CN108961559A - Intelligent vending system and its good selling method - Google Patents
Intelligent vending system and its good selling method Download PDFInfo
- Publication number
- CN108961559A CN108961559A CN201810556513.9A CN201810556513A CN108961559A CN 108961559 A CN108961559 A CN 108961559A CN 201810556513 A CN201810556513 A CN 201810556513A CN 108961559 A CN108961559 A CN 108961559A
- Authority
- CN
- China
- Prior art keywords
- user
- learning model
- deep learning
- commodity
- control unit
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F11/00—Coin-freed apparatus for dispensing, or the like, discrete articles
- G07F11/62—Coin-freed apparatus for dispensing, or the like, discrete articles in which the articles are stored in compartments in fixed receptacles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION 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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
Landscapes
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Control Of Vending Devices And Auxiliary Devices For Vending Devices (AREA)
Abstract
The invention discloses a kind of intelligent vending systems, comprising: automatic vending device, background server and deep learning model server, automatic vending device are connect by network with background server, deep learning model server.Automatic vending device includes: control unit, for carrying out data processing and management；Bio-identification unit obtains user information；Camera unit acquires the static images of surplus commodities or the real-time dynamic video of user's shopping in automatic vending device；Relay lock, for latching or unlocking the cabinet door of automatic vending device.Deep learning model server identifies this surplus commodities information for analyzing static images or real-time dynamic video.Background server is used to lack merchandise news according to this surplus commodities information and the calculating of last time surplus commodities information.Beneficial effects of the present invention: commodity identification is carried out using intelligent depth learning art, identification quickly, accurately, is suitble to a variety of shopping scenes and shopping process.
Description
Technical field
The present invention relates to the technical field of unmanned retail, in particular to a kind of intelligent vending system and its good selling method.
Background technique
Self-service machine is that popular nobody is newly sold carrier on the market, it is serviced for round-the-clock 24 hours, it is only necessary to
Having the place of network can use the equipment to carry out commercial articles vending, without keeping an eye on, without antitheft, greatly save manually at
This, facilitates the various features such as user's shopping.
Currently, on the market there are many kinds of self-service machine, major part by way of scanning the two-dimensional code or inserting coins into
Row purchase.Four major class can be divided into according to working principle:
1, the commodity of commission are placed in specific equipment, after user selects certain commodity, by inserting coins or scanning
Form of payment starting purchase operation, which pops up the commodity of purchase to come by simple mechanical spring.Its disadvantage exists
In: type of merchandize is seldom, can only sell bottled commodity (such as mineral water, pop can, beverage), not be able to satisfy client's multiple types
Demand for commodity.
2, commodity are placed in many fixed grid, judge whether customer buys by weighing principle, user sweeps
Commodity are taken out after code, the signal jump that weighing system receives, so that it is determined that whether user buys the commodity inside the grid, and
Push order.Its shortcoming is that: commodity biggish for volume ratio are difficult to place among sub-box, and cost is relatively high, it is difficult to push away
Extensively；Quantity, the type for being put into commodity are less, are not able to satisfy the demand of extensive stock.
3, it is based on the unmanned selling cabinet of RFID identification technology, each commodity is posted into RFID tag, each label carries
The goods number information.Trade company many commodity can be placed on one nobody sell in cabinet body, user pass through barcode scanning or scanning
Face characteristic information is opened the door, self-help shopping.The advantages of this unmanned selling cabinet is can to put many different types of quotient
Product, shopping are simple.But also have many problems, and such as: each RFID tag higher cost, label are not labelled to suitably
Position is easy to appear more readings or skip problem；It is the thing of very lost labor to each commodity patch RFID tag, works as restocking
When new commodity, very labor intensive is labelled in place to each commodity.Since discrimination is not high, labeling
Complexity influences the popularization of unmanned cabinet-type air conditioner naturally.
