CN110164033A - Merchandise news extracting method, merchandise news extraction element and automatically vending system - Google Patents
Merchandise news extracting method, merchandise news extraction element and automatically vending system Download PDFInfo
- Publication number
- CN110164033A CN110164033A CN201810150043.6A CN201810150043A CN110164033A CN 110164033 A CN110164033 A CN 110164033A CN 201810150043 A CN201810150043 A CN 201810150043A CN 110164033 A CN110164033 A CN 110164033A
- Authority
- CN
- China
- Prior art keywords
- commodity
- image
- information
- merchandise news
- automatic vending
- 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
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000000605 extraction Methods 0.000 title claims abstract description 38
- 238000004806 packaging method and process Methods 0.000 claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 10
- 238000009826 distribution Methods 0.000 claims abstract description 4
- 239000000047 product Substances 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 9
- 238000005303 weighing Methods 0.000 claims description 9
- 230000000007 visual effect Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 5
- 230000000153 supplemental effect Effects 0.000 claims description 4
- 239000013589 supplement Substances 0.000 claims description 2
- 230000008859 change Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 8
- 230000002159 abnormal effect Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000013523 data management Methods 0.000 description 3
- 230000033001 locomotion Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 235000013361 beverage Nutrition 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004321 preservation Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000035622 drinking Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 238000005242 forging Methods 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 239000010985 leather Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005057 refrigeration Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000013585 weight reducing agent Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F9/00—Details other than those peculiar to special kinds or types of apparatus
- G07F9/002—Vending machines being part of a centrally controlled network of vending machines
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Control Of Vending Devices And Auxiliary Devices For Vending Devices (AREA)
Abstract
The present invention provides a kind of merchandise news extracting method, merchandise news extraction element and automatically vending systems.Wherein merchandise news extracting method includes: the image for obtaining the commodity of information to be extracted, and image at least reflects the appearance of commodity and the label of commodity packaging;Image recognition is carried out to the label comprising commodity packaging, to identify the label information of commodity;Obtain the weight of the commodity of information to be extracted;Image is inputted preset commodity identification model as training sample to be trained, and is training result distribution sale mark;Label information, commodity weight and sale mark matching are saved, to obtain the sale characteristic of commodity.The solution of the present invention is automatically performed automatically extracting for commodity selling characteristic, more efficient, meets the requirement of automatically vending system.
Description
Technical field
The present invention relates to smart machine, more particularly to merchandise news extracting method, merchandise news extraction element and automatic
Vending system.
Background technique
The fast-developing upgrading for pushing transmission industry of information technology.Retail trade is also faced with quick change at present
Leather.It is to rely on internet, by using the advanced technologies means such as big data, artificial intelligence, production, circulation and pin to commodity
It sells process and carries out the new retail of upgrading just as current hot spot.
Automatic vending machine is to realize to be sold unmanned important equipment, and current automatic vending machine is mainly to intelligent network
The direction of change is developed.In order to meet the consumption demand of user, it is desirable that the type of merchandize that automatic vending machine can be sold is abundant, therefore
Automatically vending system needs handle a large amount of merchandise news.The specification of the newly-increased type of merchandise or commodity pack into
When row upgrading, merchandise news is needed to re-start acquisition, existing commodity acquisition system needs backstage manager's craft typing
A large amount of information, efficiency is lower, is easy to appear typing mistake, is not able to satisfy the demand for development of current automatically vending system.
Summary of the invention
It is an object of the present invention to provide a kind of more efficient merchandise news extracting method and device.
A further object of the present invention is to provide for automatic vending machine and automatically vending system and complete automatic vending function
Merchandise news needed for energy.
Particularly, the present invention provides a kind of merchandise news extracting methods, this method comprises: obtaining the quotient of information to be extracted
The image of product, image at least reflect the appearance of commodity and the label of commodity packaging;Figure is carried out to the label comprising commodity packaging
As identification, to identify the label information of commodity;Obtain the weight of the commodity of information to be extracted;It is defeated using image as training sample
Enter preset commodity identification model to be trained, and is training result distribution sale mark;By label information, commodity weight and
Sale mark matching saves, to obtain the sale characteristic of commodity.
Optionally, the step of obtaining the image of the commodity of information to be extracted includes: that control filming apparatus claps commodity
It takes the photograph;Feature extraction is carried out to the image that shooting obtains, and whether is accorded with according to the image that the characteristics of image of extraction judges that shooting obtains
Close the image request of commodity identification model;Retain the image for meeting image request.
Optionally, in the case where the image that shooting obtains does not meet the image request of commodity identification model further include: adjust
The acquisition parameters of whole filming apparatus re-start shooting, and re-start feature extraction.
