。
另一个实施例中,还可以将作业任务(通常包括货物识别失败的关联图像)推送给多个收银员进行处理,然后多个收银员之间的结果可以互相检查,或参考,或者合并等。具体的,本说明书所述方法的另一个实施例中,所述按照获取的网络收银员列表将所述任务调度队列中的作业任务推送给符合作业条件的收银节点包括:
将所述作业任务发送给推送给至少两个收银节点:
相应的,所述基于所述关联图像识别所述储物容器中货物的差异信息包括:根据所述至少两个收银节点的货物识别结果确定所述储物容器中货物的差异信息。
例如在一些安全性较要求较高的应用场景中,可以将一个用户id推送给多个收银员处理,多个收银员之间的结果可以互相检查审核。这样可以进一步提高货物识别的准确性。
进一步的,如果网络收银员在收银节点也无法识别出客户端拍摄上传的货物图像中的货物,则可以向客户端发送提升提示消息。所述的提示消息可以包括多种类型信息,例如告知用户订单无法生成,或者将货物识别失败的结果返回给用户,提升无法识别。或者所述提示消息也可以包括提醒用户刚才放置的货物无法生成订单,请重新按要求放置,或者直接将作业任务标识为货物未识别订单。具体的,本申请所述方法的另一个实施例中,所述方法还可以包括:若所述收银节点的货物识别结果包括货物识别失败,则执行下述之一:
S691:所述服务器向所述客户端返回提示消息,所述提示消息至少包括货物订单无法生成、重新放置货物之一的信息内容;
S692:生成所述货物识别失败的关联图像对应的未识别货物订单,以及基于所述未识别货物订单的信息设置所述未识别货物订单对应的储物容器的身份验证结果为未通过
S692中具体的可以在所述储物容器需要通过通道装置时基于查询所述储物容器中是否有未识别货物订单的结果控制所述通道装置的阀门开闭。
如果货物识别失败,并且识别失败的货物仍然在储物容器中,可以当储物容器要通过出口出去时则可以控制通道装置阀门不开启或者关闭。此时用户可以到人工收银的通道进行处理。所述的通道可以包括超市、商场或地铁等使用的控制通道阀门开启和关闭以实现控制人员或储物容器通过的装置。具体的一个示例中,用户ID可以与使用的购物车的ID进行绑定。当用户推动购物车靠近自助结算通道时,自己结算通过可以通过识别码扫描或者红外灯近场通信识别出用户ID或者购物车的ID。然后可以到后台查询用户ID或者购物车ID的记录中是否有没有被识别的货物。如果购物车中有没有识别的货物,则不能通过自助结算通道离去(可以通过身份验证控制通道阀门开启关闭来控制),需要走有人工收银的通道进行进一步处理。
本说明书上述实施例提供的一种或多种货物订单处理的方法,可以提供一种便于用户购物且商家实施成本更低的订单结算方式,在无人超市应用中,可以通过实时监控购物商品的变换(通过图像识别处理)来识别、确认用户购买清单,可以服务器远程实时生成用户的总订单。用户购物完成后自动进行订单结算,在支付完成后可以直接离开无人超市,数据交互节点更少,无需配置大量昂贵的收银设备、光幕等传感设备,实施成本更低的同时订单数据处理效率更高、更稳定。
进一步的,本说明书还提供另一种实施方式。可以在客户端或客户端所在的储物容器上使用定位装置,例如GPS定位,或者UWB(Ultra Wideband,是一种无载波通信技术,利用纳秒至微微秒级的非正弦波窄脉冲传输数据,可以用于精确定位,目前可以用于室内定位)的定位方式。然后可以设置在一定的货物分布范围内设置该区域的识别标志,将该区域内的货物统一归属于一个区间,每一个区间可以对应一个该区间内的货物信息。这样,当货物进入到不同的货物归属区间后,用户挑选的货物通常是分布在该货物归属区间内的货物。因此, 本实施例结合储物容器的定位、货物的空间分区分布来确定当用户挑选货物时其选取的货物所在的货物归属区间,这样当无法直接通过关联图像识别货物时,可以在所述货物归属区间的货物信息范围内进行识别,可以极大的减少货物搜索区间,降低图像识别搜索区间,提高识别速度的和识别成功率的同时降低识别处理的负载。因此,本说明书提供的所述方法的另一个实施例中,所述方法还可以包括:
S14:客户端获取所述储物容器的定位信息,将所述定位信息发送给所述服务器;
所述服务器基于所述定位信息确定所述储物容器中货物发生变化的货物归属区间;
相应的,所述服务器基于所述关联图像识别所述储物容器中货物的差异信息包括:在所述货物归属区间的范围内识别所述关联图像中的货物信息,基于识别出的货物信息确定所述储物容器中货物变化的差异信息。
图8是本说明书提供的所述方法的另一个实施例的流程示意图。上述对本说明书特定实施例进行了描述。其它实施例在所附权利要求书的范围内。在一些情况下,在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序来执行并且仍然可以实现期望的结果。另外,在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中,多任务处理和并行处理也是可以的或者可能是有利的。
上述实施例中所述的服务器在一些应用场景中可以包括单独的一台服务器,也可以划分为不同的逻辑处理单元。例如可以包括接收和存储客户端上传的图像的模块,用于图像分析识别货物的模块、货物自动识别失败时进行人工任务调度的模块、网络收银员管理模块等多个逻辑终端,这些多个逻辑终端可以统一视为服务器一侧,也可以独立与服务器分别单独作为与服务器连接或耦合的设备。类似的,所述的客户端可以包括安装在购物车上的拍摄设备和通信设备,其他的应用场景中也可以将拍摄设备、通信设备、购物车等一同作为客 户端,或者再加上购物车的定位设备作为客户端。本说明书的一个或多个实施例中可以以检测储物容器中货物是否发生变化的图像检测的装置1与图像识别以及订单、处理等装置2的数据不同处理方/不同处理阶段来划分客户端、服务器,具体的可以根据实际应用环境来定义客户端、服务器以及服务器一侧的各个功能模块。
下图9是本说明书提供的一种货物订单处理方法的实施例场景示意图,在图9中,购物车上可以安装支架,用于放置手机或网络摄像头。服务器一侧可以划分为多个不同的处理单元。具体的,
101:购物车可以为商场或超市的普通购物车,有60升、80升、100升、125升等各种容量。支架可以为外装的智能手机支架,可以固定在购物车推手处前侧。用户购物时,商品可以认为一件件放置到购物车中。
100:手机可以包括高清后置摄像头和屏幕,可以放置在101的支架上。支架和手机安装时可以调整角度使得屏幕倾斜朝向用户,摄像头可以拍摄到购物车空间。手机内运行拍照应用,可以含图像变化检测模块,在图像变化时刻前后录制差异的图像和视频上传云端服务器,将变化的图像区域上传到云端,以便服务器一侧自动或人工分析单元对图像进行处理。这里可以设置为一个手机和一个用户一一绑定,当用户进行扫描或者其他方式确认使用某个购物车时,可以将购物车上手机与用户的ID(identification,识别标志)进行绑定。
102:云服务器或后台服务,与手机通过wifi或者运营商网络等通信连接,可以存储摄像头拍摄的购物车图片和视频。云服务器可以负责存储用户订单,以及向105和103输送分析内容。
105:视频自动分析单元,可以包括条型码检测识别模块、报警模块。其他的实施例中可以包括机器学习商品检测模块。机器学习检测可以基于SVM,RCNN等方法进行。当视频和图像变化时,条形码检测识别模块输入差异的图像,并搜索有规律的条形码。如果识别成功,则加入用户购物订单。如果条形码检测失败,可以将图像和视频信息发送给机器学习商品检测模块。在机器学 习商品检测模块中,如果检测出对应的商品并且查找到对应的唯一价格,则可以加入用户购物订单。如果检测失败,则可以将此图像发送给报警模块。报警模块可以将此信息封装成一项作业任务发送给人工任务调度单元。作业任务中可以包含用户id+报警图像或视频内容+用户id对应的当前订单等信息。
103:人工任务调度单元,可以保存有一个任务队列,该任务队列每一项可以包含用户id+报警图像或视频内容+用户id对应的订单。也可以保存有一个网络收银员的列表,以及每个网络收银员的任务列表。新来一个报警任务项时,可以查找用户id,确认是否已有网络收银员在绑定处理,如果已有,则判断其工作量是否饱和,若不饱和,优先推送给该收银员。如果其工作量饱和,则可以推送给新的收银员处理,并将用户id和新的收银员id绑定。如果用户id在当前时刻没有网络收银员处理,则推送给新的收银员处理,并将用户id和新的收银员id绑定。如前述一些实施例所述,可以在一些安全性要求较高的实施场景中,可以将一个用户id推送给多个收银员处理,多个收银员之间的结果可以互相检查审核。
104:网络收银员人工操控单元,可以通过以太网或者运营商网络与人工任务调度单元通信连接。可以设置屏幕,用于显示报警的图像或视频。网络收银员可以为真实的作业人员,网络收银员可以对报警的图像进行肉眼观察,寻找条形码区域,并进行识别,识别成功后加入用户订单。若条形码识别失败(例如遮挡或者其他原因),则可以从商店的货物列表中寻找有无完全对应的商品,如果寻找成功,加入用户订单。如果寻找失败,可以将商品标注后,返回结果到103。103可以将结果反馈给用户屏幕100,提醒用户刚才放置的商品无法生成订单,请重新按要求放置,或者直接将该商品放入用户订单未识别项。