4, based on traditional unmanned selling cabinet of image recognition technology, which passes through detection by placing some cameras
The position of each layer of commodity, judges whether commodity move or whether be removed, and implements and needs to consider many shopping
Scene, technology are complicated.The disadvantage is that: discrimination is low, and the commodity of different layer frames cannot mutually replace position, does not conform to there can be no some
The shopping process of reason.Therefore, the strong influence experience sense of user.
Summary of the invention
In view of the problems of the existing technology, the main object of the present invention is to provide a kind of intelligent vending system, to mention
High user's shopping experience.
To achieve the above object, intelligent vending system proposed by the present invention, the system include: automatic vending device, after
Platform server and deep learning model server, the automatic vending device pass through network and the background server, described
The connection of deep learning model server；
The automatic vending device includes: control unit, for carrying out data processing and management；Bio-identification unit is used
Family obtains user information and the user information that will acquire is sent to control unit；Shooting unit, for acquire the cabinet body this
The static images of surplus commodities or the real-time dynamic video of user's shopping, and the static images of acquisition or real-time dynamic vision are taken place frequently
It send to control unit；Relay lock, for latching or unlocking the cabinet door of automatic vending device；
Deep learning model server, the static images or real-time dynamic video sent for reception control unit；It is described
The first deep learning model is equipped in deep learning model server, for the commodity in static images or real-time dynamic video
Information is identified, and the merchandise news of identification is sent to control unit；
Or the first deep learning model is deployed in control unit, control unit is by the first deep learning model to quiet
Merchandise news in state picture or real-time dynamic video is identified；
Background server, for carrying out pair the merchandise news of this surplus commodities and the merchandise news of last time surplus commodities
Than calculating the merchandise news for lacking commodity and the merchandise news for lacking commodity being sent to control unit；
Described control unit is also used to judge whether the user information can be by verifying, if it can, then by door open command
It is sent to relay lock, relay lock is made to unlock and open cabinet door；Described control unit is also used to according to the commodity for lacking commodity
Information generates corresponding purchase order, and is sent to the purchase order and user terminal by background server.
Preferably, the bio-identification unit includes: recognition of face device, Fingerprint Identification Unit and iris recognition device.
Preferably, the camera unit is big wide-angle camera.
Further, the camera unit is high-definition camera.
Preferably, the deep learning model server is equipped with the second deep learning model, for analyzing the shopping of user
Operation, judges the legitimacy of user's shopping operation, and judging result is sent to control unit；
Or the second deep learning model is deployed in control unit, control unit passes through the second deep learning model point
The shopping operation for analysing user, judges the legitimacy of user's shopping operation, and generates corresponding judging result；
The automatic vending device includes warning prompt device, for being generated according to the judging result of the second deep learning model
And issue prompt information.
The present invention also proposes that a kind of good selling method of intelligent vending system, the good selling method include the following steps:
S1 acquires user information；
S2 judges whether the user information can be by verifying, if it can, then sending door open command to automatic vending device
Relay lock, make relay lock unlock and open cabinet door；
S3 sends door signal to the control unit of automatic vending device, control if the trigger Receiving collision of relay lock
Unit processed locks a door to instruct according to door signal transmission to be locked to relay, and relay is locked a door；
S4, the static images of this interior surplus commodities of acquisition automatic vending device or the real-time dynamic vision of user's shopping
Frequently；
S5 identifies this surplus commodities by the static images of first deep learning model analysis this surplus commodities
Merchandise news；Or the real-time dynamic video done shopping by the first deep learning model analysis user, it calculates user and takes out commodity
Merchandise news, go to step S7；
S6, according to the merchandise news of the merchandise news of this surplus commodities and last time surplus commodities, calculating lacks commodity
Merchandise news；
S7 generates corresponding electronic business transaction order according to lacking merchandise news or taking out the information of commodity, and by the electricity
Sub- purchase order is sent to and user terminal.
Specifically, in step sl, the user information includes: face information, finger print information and iris information.
Preferably, in step s 4, the static images of this surplus commodities or user's purchase in acquisition automatic vending device
Further include the shopping operation for analyzing user after the real-time dynamic video of object, judges the legitimacy of user's shopping operation and generation is sentenced
Break as a result, and sending prompt information to user according to judging result.
Technical solution of the present invention acquires the static images of surplus commodities in automatic vending device by camera unit, then
Static images are identified by the deep learning model in deep learning model server, this residue is calculated by background server
Merchandise news lacks the merchandise news of commodity relative to last time surplus commodities information, i.e., front and back is opened the door twice lacks the commodity of commodity
Information, control unit generate corresponding electronic business transaction order according to the merchandise news for lacking commodity, then by background server to
User terminal sends electronic business transaction order；
Or the real-time dynamic video of user's shopping is acquired by camera unit, by deep learning model server
Deep learning model identifies that user in real-time dynamic video takes out the merchandise news of commodity, and control unit takes out quotient according to user
The merchandise news of product generates corresponding electronic business transaction order, then sends electronic business transaction to user terminal by background server and order
It is single.
The beneficial effects of the present invention are: commodity identification is carried out using artificial intelligence deep learning model, need to only be adopted in advance
Deep learning model is trained with a large amount of commodity picture or dynamic video, the feature extraction processing without carrying out additional can be known
Merchandise news in other picture, can quickly, accurately identify in the static images or real-time dynamic video of commodity commodity letter
Breath；Also, either commodity part stack or commodity placement angle it is inclined it is equal can accurately identify commodity, be suitble to a variety of shopping
Scene and shopping process can effectively improve the experience sense of user.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
The structure shown according to these attached drawings obtains other attached drawings.
Fig. 1 is the composition block diagram of the present invention one embodiment of intelligent vending system；
Fig. 2 is the cabinet body structure schematic diagram of automatic vending device；
Fig. 3 is the composition block diagram of control unit；
Fig. 4 is the flow chart of an embodiment of the good selling method of the intelligent vending system of the present invention；
The object of the invention is realized, the embodiments will be further described with reference to the accompanying drawings for functional characteristics and advantage.
Specific embodiment
The present invention proposes a kind of intelligent vending system.
Referring to Fig.1, Fig. 1 is the structural schematic diagram of the present invention one embodiment of intelligent vending system.
As shown in Figure 1, in embodiments of the present invention, which includes: automatic vending device, background service
Device 200 and deep learning model server 300, automatic vending device pass through network and background server 200, deep learning mould
Type server 300 connects.
As Figure 2-3, in the present embodiment, which includes cabinet body 110, control unit 120, camera shooting list
Member 130, bio-identification unit 140 and relay lock 150.
As shown in Fig. 2, being equipped with several layer frames 111 setting up and down in cabinet body 110, commodity 400 are placed in layer frame 111,
The top of each layer frame 111 is equipped with camera unit 130, for acquiring all commodity of the corresponding layer frame 111 of the camera unit 130
Information.Specifically, which uses big wide-angle camera, shoots quotient all in entire layer frame 111 to meet
Product.It should be noted that each corresponding camera unit 130 of layer frame 111 is according to cabinet body 110 at a distance from the layer frame 111
Depending on height and 111 numbers of layer frame.In addition, the clarity of the still photo in order to guarantee the acquisition of camera unit 130, the camera shooting list
Member 130 be high-definition camera, the picture or video of tri- kinds of Color Channels of RGB can be shot, and have anti-fog (heating or other
Defrosting) etc. functions.
As shown in figure 3, control unit 120 is a kind of equipment with operating system, such as Android device, ios device, PC
Equipment, WinPhone equipment etc. comprising user information authentication module 121, data processing module 122 and relay lock control
Molding block 123.User information authentication module 121 is for receiving user information, and by the user information received and background service
User information in device 200 is matched, to verify user information.Data processing module 122 is used to handle data information, such as
Handle camera unit 130 acquire commodity static images or user shopping real-time dynamic video, generate electronic business transaction bill,
Or issue alarm request etc..Relay lock control module 123 can control the switch of relay lock 150, such as receives enabling and refer to
Relay lock 150 is then opened in order, if the trigger Receiving collision of relay lock 150, trigger send door signal relay
Device lock control module 123 enters locking state by the control of relay lock control module 123, refers to if not being received again by enabling
It enables, then the cabinet door of automatic vending device cannot be opened.