另外,其他的实施方式中,101的购物车上可以安装类似UWB的定位方式,一方面可以帮助顾客导航以及客流统计,另外是可以通过定位,知道周围货架的商品SKU范围(SKU,Stock Keeping Unit库存量单位,可以是以件,盒,托盘等为单位。SKU通常被引申为产品统一编号的简称,每种产品均对应有唯一 的SKU号),减少图像识别搜索空间,提高商品机器/人工识别成功率。
由上述实施例的描述可以看出,本说明书一个或多个实施例提供的一种货物订单处理方法,可以在购物篮、购物车等储物容器中安装设置拍摄设备,远程服务器可以通过实时监控储物容器中货物的变化来识别和确认用户订单的变化,并可以生成或更新该用户的购物订单。这样,利用智能购物车等终端设备结合货物识别、订单处理的服务器来可以实现无人超市订单处理订单数据,节省大量自助收银设备、传感设备等。利用本说明书提供的实施方案,可以通过客户端图像或视频拍摄设备将货物检测的图像数据发送给服务器,由服务器进行货物识别、订单结算等处理,总体硬件实施成功更低,并且可以减少数据交互节点,方案实施更加简单便捷。在实际应用中可以大大降低无人超市货物订单自动结算处理的技术方案实施成本,便于实施,稳定性更好,无人超市面积、客流量、消费者行为特征等实施环境、因素的影响较小,利于大规模推广使用。
上述实施例从客户端与服务器,或者可以理解为从客户端一侧的终端设备与服务器一侧的终端设备的多方交互描述了本说明书提供的一个或多个货物订单处理方法的实施方式。基于上述实施例的描述,本说明书还提供一种可以用于服务器一侧的货物订单处理方法,所述的服务器可以包括单独的服务器作业系统,也可以包括图像存储、图像分析、作业调度等多个处理单元的单个服务器或服务器集群或分布式系统等。具体的,本说明书提供的一种可以用于服务器一侧的货物订单处理方法,图10是本说明书提供的可用于服务器一侧的所述方法一个的实施例流程示意图,如图10所示,可以包括:
S100:接收客户端上传的关联图像,所述关联图像包括客户端基于图像识别的方式检测到储物容器中货物发生变化时获取的图像;
S101:基于所述关联图像识别所述储物容器中货物的差异信息;
S102:利用所述差异信息生成相应的货物订单变更信息;
S103:根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
S104:将更新后的用户订单信息发送给所述客户端。
所述方法的一个实施例中,服务器接收的关联图像可以包括由客户端检测储物容器中的货物发生变化时拍摄获取的图像或者视频图像(在此将视频视为一种连续的图像)。客户端在检测判断储物容器中的货物是否变化时可以采用对储物容器拍摄的图像进行检测识别的方式,例如可以使前后两帧的货物图像相减,在划定区域内图像差绝对值如果超过一定阈值则可以认为储物容器中的货物发生变化。具体的,所述基于图像识别的方式检测储物容器中的货物是否发生变化可以包括:
S1001:通过客户端摄像设备拍摄所述储物容器中的货物图像;
S1002:按时间顺序检测所述货物图像之间差异数据;
S1003:若所述差异数据大于预设的变化阈值,则确定所述储物容器中的货物发生变化。
所述的关联图像可以包括采用下述至少一种方式确定的图像:
S1004:大于所述变化阈值的差异数据对应的货物图像;
S1005:当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定窗口范围内的货物图像;
S1006:当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定时间范围内的货物图像。
所述基于所述关联图像识别所述储物容器中的货物差异信息可以包括:
S1011:检测所述关联图像中的货物识别标识,以检测到的所述货物识别标识作为所述货物差异信息。
所述方法的另一种实施中,可以采用下述中的任意一种实施方式:
S1012:获取所述关联图像后,利用机器学习算法得到的货物检测模型来检测所述关联图像中的货物信息,确定货物的差异信息;
S1013:检测所述关联图像中的货物识别标识,以检测到的所述货物识别标识作为所述货物差异信息;
当对所述关联图像中的货物识别标识识别失败时,利用机器学习算法得到 的货物检测模型检测所述关联图像中的货物信息,确定货物的差异信息。
进一步的,本说明书提供的所述方法的另一个实施例中,所述方法还可以包括:
S1014:若所述关联图像中的货物识别失败,则生成关联图像检测失败的作业任务,将所述作业任务置于任务调度队列中;
S1015:按照获取的网络收银员列表将所述任务调度队列中的作业任务推送给符合作业条件的收银节点,所述网络收银员列表记录有收银节点处理任务的作业状态。
进一步的,本说明书提供的所述方法的另一个实施例中,所述方法还可以包括:
S1016:查询所述网络收银员列表中是否有与作业任务绑定的网络收银员;
若有,则将对应的作业任务发送给相应的绑定的网络收银员所在的收银节点进行处理。
进一步的,本说明书提供的所述方法的另一个实施例中,所述方法还可以包括:
S1017:确定作业任务所绑定的网络收银员后,查询所述绑定的收银节点的作业状态;
若所述作业状态为作业量饱和,则将所述作业任务发送给所述网络收银员列表中作业状态为非作业量饱和的网络收银员所在的收银节点进行处理。
进一步的,本说明书提供的所述方法的另一个实施例中,所述按照获取的网络收银员列表将所述任务调度队列中的作业任务推送给符合作业条件的收银节点包括:
S1018:将所述作业任务发送给推送给至少两个收银节点:
相应的,所述基于所述关联图像识别所述储物容器中货物的差异信息包括:根据所述至少两个收银节点的货物识别结果确定所述储物容器中货物的差异信息。
进一步的,本说明书提供的所述方法的另一个实施例中,所述方法还可以包括:
S1019:若所述收银节点的货物识别结果包括货物识别失败,则执行下述之一:
向所述客户端返回提示消息,所述提示消息至少包括货物订单无法生成、重新放置货物之一的信息内容;
生成所述货物识别失败的关联图像对应的未识别货物订单,以及基于所述未识别货物订单的信息设置所述未识别货物订单对应的储物容器的身份验证结果为未通过。
进一步的,本说明书提供的所述方法的另一个实施例中,所述方法还可以包括:
S10110:接收客户端上传的所述储物容器的定位信息,基于所述定位信息确定所述储物容器中货物发生变化的货物归属区间;
相应的,所述基于所述关联图像识别所述储物容器中货物的差异信息包括:在所述货物归属区间的范围内识别所述关联图像中的货物信息,基于识别出的货物信息确定所述储物容器中货物变化的差异信息。
上述实施例描述了服务器一侧的货物订单处理方法的多个实施例,其具体的实施过程及实现方式可以参照前述客户端与服务器交互实施例的相关描述,在此不做赘述。
当然,基于上述客户端与服务器交互实施例的描述,本说明书还提供一种可以用于客户端一侧的货物订单处理方法,所述的客户端可以包括安装在例如购物车上的具有拍摄和数据通信能力的终端装置,所述的客户端中也可以包括图像变化检测的模块,可以基于拍摄装置获取的图像检测到储物容器中的货物发生变化。其他的一些实施场景中,也可以包括将购物车、支架设备、显示屏或者定位设备等的一个或多个与所述拍摄装置、图像传输的数据通信装置一同视为客户端。所述的方法可以用于客户端一侧,可以实现检测购物中货物的变 化,并将货物发生变化时的图片或视频发送给服务器进行识别和处理,生成用户的订单信息。具体的,本说明书提供的一种可以用于客户端一侧的货物订单处理方法,图11是本说明书提供的可以用于客户端一侧所述方法的一个实施例流程示意图,如图11所示,可以包括:
S200:基于图像识别的方式检测储物容器中的货物是否发生变化;
S201:若检测到货物发生变化,则将所述储物容器中货物发生变化的关联图像发送至服务器;
S202:接收所述服务器返回的更新后的用户订单信息;
S203:展示所述更新后的用户订单信息。
参照前述实施例的描述,本说明书还提供的另一种可以用于客户端一侧的货物订单处理方法中,所述基于图像识别的方式检测储物容器中的货物是否发生变化可以包括:
S2001:通过摄像设备拍摄所述储物容器中的货物图像;
S2002:按时间顺序检测所述货物图像之间差异数据;
S2003:若所述差异数据大于预设的变化阈值,则确定所述储物容器中的货物发生变化。
所述的关联图像可以包括多种确定方式。本说明书提供的所述方法的另一个实施例中,所述的关联图像可以包括采用下述至少一种方式确定的图像:
S2004:大于所述变化阈值的差异数据对应的货物图像;
S2005:当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定窗口范围内的货物图像;
S2006:当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定时间范围内的货物图像。