Bio-identification unit 140 is arranged on cabinet body 110, and the user information for obtaining user information and will acquire is sent
To control unit 120.The bio-identification unit 140 can be the screen (including touch screen and non-touch screen) of certain size,
It can be the equipment for having operating system, in order to trade company's development and maintenance.The bio-identification unit 140 is known including face
One of other device, Fingerprint Identification Unit and iris recognition device are a variety of, and face, fingerprint or iris for acquiring user etc. are living
Property human body information.
Specifically, the process of the acquisition user information of the bio-identification unit 140 is as follows:
User stand before cabinet body 110, by bio-identification unit 140 obtain user information (such as shooting people
Face information, acquisition finger print information, acquisition iris information), and the user information that will acquire is sent to user information authentication module
121, user information authentication module 121 matches the user information received with the user information in background server 200,
Judge whether user has registered, determines whether the user has permission using the automatic vending device.
If judging, the user is not registered, and bio-identification unit 140 provides wechat or Alipay or movement on the screen
Tri- kinds of different modes of APP guide client to provide the information such as face, fingerprint, iris and register, and these information are uploaded to
Background service carries out feature information extraction；
If judging, the user has been registered, relay lock control module 123 send door open command to relay lock 150 into
Row is unlocked, and user, that is, openable cabinet body 110 cabinet door is done shopping.
Background server 200 is for storing registration user information, calculating the merchandise news for lacking commodity and lacking commodity
Merchandise news be sent to data processing module 122, corresponding electronic business transaction order is generated by data processing module 122, and will
The electronic business transaction order of generation is by network push to user terminal.
Deep learning model server 300 includes the first identification module, is deployed in first identification module trained
The first good deep learning model, this surplus commodities that the first deep learning model is used to be sent according to control unit 120
The static images real-time dynamic video that perhaps user does shopping carrys out the commodity in discriminance analysis static images or real-time dynamic video
Information, and the merchandise news of identification is sent to control unit 120；Or it is trained first deep learning model is direct
It is deployed to control unit 120, the static images of first deep learning model analysis this surplus commodities are passed through by control unit 120
Or dynamic video, identify this surplus commodities information.
Wherein, the training process of the first deep learning model is as follows:
1, largely acquisition commodity in layer frame 111 under different placement states (such as part commodity tilt or partly overlap)
Static images or real-time dynamic video construct commodity test set；During constructing commodity test set, by each commodity
Carry out labeling, it is indicated that there are which commodity or point out commodity in the position of picture, so that depth in the picture or video
The commodity in picture or video can be identified by practising model, generate better parameter.By the commodity for acquiring different placement states
Still photo or dynamic video may make trained model to have better generalization ability.
2, deep learning model (such as convolution mind is created by existing deep learning frame (such as TensorFlow, Caffe)
Through network model), the number of plies of learning rate and deep learning is set, so that the loss function considerable convergence of deep learning model,
Reach the loss index of a higher discrimination and very little.
3, by commodity test set training deep learning model created, the parameter of the deep learning model is adjusted, is made
Model parameter is optimal, to obtain the first deep learning model.
It should be noted that deep learning model training can be realized by existing training method, then not superfluous in detail herein
It states.
In embodiments of the present invention, which further includes the second identification module, second identification
There are trained second deep learning model, this residue of the second deep learning model for basis in inside modules administration
The static images of commodity or the real-time dynamic video of user's shopping judge that user does shopping and grasp to analyze the shopping operation of user
The legitimacy of work, and judging result is sent to control unit 120；Or trained second deep learning model is sent
To control unit 120, is operated by control unit 120 by the shopping of the second deep learning model analysis user, judge that user purchases
The legitimacy of object operation, and generate corresponding judging result.Correspondingly, automatic vending device includes warning prompt device 160, the police
Prompt information can be generated and issued to user terminal according to the judging result of the second deep learning model by accusing prompting device 160.As a result,
When user's shopping operation does not meet normal Shopping Behaviors, user can be prompted to abide by shopping specification.
By taking static images as an example, the building process of the second deep learning model is as follows:
1, a large amount of shopping process lack of standardization is simulated, and acquires the static images after lack of standardization or illegal operation；
2, labeling processing is carried out to the static images after the lack of standardization or illegal operation of acquisition, constructs non-standard operation
Pictures；
3, construct convolutional neural networks model, set learning rate, the deep learning number of plies, training convolutional neural networks model with
And parameter, until convolutional neural networks model is greater than or equal to the discrimination of the static images after lack of standardization or illegal operation
The discrimination of setting saves the convolutional neural networks model and parameter, and generates the second deep learning model.
Specifically, in the present embodiment, when the shopping operation of the second deep learning model identification user, three kinds are produced
Judging result is respectively as follows:
1, judging result is normal operating, and the data processing module 122 of control unit 120 sends null message to warning at this time
Prompting device 160 after warning prompt device 160 receives the null message, is operated without alarm.
2, judging result is non-standard operation (such as stacking part commodity), and Data Data handles mould module to warning at this time
Prompting device 160 issues alarm request and carries out alarm operation, and issues prompt request to background server 200, by background server
200 send non-standard operation prompt to user terminal, and user is allowed to change the problem of non-standard operation is left.
3, judging result is criminal manipulation (as destroyed commodity, firmly beaing damage cabinet body 110), and Data Data is handled at this time
Mould module issues alarm request to warning prompt device 160 and carries out alarm operation, to warn user to stop criminal manipulation, while will tool
The user information and photo site or dynamic video of body, which are sent in background server 200, to be saved, as alarm material.
The present invention also proposes a kind of good selling method of intelligent vending system.
Referring to the flow chart of an embodiment of the good selling method that Fig. 4, Fig. 4 are the intelligent vending system of the present invention.
As shown in figure 4, the good selling method includes the following steps:
S1 is acquired any in face information, finger print information and the iris information of user by bio-identification unit 140
It is one or more；
S2, the user information that will acquire issue control unit 120, and control unit 120 is by the user information and background service
User information carries out matching verifying in device 200, if matching is proved to be successful, control unit 120 sends door open command to automatic selling
The relay lock 150 of goods device, makes relay lock 150 unlock and open cabinet door, user buys goods；
S3, if user buy goods after close the door, collision relay lock 150 trigger, trigger send door signal to
Control unit 120 locks a door instruction to relay lock 150,150 lock of relay lock according to door signal transmission by control unit 120
Door；
S4 starts camera unit 130, the static images of this surplus commodities or user's purchase in acquisition automatic vending device
The real-time dynamic video of object；
S5 passes through the first deep learning model analysis sheet in deep learning model server 300 or control unit 120
The static images of secondary surplus commodities identify this surplus commodities quantity and information；Or directly pass through the first deep learning mould
The real-time dynamic video of type analysis user shopping, identification user take out the quantity and information of commodity, and go to step S7；
The information of this surplus commodities of identification is uploaded to background server 200 by S6, right by background server 200
Than the merchandise news of this surplus commodities and the merchandise news of last time surplus commodities, the type and quantity letter for lacking commodity is calculated
Breath；
S7 generates corresponding electronic business transaction according to the merchandise news for lacking commodity or the merchandise news for taking out commodity and orders
It is single, and the electronic business transaction order is sent to and user terminal.