图12是本说明书提供的可以用于客户端一侧所述方法的另一个实施例流程示意图。另一种实施例中,在客户端上传关联图像的处理中,可以仅将货物图像发送变化的图像内容发送给服务器进行处理,这样可以节省网络带宽浏览, 提高处理速度。具体的,所述方法的另一个实施例中,获取所述关联图像后,所述方法还可以包括:
S2007:识别所述关联图像中货物发生变化的货物边界;
相应的,所述将所述关联图像发送至服务器可以包括:将识别出的所述货物边界中的图像发送至所述服务器。
进一步的,还可以在购物车上安装类似UWB的定位方式,一方面可以帮助顾客导航以及客流统计,另一方面可以通过定位周围货架的商品SKU范围,减少图像识别搜索空间,提高商品机器/人工识别成功率。具体的,所述方法的另一个实施例中,所述方法还可以包括:
S2008:获取所述储物容器的定位信息,将所述定位信息发送给所述服务器,以使所述服务器基于所述定位信息确定所述储物容器中货物的货物归属区间。
通过上述客户端与服务器交互、服务器一侧、客户端一侧的多个实施例可以看出,本说明书提供的货物订单处理方法,可以用于无人超市类型的消费者自助购物、平台自动识别货物并进行订单结算的应用场景中。利用本说明书提供的实施例,可以在购物篮、购物车等储物容器中安装设置拍摄设备,远程服务器可以通过实时监控储物容器中货物的变化来识别和确认用户订单的变化,并可以生成或更新该用户的购物订单。这样,利用智能购物车等终端设备结合货物识别、订单处理的服务器来可以实现无人超市订单处理订单数据,节省大量自助收银设备、传感设备等。利用本说明书提供的实施方案,可以通过客户端图像或视频拍摄设备将货物检测的图像数据发送给服务器,由服务器进行货物识别、订单结算等处理,总体硬件实施成功更低,并且可以减少数据交互节点,方案实施更加简单便捷。在实际应用中可以大大降低无人超市货物订单自动结算处理的技术方案实施成本,便于实施,稳定性更好,无人超市面积、客流量、消费者行为特征等实施环境、因素的影响较小,利于大规模推广使用。
基于上述所述的货物订单处理方法,本说明书一个或多个实施例还提供一种货物订单处理装置。所述的装置可以包括使用了本说明书实施例所述方法的 系统(包括分布式系统)、软件(应用)、模块、组件、服务器、客户端等并结合必要的实施硬件的装置。基于同一创新构思,本说明书实施例提供的一个或多个实施例中的装置如下面的实施例所述。由于装置解决问题的实现方案与方法相似,因此本说明书实施例具体的装置的实施可以参见前述方法的实施,重复之处不再赘述。以下所使用的,术语“单元”或者“模块”可以实现预定功能的软件和/或硬件的组合。尽管以下实施例所描述的装置较佳地以软件来实现,但是硬件,或者软件和硬件的组合的实现也是可能并被构想的。具体的,图13是本说明书提供的一种可以用于服务器一侧的货物订单处理装置实施例的模块结构示意图,如图13所示,所述装置可以包括:
图像接收模块101,可以用于接收客户端上传的关联图像,所述关联图像包括客户端基于图像识别的方式检测到储物容器中货物发生变化时获取的图像;
图像识别模块102,可以用于基于所述关联图像识别所述储物容器中货物的差异信息;
订单生成模块103,可以用于利用所述差异信息生成相应的货物订单变更信息;
订单更新模块104,可以用于根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
信息反馈模块105,可以用于将更新后的用户订单信息发送给所述客户端。
所述装置的另一个实施例中,所述基于图像识别的方式检测储物容器中的货物是否发生变化可以包括:
通过客户端摄像设备拍摄所述储物容器中的货物图像;
按时间顺序检测所述货物图像之间差异数据;
若所述差异数据大于预设的变化阈值,则确定所述储物容器中的货物发生变化。
所述装置的另一个实施例中,所述的关联图像可以包括采用下述至少一种 方式确定的图像:
大于所述变化阈值的差异数据对应的货物图像;
当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定窗口范围内的货物图像;
当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定时间范围内的货物图像。
图14是本说明书提供的所述装置中图像识别模块一个实施例的模块结构示意图。如图14所示,所述图像识别模块102可以包括:
识别码检测模块1021,可以用于检测所述关联图像中的货物识别标识,以检测到的所述货物识别标识作为所述货物差异信息。
如图14所示,所述装置的一个实施例中所述图像识别模块102可以包括:
货物检测模块1022,所述货物检测模块1022包括基于选取的机器学习算法经过样本训练后得到的货物检测模型,可以用于获取所述关联图像后检测所述关联图像中的货物信息,确定货物的差异信息;或者,可以用于当对所述关联图像中的货物识别标识识别失败时,检测所述关联图像中的货物信息,确定货物的差异信息。
图15是本说明书提供的所述装置中另一个实施例的模块结构示意图。如图15所示,所述装置还可以包括:
人工调度模块106,可以用于在所述关联图像中的货物识别标识检测识别或者所述货物检测模型检测失败,生成关联图像检测失败的作业任务,将所述作业任务置于任务调度队列中;以及,
按照获取的网络收银员列表将所述任务调度队列中的作业任务推送给符合作业条件的收银节点,所述网络收银员列表记录有收银节点处理任务的作业状态。
图16是本说明书提供的所述装置中另一个实施例的模块结构示意图。如图16所示,所述装置的另一个实施例中,所述装置还可以包括:
收银员控制模块107,可以用于查询所述网络收银员列表中是否有与作业任务绑定的网络收银员;
若有,则将对应的作业任务发送给相应的绑定的网络收银员所在的收银节点进行处理。
所述装置的另一个实施例中,所述收银员控制模块107确定作业任务所绑定的网络收银员后,还可以用于查询所述绑定的收银节点的作业状态;以及,
若所述作业状态为作业量饱和,则将所述作业任务发送给所述网络收银员列表中作业状态为非作业量饱和的网络收银员所在的收银节点进行处理。
所述装置的另一个实施例中,所述人工调度模块106按照获取的网络收银员列表将所述任务调度队列中的作业任务推送给符合作业条件的收银节点包括:
将所述作业任务发送给推送给至少两个收银节点:
相应的,所述图像识别模块102基于所述关联图像识别所述储物容器中货物的差异信息包括:根据所述至少两个收银节点的货物识别结果确定所述储物容器中货物的差异信息。
所述装置的另一个实施例中,所述信息反馈模块105可以在所述收银节点的货物识别结果包括货物识别失败时,则执行下述之一:
向所述客户端返回提示消息,所述提示消息至少包括货物订单无法生成、重新放置货物之一的信息内容;
生成所述货物识别失败的关联图像对应的未识别货物订单,以及基于所述未识别货物订单的信息设置所述未识别货物订单对应的储物容器的身份验证结果为未通过。
图17是本说明书提供的所述装置中另一个实施例的模块结构示意图。如图17所示,所述装置的另一个实施例中,所述装置还可以包括:
定位处理模块108,可以用于接收客户端上传的所述储物容器的定位信息,基于所述定位信息确定所述储物容器中货物发生变化的货物归属区间;
相应的,所述图像识别模块102基于所述关联图像识别所述储物容器中货物的差异信息包括:在所述货物归属区间的范围内识别所述关联图像中的货物信息,基于识别出的货物信息确定所述储物容器中货物变化的差异信息。
需要说明的,上述所述的装置根据客户端与服务器交互的实施例或者服务器一侧方法实施例的描述还可以包括其他的实施方式,具体的实现方式可以参照相关方法实施例的描述,在此不作一一赘述。
上述对本说明书特定实施例进行了描述。其它实施例在所附权利要求书的范围内。在一些情况下,在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序来执行并且仍然可以实现期望的结果。另外,在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中,多任务处理和并行处理也是可以的或者可能是有利的。例如一个示例中,可以将人工调度模块106与收银员控制模块107合并为一个处理单元或功能模块实施其步骤。
本说明书还可以提供一种可以用于客户端一侧的货物订单处理装置。图18是本说明书一个的一种货物订单处理装置的实施例模块结构示意图,如图18所示,所述装置可以包括:
货物变化检测模块201,可以用于基于图像识别的方式检测储物容器中的货物是否发生变化;
图像发送模块202,可以用于在检测到货物发生变化时,将所述储物容器中货物发生变化的关联图像发送至服务器;
信息接收模块203,可以用于接收所述服务器返回的更新后的用户订单信息;
显示模块204,可以用于展示所述更新后的用户订单信息。
图19是本说明书所述装置中货物变化检测模块一种实施例的模块结构示意图,如图19所示,所述货物变化检测模块201可以包括:
拍摄单元2011,可以用于拍摄所述储物容器中的货物图像;