When user's shopping operation does not meet normal Shopping Behaviors, need that user is prompted to abide by shopping specification, therefore, therefore in step
In rapid S4, further includes the real-time dynamic video done shopping according to the static images of this surplus commodities or user, analyze user's
Shopping operation judges the legitimacy of user's shopping operation and generates judging result；According to judging result send prompt information to
Family terminal.
In the present embodiment, judging result is roughly divided into three kinds, is respectively as follows:
1, judging result is normal operating, and the data processing module 122 of control unit 120 sends null message to warning at this time
Prompting device 160 after warning prompt device 160 receives the null message, is operated without alarm.
2, judging result is non-standard operation (such as stacking part commodity), and Data Data handles mould module to warning at this time
Prompting device 160 issues alarm request and carries out alarm operation, and issues prompt request to background server 200, by background server
200 send non-standard operation prompt to user terminal, and user is allowed to change the problem of non-standard operation is left.
3, judging result is criminal manipulation (as destroyed commodity, firmly beaing damage cabinet body 110), and Data Data is handled at this time
Mould module issues alarm request to warning prompt device 160 and carries out alarm operation, to warn user to stop criminal manipulation, while will tool
The user information and photo site or dynamic video of body, which are sent in background server 200, to be saved, as alarm material.
The above description is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all at this
Under the inventive concept of invention, using equivalent structure transformation made by description of the invention and accompanying drawing content, or directly/use indirectly
It is included in other related technical areas in scope of patent protection of the invention.
Claims (8)
1. a kind of intelligence vending system, which is characterized in that the system includes: automatic vending device, background server and depth
Learning model server is spent, the automatic vending device is taken by network and the background server, the deep learning model
Business device connection；
The automatic vending device includes: control unit, for carrying out data processing and management；Bio-identification unit, user obtain
It takes user information and the user information that will acquire is sent to control unit；Shooting unit, for acquiring the cabinet body, this is remaining
The static images of commodity or the real-time dynamic video of user's shopping, and the static images of acquisition or real-time dynamic video are sent to
Control unit；Relay lock, for latching or unlocking the cabinet door of automatic vending device；
Deep learning model server, the static images or real-time dynamic video sent for reception control unit；The depth
The first deep learning model is equipped in learning model server, for the merchandise news in static images or real-time dynamic video
It is identified, and the merchandise news of identification is sent to control unit；
Or the first deep learning model is deployed in control unit, control unit is by the first deep learning model to static map
Merchandise news in piece or real-time dynamic video is identified；
Background server, for the merchandise news of the merchandise news of this surplus commodities and last time surplus commodities to be compared,
It calculates the merchandise news for lacking commodity and the merchandise news for lacking commodity is sent to control unit；
Described control unit is also used to judge whether the user information can be by verifying, if it can, then sending door open command
It is locked to relay, relay lock is made to unlock and open cabinet door；Described control unit is also used to according to the merchandise news for lacking commodity
It generates corresponding purchase order, and is sent to the purchase order and user terminal by background server.
2. intelligence vending system as described in claim 1, which is characterized in that the bio-identification unit includes: that face is known
Other device, Fingerprint Identification Unit and iris recognition device.
3. intelligence vending system as described in claim 1, which is characterized in that the camera unit is big wide-angle camera.
4. intelligence vending system as claimed in claim 3, which is characterized in that the camera unit is high-definition camera.
5. intelligence vending system according to any one of claims 1-4, which is characterized in that the deep learning model service
Device is equipped with the second deep learning model, and the shopping for analyzing user operates, and judges the legitimacy of user's shopping operation, and will sentence
Disconnected result is sent to control unit；
Or the second deep learning model is deployed in control unit, control unit is used by the second deep learning model analysis
The shopping at family operates, and judges the legitimacy of user's shopping operation, and generates corresponding judging result；
The automatic vending device includes warning prompt device, concurrent for being generated according to the judging result of the second deep learning model
Prompt information out.