图像差异处理单元2012,可以用于按时间顺序检测所述货物图像之间差异数据;以及,
货物变化确定模块2013,可以用于在所述差异数据大于预设的变化阈值时,确定所述储物容器中的货物发生变化。
所述装置的另一个实施例中,所述图像发送模块202中的关联图像,可以包括采用下述至少一种方式确定的图像:
大于所述变化阈值的差异数据对应的货物图像;
当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定窗口范围内的货物图像;
当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定时间范围内的货物图像。
图20是本说明书所述装置中货物变化检测模块另一种实施例的模块结构示意图,如图20所示,所述货物变化检测模块201可以包括:
变化边界识别模块2014,可以用于识别所述关联图像中货物发生变化的货物边界;
相应的,所述图像发送模块202将所述关联图像发送至服务器可以包括:将识别出的所述货物边界中的图像发送至所述服务器。
图21是本说明书所述装置另一种实施例的模块结构示意图,如图21所示,所述装置还可以包括:
位置定位模块205,可以用于获取所述储物容器的定位信息,将所述定位信息发送给所述服务器,以使所述服务器基于所述定位信息确定所述储物容器中货物的货物归属区间。
需要说明的,上述所述的装置根据客户端与服务器交互的实施例或者客户端一侧方法实施例的描述还可以包括其他的实施方式,具体的实现方式可以参照相关方法实施例的描述,在此不作一一赘述。
上述对本说明书特定实施例进行了描述。其它实施例在所附权利要求书的 范围内。在一些情况下,在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序来执行并且仍然可以实现期望的结果。另外,在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中,多任务处理和并行处理也是可以的或者可能是有利的。例如一个示例中,图像发送模块202和信息接收模块203可以合并在一个通信模块中实现其实施步骤。
本说明书一个或多个实施例提供的一种货物订单处理装置,可以在购物篮、购物车等储物容器中安装设置拍摄设备,远程服务器可以通过实时监控储物容器中货物的变化来识别和确认用户订单的变化,并可以生成或更新该用户的购物订单。这样,利用智能购物车等终端设备结合货物识别、订单处理的服务器来可以实现无人超市订单处理订单数据,节省大量自助收银设备、传感设备等。利用本说明书提供的实施方案,可以通过客户端图像或视频拍摄设备将货物检测的图像数据发送给服务器,由服务器进行货物识别、订单结算等处理,总体硬件实施成功更低,并且可以减少数据交互节点,方案实施更加简单便捷。在实际应用中可以大大降低无人超市货物订单自动结算处理的技术方案实施成本,便于实施,稳定性更好,无人超市面积、客流量、消费者行为特征等实施环境、因素的影响较小,利于大规模推广使用。
本说明书实施例提供的上述货物订单处理方法或装置可以在计算机中由处理器执行相应的程序指令来实现,如使用windows操作系统的c++语言在PC端实现、linux系统实现,或其他例如使用android、iOS系统程序设计语言在智能终端实现,以及基于量子计算机的处理逻辑实现等。所述装置可以用于多种订单结算系统平台、自助购物订单处理系统、购物云服务系统等的多种服务器中,实现对用户购物物品的低成本、快速、高效、可靠的订单处理。具体的,本说明书提供一种服务器,如图22所示,可以包括至少一个处理器以及用于存储处理器可执行指令的存储器,所述处理器执行所述指令时实现:
接收客户端上传的关联图像,所述关联图像包括客户端基于图像识别的方 式检测到储物容器中货物发生变化时获取的图像;
基于所述关联图像识别所述储物容器中货物的差异信息;
利用所述差异信息生成相应的货物订单变更信息;
根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
将更新后的用户订单信息发送给所述客户端。
需要说明的是,上述所述的服务器根据方法或装置实施例的描述还可以包括其他的实施方式,具体的实现方式可以参照相关方法实施例的描述,在此不作一一赘述。
所述的服务器可以包括单独的服务器,也可以包括包含多个服务器的服务器系统,如存储用户订单的云服务器、人工作业任务调度的服务器、网络收银员操作的服务器、视频或图像自动分析的服务器等,也可以是分布式服务器或服务器集群的构架。
需要说明的是,本说明书上述所述的装置或电子设备根据相关方法实施例的描述还可以包括其他的实施方式,具体的实现方式可以参照方法实施例的描述,在此不作一一赘述。本说明书中的各个实施例均采用递进的方式描述,各个实施例之间相同相似的部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。尤其,对于硬件+程序类实施例而言,由于其基本相似于方法实施例,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
当然,本说明书基于前述客户端一侧方法或装置的实施例描述,还可以提供一种购物终端设备,所述的购物终端可以包括前述的例如具有拍摄和图像数据传输功能的手机、或者还包括购物车、定位设备等的终端设备。所述的客户端可以包括图像拍摄和检测单元,可以由处理器执行相应的程序指令来实现,如使用windows操作系统的c++语言在PC端实现、linux系统实现,或其他例如使用android、iOS系统程序设计语言在智能终端实现,以及基于量子计算机的处理逻辑实现等。所述购物终端设备可以拍摄例如购物车等储物容器中的储物空间图像,当消费者放入或拿出货物时可以检测到货物的变化,并拍摄图像 上传至服务器,以便服务器对图像中的货物进行识别,进而生成或处理用户的订单信息,实现对用户购物物品的低成本、快速、高效、可靠的订单处理。具体的,本说明书提供一种客户端,可以包括至少一个处理器以及用于存储处理器可执行指令的存储器,所述处理器执行所述指令时实现:
基于图像识别的方式检测储物容器中的货物是否发生变化;
若检测到货物发生变化,则将所述储物容器中货物发生变化的关联图像发送至服务器;
接收所述服务器返回的更新后的用户订单信息;
展示所述更新后的用户订单信息。
需要说明的是,上述所述的客户端根据方法或装置实施例的描述还可以包括其他的实施方式,具体的实现方式可以参照相关方法实施例的描述,在此不作一一赘述。
上述对本说明书特定实施例进行了描述。其它实施例在所附权利要求书的范围内。在一些情况下,在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序来执行并且仍然可以实现期望的结果。另外,在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中,多任务处理和并行处理也是可以的或者可能是有利的。
基于上述所述的方法或装置或客户端或服务器的描述,本说明书还提供一个或多个货物订单处理系统。图23是本说明书提供的所述系统一种实施例的架构示意图,所述系统可以包括购物终端1和服务器2,
所述服务器1可以包括上述任意一个服务器一侧实施例所述的装置或实现任意一个服务器一侧所述的方法的步骤,
所述购物终端2可以包括上述任意一个实施例所述的购物终端一侧实施例所述的装置或者实现任意一个购物终端一侧所述的方法的步骤。
本说明书一个或多个实施例提供的一种货物订单处理方法、装置、服务器、购物终端及系统,可以在购物篮、购物车等储物容器中安装设置拍摄设备,远 程服务器可以通过实时监控储物容器中货物的变化来识别和确认用户订单的变化,并可以生成或更新该用户的购物订单。这样,利用智能购物车等终端设备结合货物识别、订单处理的服务器来可以实现无人超市订单处理订单数据,节省大量自助收银设备、传感设备等。利用本说明书提供的实施方案,可以通过客户端图像或视频拍摄设备将货物检测的图像数据发送给服务器,由服务器进行货物识别、订单结算等处理,总体硬件实施成功更低,并且可以减少数据交互节点,方案实施更加简单便捷。在实际应用中可以大大降低无人超市货物订单自动结算处理的技术方案实施成本,便于实施,稳定性更好,无人超市面积、客流量、消费者行为特征等实施环境、因素的影响较小,利于大规模推广使用。
尽管说明书实施例内容中提到采用智能手机进行图像拍摄和传输、使用RCNN方法构建检测模型、设置收银员作业状态实现任务分配、以及图像中货物识别失败时的多种处理方式等之类的数据/定义、存储、获取、交互、计算、判断等描述,但是,本说明书实施例并不局限于必须是符合行业通信标准、机器学习模型、标准计算机数据处理和数据存储规则或本说明书一个或多个实施例所描述的情况。某些行业标准、机器学习模型或者使用自定义方式或实施例描述的实施基础上略加修改后的实施方案也可以实现上述实施例相同、等同或相近、或变形后可预料的实施效果。应用这些修改或变形后的数据获取、存储、判断、处理方式等获取的实施例,仍然可以属于本说明书实施例的可选实施方案范围之内。