6. a kind of good selling method of intelligence vending system, which is characterized in that the good selling method includes the following steps:
S1 acquires user information；
S2, judge the user information whether can by verifying, if it can, then send door open command to automatic vending device after
Electrical apparatus lock makes relay lock unlock and open cabinet door；
S3 sends door signal to the control unit of automatic vending device, control is single if the trigger Receiving collision of relay lock
Member locks a door to instruct according to door signal transmission locks to relay, and relay is locked a door；
S4, the static images of this interior surplus commodities of acquisition automatic vending device or the real-time dynamic video of user's shopping；
S5 identifies the commodity of this surplus commodities by the static images of first deep learning model analysis this surplus commodities
Information；Or the real-time dynamic video done shopping by the first deep learning model analysis user, calculate the commodity that user takes out commodity
Information, go to step S7；
S6 calculates the commodity for lacking commodity according to the merchandise news of the merchandise news of this surplus commodities and last time surplus commodities
Information；
S7 generates corresponding electronic business transaction order according to the information for lacking merchandise news or taking-up commodity, and the electronics is purchased
Object order is sent to and user terminal.
7. intelligence vending system as claimed in claim 6, which is characterized in that in step sl, the user information includes:
Face information, finger print information and iris information.
8. the good selling method of intelligence vending system as claimed in claim 6, which is characterized in that in step s 4, acquisition is certainly
It further include the purchase for analyzing user after the dynamic video that the static images of this surplus commodities or user do shopping in dynamic goods selling equipment
Object operation, judge user do shopping operation legitimacy and generate judging result, and according to judging result send prompt information to
Family.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810556513.9A CN108961559A (en) | 2018-06-01 | 2018-06-01 | Intelligent vending system and its good selling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810556513.9A CN108961559A (en) | 2018-06-01 | 2018-06-01 | Intelligent vending system and its good selling method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108961559A true CN108961559A (en) | 2018-12-07 |
Family
ID=64492999
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810556513.9A Pending CN108961559A (en) | 2018-06-01 | 2018-06-01 | Intelligent vending system and its good selling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108961559A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109461004A (en) * | 2018-12-28 | 2019-03-12 | 南京美果电子商务有限公司 | A kind of self-service selling system with intelligent marketing function |
| CN111222870A (en) * | 2019-01-24 | 2020-06-02 | 图灵通诺（北京）科技有限公司 | Settlement method, device and system |
| CN113496442A (en) * | 2020-03-19 | 2021-10-12 | 荷盛崧钜智财顾问股份有限公司 | Graph representation generation system, graph representation generation method and graph representation intelligent module thereof |
| WO2022190061A1 (en) * | 2021-03-11 | 2022-09-15 | Rk.Ai - Serviços De Processamento De Imagens E Análise De Dados Lda. | Storage cabinet, methods and uses thereof |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000207452A (en) * | 1999-01-13 | 2000-07-28 | Tsubasa System Co Ltd | System and method for promoting used car distribution |
| JP2001076007A (en) * | 1999-07-07 | 2001-03-23 | Matsushita Electric Ind Co Ltd | Data-managing method/data managing system and device used for the same |
| CN101226664A (en) * | 2008-02-02 | 2008-07-23 | 北京海鑫科金高科技股份有限公司 | Intelligent monitoring system and method for self-help bank and ATM |
| CN106781017A (en) * | 2017-03-07 | 2017-05-31 | 深圳市楼通宝实业有限公司 | Self-service vending method and system |
| CN107134053A (en) * | 2017-04-19 | 2017-09-05 | 石道松 | Intelligence is sold goods shops |
| CN107729923A (en) * | 2017-09-19 | 2018-02-23 | 浙江师范大学 | A kind of commodity Computer Vision Recognition method based on TensorFlow |