上述对本说明书特定实施例进行了描述。其它实施例在所附权利要求书的范围内。在一些情况下,在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序来执行并且仍然可以实现期望的结果。另外,在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中,多任务处理和并行处理也是可以的或者可能是有利的。
在20世纪90年代,对于一个技术的改进可以很明显地区分是硬件上的改进(例如,对二极管、晶体管、开关等电路结构的改进)还是软件上的改进(对 于方法流程的改进)。然而,随着技术的发展,当今的很多方法流程的改进已经可以视为硬件电路结构的直接改进。设计人员几乎都通过将改进的方法流程编程到硬件电路中来得到相应的硬件电路结构。因此,不能说一个方法流程的改进就不能用硬件实体模块来实现。例如,可编程逻辑器件(Programmable Logic Device,PLD)(例如现场可编程门阵列(Field Programmable Gate Array,FPGA))就是这样一种集成电路,其逻辑功能由用户对器件编程来确定。由设计人员自行编程来把一个数字系统“集成”在一片PLD上,而不需要请芯片制造厂商来设计和制作专用的集成电路芯片。而且,如今,取代手工地制作集成电路芯片,这种编程也多半改用“逻辑编译器(logic compiler)”软件来实现,它与程序开发撰写时所用的软件编译器相类似,而要编译之前的原始代码也得用特定的编程语言来撰写,此称之为硬件描述语言(Hardware Description Language,HDL),而HDL也并非仅有一种,而是有许多种,如ABEL(Advanced Boolean Expression Language)、AHDL(Altera Hardware Description Language)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL(Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(Ruby Hardware Description Language)等,目前最普遍使用的是VHDL(Very-High-Speed Integrated Circuit Hardware Description Language)与Verilog。本领域技术人员也应该清楚,只需要将方法流程用上述几种硬件描述语言稍作逻辑编程并编程到集成电路中,就可以很容易得到实现该逻辑方法流程的硬件电路。
控制器可以按任何适当的方式实现,例如,控制器可以采取例如微处理器或处理器以及存储可由该(微)处理器执行的计算机可读程序代码(例如软件或固件)的计算机可读介质、逻辑门、开关、专用集成电路(Application Specific Integrated Circuit,ASIC)、可编程逻辑控制器和嵌入微控制器的形式,控制器的例子包括但不限于以下微控制器:ARC 625D、Atmel AT91SAM、Microchip PIC18F26K20以及Silicone Labs C8051F320,存储器控制器还可以被实现为存储器的控制逻辑的一部分。本领域技术人员也知道,除了以纯计算机可读程序 代码方式实现控制器以外,完全可以通过将方法步骤进行逻辑编程来使得控制器以逻辑门、开关、专用集成电路、可编程逻辑控制器和嵌入微控制器等的形式来实现相同功能。因此这种控制器可以被认为是一种硬件部件,而对齐内包括的用于实现各种功能的装置也可以视为硬件部件内的结构。或者甚至,可以将用于实现各种功能的装置视为既可以是实现方法的软件模块又可以是硬件部件内的结构。
上述实施例阐明的系统、装置、模块或单元,具体可以由计算机芯片或实体实现,或者由具有某种功能的产品来实现。一种典型的实现设备为计算机。具体的,计算机例如可以为个人计算机、膝上型计算机、车载人机交互设备、蜂窝电话、相机电话、智能电话、个人数字助理、媒体播放器、导航设备、电子邮件设备、游戏控制台、平板计算机、可穿戴设备或者这些设备中的任何设备的组合。
虽然本说明书一个或多个实施例提供了如实施例或流程图所述的方法操作步骤,但基于常规或者无创造性的手段可以包括更多或者更少的操作步骤。实施例中列举的步骤顺序仅仅为众多步骤执行顺序中的一种方式,不代表唯一的执行顺序。在实际中的装置或终端产品执行时,可以按照实施例或者附图所示的方法顺序执行或者并行执行(例如并行处理器或者多线程处理的环境,甚至为分布式数据处理环境)。术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、产品或者设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、产品或者设备所固有的要素。在没有更多限制的情况下,并不排除在包括所述要素的过程、方法、产品或者设备中还存在另外的相同或等同要素。
为了描述的方便,描述以上装置时以功能分为各种模块分别描述。当然,在实施本说明书一个或多个时可以把各模块的功能在同一个或多个软件和/或硬件中实现,也可以将实现同一功能的模块由多个子模块或子单元的组合实现等。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,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。
本领域技术人员也知道,除了以纯计算机可读程序代码方式实现控制器以外,完全可以通过将方法步骤进行逻辑编程来使得控制器以逻辑门、开关、专用集成电路、可编程逻辑控制器和嵌入微控制器等的形式来实现相同功能。因此这种控制器可以被认为是一种硬件部件,而对齐内部包括的用于实现各种功能的装置也可以视为硬件部件内的结构。或者甚至,可以将用于实现各种功能的装置视为既可以是实现方法的软件模块又可以是硬件部件内的结构。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
在一个典型的配置中,计算设备包括一个或多个处理器(CPU)、输入/输出 接口、网络接口和内存。
内存可能包括计算机可读介质中的非永久性存储器,随机存取存储器(RAM)和/或非易失性内存等形式,如只读存储器(ROM)或闪存(flash RAM)。内存是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括,但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带,磁带磁磁盘存储、石墨烯存储或其他磁性存储设备或任何其他非传输介质,可用于存储可以被计算设备访问的信息。按照本文中的界定,计算机可读介质不包括暂存电脑可读媒体(transitory media),如调制的数据信号和载波。
本领域技术人员应明白,本说明书一个或多个实施例可提供为方法、系统或计算机程序产品。因此,本说明书一个或多个实施例可采用完全硬件实施例、完全软件实施例或结合软件和硬件方面的实施例的形式。而且,本说明书一个或多个实施例可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本说明书一个或多个实施例可以在由计算机执行的计算机可执行指令的一般上下文中描述,例如程序模块。一般地,程序模块包括执行特定任务或实现特定抽象数据类型的例程、程序、对象、组件、数据结构等等。也可以在分布式计算环境中实践本本说明书一个或多个实施例,在这些分布式计算环境中,由通过通信网络而被连接的远程处理设备来执行任务。在分布式计算环境中,程序模块可以位于包括存储设备在内的本地和远程计算机存储介质中。
本说明书中的各个实施例均采用递进的方式描述,各个实施例之间相同相似的部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。尤其,对于系统实施例而言,由于其基本相似于方法实施例,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。在本说明书的描述中,参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本说明书的至少一个实施例或示例中。在本说明书中,对上述术语的示意性表述不必须针对的是相同的实施例或示例。而且,描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外,在不相互矛盾的情况下,本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
以上所述仅为本说明书一个或多个实施例的实施例而已,并不用于限制本本说明书一个或多个实施例。对于本领域技术人员来说,本说明书一个或多个实施例可以有各种更改和变化。凡在本申请的精神和原理之内所作的任何修改、等同替换、改进等,均应包含在权利要求范围之内。

Claims (36)

  1. 一种货物订单处理方法,所述方法包括:
    客户端基于图像识别的方式检测储物容器中的货物是否发生变化;
    若检测到货物发生变化,则所述客户端将所述储物容器中货物发生变化的关联图像发送至服务器;
    所述服务器基于所述关联图像识别所述储物容器中货物的差异信息;
    所述服务器利用所述差异信息生成相应的货物订单变更信息;
    所述服务器根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
    所述服务器将更新后的用户订单信息发送给所述客户端;
    所述客户端展示所述更新后的用户订单信息。
  2. 一种货物订单处理方法,所述方法包括:
    接收客户端上传的关联图像,所述关联图像包括客户端基于图像识别的方式检测到储物容器中货物发生变化时获取的图像;
    基于所述关联图像识别所述储物容器中货物的差异信息;
    利用所述差异信息生成相应的货物订单变更信息;
    根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
    将更新后的用户订单信息发送给所述客户端。
  3. 如权利要求2所述的一种货物订单处理方法,所述基于图像识别的方式检测储物容器中的货物是否发生变化包括:
    通过客户端摄像设备拍摄所述储物容器中的货物图像;
    按时间顺序检测所述货物图像之间差异数据;
    若所述差异数据大于预设的变化阈值,则确定所述储物容器中的货物发生变化。
  4. 如权利要求3所述的一种货物订单处理方法,所述的关联图像包括采用下述至少一种方式确定的图像:
    大于所述变化阈值的差异数据对应的货物图像;
    当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定窗口范围内的货物图像;
    当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定时间范围内的货物图像。
  5. 如权利要求2所述的一种货物订单处理方法,所述基于所述关联图像识别所述储物容器中的货物差异信息包括:
    获取所述关联图像后,利用机器学习算法得到的货物检测模型来检测所述关联图像中的货物信息,确定货物的差异信息。
  6. 如权利要求2所述的一种货物订单处理方法,所述基于所述关联图像识别所述储物容器中的货物差异信息包括:
    检测所述关联图像中的货物识别标识,以检测到的所述货物识别标识作为所述货物差异信息;
    当对所述关联图像中的货物识别标识识别失败时,利用机器学习算法得到的货物检测模型检测所述关联图像中的货物信息,确定货物的差异信息。
  7. 如权利要求6中所述的一种货物订单处理方法,所述方法还包括:
    若所述关联图像中的货物识别失败,则生成所述关联图像检测失败的作业任务,将所述作业任务置于任务调度队列中;
    按照获取的网络收银员列表将所述任务调度队列中的作业任务推送给符合作业条件的收银节点,所述网络收银员列表记录有收银节点处理任务的作业状态。
  8. 如权利要求7中所述的一种货物订单处理方法,所述方法还包括:
    查询所述网络收银员列表中是否有与作业任务绑定的网络收银员;
    若有,则将对应的作业任务发送给相应的绑定的网络收银员所在的收银节点进行处理。
  9. 如权利要求8中所述的一种货物订单处理方法,所述方法还包括:
    确定作业任务所绑定的网络收银员后,查询所述绑定的收银节点的作业状态;
    若所述作业状态为作业量饱和,则将所述作业任务发送给所述网络收银员列表中作业状态为非作业量饱和的网络收银员所在的收银节点进行处理。
  10. 如权利要求7中所述的一种货物订单处理方法,所述按照获取的网络收银员列表将所述任务调度队列中的作业任务推送给符合作业条件的收银节点包括:
    将所述作业任务发送给推送给至少两个收银节点:
    相应的,所述基于所述关联图像识别所述储物容器中货物的差异信息包括:根据所述至少两个收银节点的货物识别结果确定所述储物容器中货物的差异信息。
  11. 如权利要求7中所述的一种货物订单处理方法,所述方法还包括:
    若所述收银节点的货物识别结果包括货物识别失败,则执行下述至少之一:
    向所述客户端返回提示消息,所述提示消息至少包括货物订单无法生成、重新放置货物之一的信息内容;
    生成所述货物识别失败的关联图像对应的未识别货物订单,以及基于所述未识别货物订单的信息设置所述未识别货物订单对应的储物容器的身份验证结果为未通过。
  12. 如权利要求2中所述的一种货物订单处理方法,所述方法还包括:
    接收客户端上传的所述储物容器的定位信息,基于所述定位信息确定所述储物容器中货物发生变化的货物归属区间;
    相应的,所述基于所述关联图像识别所述储物容器中货物的差异信息包括:在所述货物归属区间的范围内识别所述关联图像中的货物信息,基于识别出的货物信息确定所述储物容器中货物变化的差异信息。
  13. 一种货物订单处理方法,所述方法包括:
    基于图像识别的方式检测储物容器中的货物是否发生变化;
    若检测到货物发生变化,则将所述储物容器中货物发生变化的关联图像发送至服务器;
    接收所述服务器返回的更新后的用户订单信息;
    展示所述更新后的用户订单信息。
  14. 如权利要求13所述的一种货物订单处理方法,所述基于图像识别的方式检测储物容器中的货物是否发生变化包括:
    通过摄像设备拍摄所述储物容器中的货物图像;
    按时间顺序检测所述货物图像之间差异数据;
    若所述差异数据大于预设的变化阈值,则确定所述储物容器中的货物发生变化。
  15. 如权利要求13所述的一种货物订单处理方法,所述的关联图像包括采用下述至少一种方式确定的图像:
    大于所述变化阈值的差异数据对应的货物图像;
    当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定窗口范围内的货物图像;
    当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定时间范围内的货物图像。
  16. 如权利要求13所述一种货物订单处理方法,获取所述关联图像后,所述方法还包括:
    识别所述关联图像中货物发生变化的货物边界;
    相应的,所述将所述关联图像发送至服务器包括:将识别出的所述货物边界中的图像发送至所述服务器。
  17. 如权利要求13所述的一种货物订单处理方法,所述方法还包括:
    获取所述储物容器的定位信息,将所述定位信息发送给所述服务器,以使所述服务器基于所述定位信息确定所述储物容器中货物的货物归属区间。
  18. 一种货物订单处理装置,所述装置包括:
    图像接收模块,用于接收客户端上传的关联图像,所述关联图像包括客户端基于图像识别的方式检测到储物容器中货物发生变化时获取的图像;
    图像识别模块,用于基于所述关联图像识别所述储物容器中货物的差异信息;
    订单生成模块,用于利用所述差异信息生成相应的货物订单变更信息;
    订单更新模块,用于根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
    信息反馈模块,用于将更新后的用户订单信息发送给所述客户端。
  19. 如权利要求18所述的一种货物订单处理装置,所述基于图像识别的方式检测储物容器中的货物是否发生变化包括:
    通过客户端摄像设备拍摄所述储物容器中的货物图像;
    按时间顺序检测所述货物图像之间差异数据;
    若所述差异数据大于预设的变化阈值,则确定所述储物容器中的货物发生变化。
  20. 如权利要求19所述的一种货物订单处理装置,所述的关联图像包括采用下述至少一种方式确定的图像:
    大于所述变化阈值的差异数据对应的货物图像;
    当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定窗口范围内的货物图像;
    当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定时间范围内的货物图像。
  21. 如权利要求18所述的一种货物订单处理装置,所述图像识别模块包括:
    识别码检测模块,用于检测所述关联图像中的货物识别标识,以检测到的所述货物识别标识作为所述货物差异信息。