| CN107895306A (en) * | 2017-12-01 | 2018-04-10 | 深圳市迪比科电子科技有限公司 | Commodity combination identification method, commodity combination identification system and commodity combination selling system |
-
2018
- 2018-06-01 CN CN201810556513.9A patent/CN108961559A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000207452A (en) * | 1999-01-13 | 2000-07-28 | Tsubasa System Co Ltd | System and method for promoting used car distribution |
| JP2001076007A (en) * | 1999-07-07 | 2001-03-23 | Matsushita Electric Ind Co Ltd | Data-managing method/data managing system and device used for the same |
| CN101226664A (en) * | 2008-02-02 | 2008-07-23 | 北京海鑫科金高科技股份有限公司 | Intelligent monitoring system and method for self-help bank and ATM |
| CN106781017A (en) * | 2017-03-07 | 2017-05-31 | 深圳市楼通宝实业有限公司 | Self-service vending method and system |
| CN107134053A (en) * | 2017-04-19 | 2017-09-05 | 石道松 | Intelligence is sold goods shops |
| CN107729923A (en) * | 2017-09-19 | 2018-02-23 | 浙江师范大学 | A kind of commodity Computer Vision Recognition method based on TensorFlow |
| CN107895306A (en) * | 2017-12-01 | 2018-04-10 | 深圳市迪比科电子科技有限公司 | Commodity combination identification method, commodity combination identification system and commodity combination selling system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109461004A (en) * | 2018-12-28 | 2019-03-12 | 南京美果电子商务有限公司 | A kind of self-service selling system with intelligent marketing function |
| CN111222870A (en) * | 2019-01-24 | 2020-06-02 | 图灵通诺（北京）科技有限公司 | Settlement method, device and system |
| CN111222870B (en) * | 2019-01-24 | 2024-02-27 | 图灵通诺（北京）科技有限公司 | Settlement method, device and system |
| CN113496442A (en) * | 2020-03-19 | 2021-10-12 | 荷盛崧钜智财顾问股份有限公司 | Graph representation generation system, graph representation generation method and graph representation intelligent module thereof |
| WO2022190061A1 (en) * | 2021-03-11 | 2022-09-15 | Rk.Ai - Serviços De Processamento De Imagens E Análise De Dados Lda. | Storage cabinet, methods and uses thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108269371B (en) | Automatic commodity settlement method and device and self-service cash register | |
| CN108320404B (en) | Commodity identification method and device based on neural network and self-service cash register | |
| CN108491799B (en) | Intelligent sales counter commodity management method and system based on image recognition | |
| CN108053204B (en) | Automatic settlement method and selling equipment | |
| CN206179114U (en) | Commodity lease machine | |
| US10800440B2 (en) | Smart shopping carts and corresponding methods thereof, and smart shopping systems | |
| WO2019174553A1 (en) | Automatic vending method and apparatus | |
| CN108961559A (en) | Intelligent vending system and its good selling method | |
| CN108564420B (en) | Intelligent retail system network | |
| CN107862515A (en) | A kind of good selling method and system | |
| CN108416909A (en) | A kind of intelligence sales counter and intelligent good selling method | |
| CN106384442A (en) | Commodity lease machine | |
| CN108520409B (en) | Rapid checkout method and device and electronic equipment | |
| CN107146325A (en) | Automatic vending machine and its control method, control device and system | |
| CN107464116A (en) | A kind of order settlement method and system | |
| CN108647671B (en) | Optical identification visual identification method and unmanned sales counter based on same | |
| EP3901841B1 (en) | Settlement method, apparatus, and system | |
| US20200065791A1 (en) | Self-service product return using computer vision and artificial intelligence | |
| CN111222870B (en) | Settlement method, device and system | |
| CN108038693A (en) | The consistent method and system of automatic vending machine identification adult customer's testimony of a witness | |
| CN109447756A (en) | A kind of unattended supermarket shopping method and system | |
| CN108510673B (en) | Automatic radio frequency quick checkout method and device and electronic equipment | |
| CN110164032A (en) | Automatic vending machine and its data processing method and automatically vending system | |
| CN209132863U (en) | A kind of self-service cabinet | |
| CN109948966A (en) | A kind of intelligent article access method based on face recognition technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181207 |