  22. 如权利要求18所述的一种货物订单处理装置,所述图像识别模块包括:
    货物检测模块,所述货物检测模块包括基于选取的机器学习算法经过样本训练后得到的货物检测模型,用于获取所述关联图像后检测所述关联图像中的货物信息,确定货物的差异信息;或者,用于当对所述关联图像中的货物识别标识识别失败时,检测所述关联图像中的货物信息,确定货物的差异信息。
  23. 如权利要求22所述的一种货物订单处理装置,所述装置还包括:
    人工调度模块,用于在所述关联图像中的货物识检测失败时,生成所述关联图像检测失败的作业任务,将所述作业任务置于任务调度队列中;以及,
    按照获取的网络收银员列表将所述任务调度队列中的作业任务推送给符合作业条件的收银节点,所述网络收银员列表记录有收银节点处理任务的作业状态。
  24. 如权利要求23所述的一种货物订单处理装置,所述装置还包括:
    收银员控制模块,用于查询所述网络收银员列表中是否有与作业任务绑定的网络收银员;
    若有,则将对应的作业任务发送给相应的绑定的网络收银员所在的收银节点进行处理。
  25. 如权利要求24所述的一种货物订单处理装置,所述收银员控制模块确定作业任务所绑定的网络收银员后,还用于查询所述绑定的收银节点的作业状态;以及,
    若所述作业状态为作业量饱和,则将所述作业任务发送给所述网络收银员列表中作业状态为非作业量饱和的网络收银员所在的收银节点进行处理。
  26. 如权利要求23所述的一种货物订单处理装置,所述人工调度模块按照获取的网络收银员列表将所述任务调度队列中的作业任务推送给符合作业条件的收银节点包括:
    将所述作业任务发送给推送给至少两个收银节点:
    相应的,所述图像识别模块基于所述关联图像识别所述储物容器中货物的差异信息包括:根据所述至少两个收银节点的货物识别结果确定所述储物容器 中货物的差异信息。
  27. 如权利要求23所述的一种货物订单处理装置,所述信息反馈模块在所述收银节点的货物识别结果包括货物识别失败时,则执行下述之一:
    向所述客户端返回提示消息,所述提示消息至少包括货物订单无法生成、重新放置货物之一的信息内容;
    生成所述货物识别失败的关联图像对应的未识别货物订单,以及基于所述未识别货物订单的信息设置所述未识别货物订单对应的储物容器的身份验证结果为未通过。
  28. 如权利要求18所述的一种货物订单处理装置,所述装置还包括:
    定位处理模块,用于接收客户端上传的所述储物容器的定位信息,基于所述定位信息确定所述储物容器中货物发生变化的货物归属区间;
    相应的,所述图像识别模块基于所述关联图像识别所述储物容器中货物的差异信息包括:在所述货物归属区间的范围内识别所述关联图像中的货物信息,基于识别出的货物信息确定所述储物容器中货物变化的差异信息。
  29. 一种货物订单处理装置,所述装置包括:
    货物变化检测模块,用于基于图像识别的方式检测储物容器中的货物是否发生变化;
    图像发送模块,用于在检测到货物发生变化时,将所述储物容器中货物发生变化的关联图像发送至服务器;
    信息接收模块,用于接收所述服务器返回的更新后的用户订单信息;
    显示模块,用于展示所述更新后的用户订单信息。
  30. 如权利要求29所述的一种货物订单处理装置,所述货物变化检测模块包括:
    拍摄单元,用于拍摄所述储物容器中的货物图像;
    图像差异处理单元,用于按时间顺序检测所述货物图像之间差异数据;以及,
    货物变化确定模块,用于在所述差异数据大于预设的变化阈值时,确定所述储物容器中的货物发生变化。
  31. 如权利要求30所述的一种货物订单处理装置,所述图像发送模块中的关联图像包括采用下述至少一种方式确定的图像:
    大于所述变化阈值的差异数据对应的货物图像;
    当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定窗口范围内的货物图像;
    当确定所述储物容器中的货物发生变化时,在货物变化时刻对应的货物图像预定时间范围内的货物图像。
  32. 如权利要求29所述的一种货物订单处理装置,所述货物变化检测模块包括:
    变化边界识别模块,用于识别所述关联图像中货物发生变化的货物边界;
    相应的,所述图像发送模块将所述关联图像发送至服务器包括:将识别出的所述货物边界中的图像发送至所述服务器。
  33. 如权利要求29所述的一种货物订单处理装置,所述装置还包括:
    位置定位模块,用于获取所述储物容器的定位信息,将所述定位信息发送给所述服务器,以使所述服务器基于所述定位信息确定所述储物容器中货物的货物归属区间。
  34. 一种服务器,包括至少一个处理器以及用于存储处理器可执行指令的存储器,所述处理器执行所述指令时实现:
    接收客户端上传的关联图像,所述关联图像包括客户端基于图像识别的方式检测到储物容器中货物发生变化时获取的图像;
    基于所述关联图像识别所述储物容器中货物的差异信息;
    利用所述差异信息生成相应的货物订单变更信息;
    根据所述货物订单变更信息更新所述客户端对应的用户订单信息;
    将更新后的用户订单信息发送给所述客户端。
  35. 一种购物终端,包括至少一个处理器以及用于存储处理器可执行指令的存储器,所述处理器执行所述指令时实现:
    基于图像识别的方式检测储物容器中的货物是否发生变化;
    若检测到货物发生变化,则将所述储物容器中货物发生变化的关联图像发送至服务器;
    接收所述服务器返回的更新后的用户订单信息;
    展示所述更新后的用户订单信息。
  36. 一种货物订单处理系统,包括:购物终端、服务器,
    所述服务器包括权利要求18-28中任意一个所述的装置,
    所述购物终端包括权利要求29-33中任意一个所述的购物终端;
    或者,
    所述服务器实现权利要求2-12中任意一个所述的方法的步骤,
    所述购物终端实现权利要3-17中任意一个所述的方法的步骤。
PCT/CN2018/107028 2017-09-25 2018-09-21 一种货物订单处理方法、装置、服务器、购物终端及系统 WO2019057168A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP18857658.1A EP3605431A4 (en) 2017-09-25 2018-09-21 METHOD AND DEVICE FOR PROCESSING ORDERS, SERVER, SHOPPING PORTAL AND SYSTEM
SG11201910119V SG11201910119VA (en) 2017-09-25 2018-09-21 Goods order processing method and apparatus, server, shopping terminal, and system
US16/692,316 US10791199B2 (en) 2017-09-25 2019-11-22 Goods order processing method and apparatus, server, shopping terminal, and system
US16/945,650 US11019180B2 (en) 2017-09-25 2020-07-31 Goods order processing method and apparatus, server, shopping terminal, and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710873473.6 2017-09-25
CN201710873473.6A CN107833083B (zh) 2017-09-25 2017-09-25 一种货物订单处理方法、装置、服务器、购物终端及系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/692,316 Continuation US10791199B2 (en) 2017-09-25 2019-11-22 Goods order processing method and apparatus, server, shopping terminal, and system

Publications (1)

Publication Number Publication Date
WO2019057168A1 true WO2019057168A1 (zh) 2019-03-28

Family

ID=61643417

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/107028 WO2019057168A1 (zh) 2017-09-25 2018-09-21 一种货物订单处理方法、装置、服务器、购物终端及系统

Country Status (6)

Country Link
US (2) US10791199B2 (zh)
EP (1) EP3605431A4 (zh)
CN (1) CN107833083B (zh)
SG (1) SG11201910119VA (zh)
TW (1) TW201915885A (zh)
WO (1) WO2019057168A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111369317A (zh) * 2020-02-27 2020-07-03 创新奇智(上海)科技有限公司 订单生成方法、装置、电子设备及存储介质
CN111523977A (zh) * 2020-04-23 2020-08-11 北京京东振世信息技术有限公司 波次订单集合的创建方法、装置、计算设备和介质
CN114290337A (zh) * 2022-01-28 2022-04-08 北京云迹科技股份有限公司 一种机器人控制方法、装置、电子设备和存储介质

Families Citing this family (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6298562B1 (ja) * 2017-05-31 2018-03-20 伸和コントロールズ株式会社 状態監視装置、状態監視方法及びプログラム
CN107833083B (zh) 2017-09-25 2020-09-04 阿里巴巴集团控股有限公司 一种货物订单处理方法、装置、服务器、购物终端及系统
FR3077404B1 (fr) * 2018-01-31 2022-07-15 Afraite Seugnet Mehdi Methodes et systemes d'assistance a l'achat en un point de vente physique
CN108629911B (zh) * 2018-03-30 2020-11-10 厦门致联科技有限公司 一种颜色可变智慧零售商店及其颜色可变方法
CN108960271B (zh) * 2018-04-13 2021-07-20 湖南大学 一种基于图像识别的餐饮收货诚信规范方法
CN108460584A (zh) * 2018-04-15 2018-08-28 杨国超 一种无人超市管理系统
CN108710897A (zh) * 2018-04-24 2018-10-26 江苏科海智能系统有限公司 一种基于ssd-t的远端在线通用目标检测系统
CN108960038B (zh) * 2018-05-04 2020-11-10 广州图匠数据科技有限公司 一种基于图像识别技术的购物手推车及其识别方法
CN108806128A (zh) * 2018-05-28 2018-11-13 京东方科技集团股份有限公司 货物终端、服务器、购物系统及其购物方法
CN108787486A (zh) * 2018-06-05 2018-11-13 四川九哈科技股份有限公司 基于深度学习的卷烟识别分拣方法及其卷烟识别分拣装置
CN109035558B (zh) * 2018-06-12 2020-08-25 武汉市哈哈便利科技有限公司 一种用于无人售货柜的商品识别算法在线学习系统
CN109145989B (zh) * 2018-08-22 2021-07-06 深圳市东部公共交通有限公司 公交站点布设方法、装置及计算机终端
JP7333688B2 (ja) * 2018-10-18 2023-08-25 東芝テック株式会社 カメラ異常検出装置及びプログラム
CN109367608A (zh) * 2018-12-10 2019-02-22 云南中商正晓农业科技有限公司 一种用于无人超市的自动购物车及商业模式
CN111435511B (zh) * 2019-01-11 2023-06-09 阿里巴巴集团控股有限公司 订单处理方法、装置、设备、系统及可读存储介质
CN111597858B (zh) * 2019-02-20 2023-05-09 阿里巴巴集团控股有限公司 生产图像关联方法、装置及计算机存储介质
CN111860076A (zh) * 2019-04-30 2020-10-30 百度时代网络技术(北京)有限公司 用于生成信息的方法和装置
CN111951063A (zh) * 2019-05-16 2020-11-17 阿里巴巴集团控股有限公司 商品数据处理系统、方法及装置
CN111553776B (zh) * 2020-04-26 2023-08-08 成都新致云服信息技术有限公司 数据处理方法、装置和电子设备
CN111598512A (zh) * 2020-05-14 2020-08-28 陕西秦邦速运物流有限公司 一种物流信息管理系统
CN111724236A (zh) * 2020-06-19 2020-09-29 广州市兴视科信息科技有限公司 一种人工智能导购方法及终端设备
CN111754131A (zh) * 2020-06-30 2020-10-09 苏州朗动网络科技有限公司 企业情报动态监控方法、设备及介质
CN111915251B (zh) * 2020-07-27 2023-08-04 深圳市丰巢科技有限公司 快递柜监控方法、装置、电子设备及存储介质
CN112001349B (zh) * 2020-08-31 2023-09-26 杭州海康威视数字技术股份有限公司 数据审核方法、系统及电子设备
CN112242025A (zh) * 2020-09-30 2021-01-19 苏州艾隆科技股份有限公司 用于发药系统的发药方法及其装置
CN112686731A (zh) * 2021-01-04 2021-04-20 浪潮云信息技术股份公司 一种应用于政务云场景下的服务产品购物车实现方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106781121A (zh) * 2016-12-14 2017-05-31 朱明� 基于视觉分析的超市自助结账智能系统
CN106934692A (zh) * 2017-03-03 2017-07-07 陈维龙 物品信息处理系统、方法及装置
CN107833083A (zh) * 2017-09-25 2018-03-23 阿里巴巴集团控股有限公司 一种货物订单处理方法、装置、服务器、购物终端及系统

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9270841B2 (en) * 2005-04-15 2016-02-23 Freeze Frame, Llc Interactive image capture, marketing and distribution
WO2013134865A1 (en) * 2012-03-16 2013-09-19 In Situ Media Corporation System and method for scanning, tracking and collating customer shopping selections
CN104112216A (zh) * 2013-04-22 2014-10-22 学思行数位行销股份有限公司 存货管理与行销的图像辨识方法
US10268983B2 (en) 2013-06-26 2019-04-23 Amazon Technologies, Inc. Detecting item interaction and movement
US10176456B2 (en) * 2013-06-26 2019-01-08 Amazon Technologies, Inc. Transitioning items from a materials handling facility
US10290031B2 (en) * 2013-07-24 2019-05-14 Gregorio Reid Method and system for automated retail checkout using context recognition
US9473747B2 (en) * 2013-07-25 2016-10-18 Ncr Corporation Whole store scanner
US9262681B1 (en) * 2014-11-19 2016-02-16 Amazon Technologies, Inc. Probabilistic registration of interactions, actions or activities from multiple views
CN104778618A (zh) 2015-04-22 2015-07-15 郭小卫 一种购物储物设备
CN204614025U (zh) * 2015-04-22 2015-09-02 郭小卫 一种购物储物设备
CN104787102B (zh) 2015-04-28 2017-09-05 西安超嗨网络科技有限公司 智能购物车、智能购物系统及其使用方法
US11030604B2 (en) * 2016-02-29 2021-06-08 Signpost Corporation Information processing system
US10248943B2 (en) * 2016-05-18 2019-04-02 Indaflow LLC Object recognition system for checkout lane management
CN106709422A (zh) 2016-11-16 2017-05-24 南京亿猫信息技术有限公司 超市购物车手部识别方法及其识别系统
CN106204226A (zh) * 2016-07-14 2016-12-07 广州极汇信息科技有限公司 自助结账智能购物方法及其装置
US10372998B2 (en) * 2016-08-08 2019-08-06 Indaflow LLC Object recognition for bottom of basket detection
CN106408369B (zh) 2016-08-26 2021-04-06 西安超嗨网络科技有限公司 一种智能鉴别购物车内商品信息的方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106781121A (zh) * 2016-12-14 2017-05-31 朱明� 基于视觉分析的超市自助结账智能系统
CN106934692A (zh) * 2017-03-03 2017-07-07 陈维龙 物品信息处理系统、方法及装置
CN107833083A (zh) * 2017-09-25 2018-03-23 阿里巴巴集团控股有限公司 一种货物订单处理方法、装置、服务器、购物终端及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3605431A4 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111369317A (zh) * 2020-02-27 2020-07-03 创新奇智(上海)科技有限公司 订单生成方法、装置、电子设备及存储介质
CN111369317B (zh) * 2020-02-27 2023-08-18 创新奇智(上海)科技有限公司 订单生成方法、装置、电子设备及存储介质
CN111523977A (zh) * 2020-04-23 2020-08-11 北京京东振世信息技术有限公司 波次订单集合的创建方法、装置、计算设备和介质
CN111523977B (zh) * 2020-04-23 2023-12-08 北京京东振世信息技术有限公司 波次订单集合的创建方法、装置、计算设备和介质
CN114290337A (zh) * 2022-01-28 2022-04-08 北京云迹科技股份有限公司 一种机器人控制方法、装置、电子设备和存储介质
CN114290337B (zh) * 2022-01-28 2024-03-26 北京云迹科技股份有限公司 一种机器人控制方法、装置、电子设备和存储介质

Also Published As

Publication number Publication date
US11019180B2 (en) 2021-05-25
CN107833083B (zh) 2020-09-04
CN107833083A (zh) 2018-03-23
TW201915885A (zh) 2019-04-16
SG11201910119VA (en) 2019-11-28
EP3605431A4 (en) 2020-04-08
US20200092398A1 (en) 2020-03-19
US10791199B2 (en) 2020-09-29
EP3605431A1 (en) 2020-02-05
US20200366762A1 (en) 2020-11-19

Similar Documents

Publication Publication Date Title
WO2019057168A1 (zh) 一种货物订单处理方法、装置、服务器、购物终端及系统
US11132737B2 (en) Dynamic customer checkout experience within an automated shopping environment
US9418352B2 (en) Image-augmented inventory management and wayfinding
TW201945994A (zh) 智慧設備、商品盤點方法、裝置以及設備
US20150269549A1 (en) Synchronizing scan activity with loss prevention cameras
KR20190093733A (ko) 무인 점포에서의 아이템 인식 시스템 및 그 방법
US20180260800A1 (en) Unmanned vehicle in shopping environment
US11379903B2 (en) Data processing method, device and storage medium
KR20190020205A (ko) 구매행위 인식 시스템 및 그 방법
KR102637070B1 (ko) 복수의 센서들을 이용한 물품 판매 시스템 및 그 방법
EP3748565A1 (en) Environment tracking
CA3034834C (en) Sensor-enabled prioritization of processing task requests in an environment
CN111507792A (zh) 一种自助购物方法、计算机可读存储介质及系统
US9779302B1 (en) System for optimizing storage location arrangement
CN110765825A (zh) 物品摆放状态获取方法和系统
JP2023504871A (ja) 不正検出システムおよび方法
US10938890B2 (en) Systems and methods for managing the processing of information acquired by sensors within an environment
US11558539B2 (en) Systems and methods of detecting and identifying an object
US20210201431A1 (en) Dynamically controlled cameras for computer vision monitoring
Li et al. DrtNet: An improved RetinaNet for detecting beverages in unmanned vending machines
CN115840417A (zh) 基于人工智能的目标识别方法、设备及存储介质
US11948044B2 (en) Subregion transformation for label decoding by an automated checkout system
US11842376B2 (en) Method, medium, and system for data lookup based on correlation of user interaction information
CN113269014A (zh) 图像处理方法、设备及存储介质
CN108765798A (zh) 自助购物单元以及系统

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: 18857658

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2018857658

Country of ref document: EP

Effective date: 20191029

NENP Non-entry into the national phase

Ref country code: DE