Optionally, the process that control filming apparatus shoots commodity includes: that the pallet of control carrying commodity gradually revolves
Turn, and after the rotation of every step, control filming apparatus shooting is primary, to obtain the image at the multiple visual angles of commodity;From multiple visual angles
Image in select the appearance and commodity packaging that can reflect commodity label image.
Optionally, after the label information for identifying commodity the step of further include: information needed for obtaining merchandising
Type;Whether the label information identified according to information type judgement is complete, if imperfect, obtains the supplemental information of typing with right
Label information is supplemented.
Optionally, after saving the step of selling characteristic further include: to automatic vending machine and/or automatic selling cargo net
Network service equipment provides commodity identification model, so that it is known using selling images of the commodity identification model to automatic vending machine
Not, to determine sold commodity.
Optionally, after saving the step of selling characteristic further include: to automatic vending machine and/or automatic selling cargo net
Network service equipment provide commodity weight, with using commodity weight to the commodity sold determined using commodity identification model into
Row verifying.
Optionally, after saving the step of selling characteristic further include: to automatic vending machine and/or automatic selling cargo net
Network service equipment provides label information, to carry out settlement of transactions to the commodity sold using label information.
According to another aspect of the present invention, a kind of merchandise news extraction element is additionally provided comprising: rotary-tray,
Commodity for support information to be extracted;At least one filming apparatus is set to the periphery of rotary-tray, for carrying out to commodity
Shooting;Weighing device is set to the lower section of rotary-tray, to weigh to commodity;And controller, controller include storage
Device and processor, wherein memory is stored with computer program, is able to carry out when computer program is executed by processor above-mentioned
Merchandise news extracting method.
According to another aspect of the present invention, a kind of automatically vending system is additionally provided, on which includes:
State merchandise news extraction element;It is true to be configured with the sale characteristic that merchandise news extraction element obtains for automatic vending machine
Make the commodity sold;And network server device, data connection is distinguished with merchandise news extraction element and automatic vending machine, is used
It is settled accounts in the commodity of sale.
Merchandise news extracting method of the invention and device learn commodity image using machine learning algorithm, instruction
Practise commodity identification model.And by identifying that the label image of commodity packaging obtains the label information of commodity, by commodity
Weighing obtains the weight of commodity.Label information, commodity weight and sale mark matching save, so that the sale for forming commodity is special
Levy data, for automatic vending machine in vending articles by the sale image of shooting, determine the commodity that user takes out, utilize quotient
Product weight is verified to article is picked and placed, and carries out Automatic-settlement accordingly, and whole process carries out automatically, high-efficient, can be fast
Merchandise sales information needed for obtaining automatic vending machine fastly.
Further, merchandise news extracting method of the invention and device, high degree of automation can automatically snap image
The image of the required each angle of training, and image is handled, acquired data can be convenient by automatic vending machine
It is used with automatic vending network server device.
Further, the commodity selling characteristic obtained using merchandise news extracting method of the invention and device,
It is largely operated when on the one hand can be consumed to avoid user, and meets the demand that user voluntarily selects cargo, separately
On the one hand commodity bring businessman loss can be stolen to avoid certain customers again, user's selective purchase operation is few, easy-to-use.
According to the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings, those skilled in the art will be brighter
The above and other objects, advantages and features of the present invention.
Detailed description of the invention
Some specific embodiments of the present invention is described in detail by way of example and not limitation with reference to the accompanying drawings hereinafter.
Identical appended drawing reference denotes same or similar part or part in attached drawing.It should be appreciated by those skilled in the art that these
What attached drawing was not necessarily drawn to scale.In attached drawing:
Fig. 1 is the schematic block diagram of merchandise news extraction element according to an embodiment of the invention;
Fig. 2 is the structural schematic diagram of merchandise news extraction element according to an embodiment of the invention;
Fig. 3 is the schematic diagram of merchandise news extracting method according to an embodiment of the invention;And
Fig. 4 is the schematic architectural diagram of automatically vending system according to an embodiment of the invention.
Specific embodiment
The embodiment of the present invention provides firstly a kind of merchandise news extraction element, automatically extracts automatic vending machine dealer
The merchandise news of product.
Fig. 1 is the schematic block diagram of merchandise news extraction element 100 according to an embodiment of the invention, and Fig. 2 is according to this
The structural schematic diagram of the merchandise news extraction element 100 of invention one embodiment.The merchandise news extraction element 100 is generally
It may include: rotary-tray 140, filming apparatus 110, weighing device 120, controller 130.
Rotary-tray 140, for the commodity 20 of support information to be extracted, rotary-tray 140 can be driven by a motor rotation,
The revolving speed and angle of motor can be controlled.Such as motor can control rotary-tray 140 and gradually revolve according to the angle of setting
Turn, and after the rotation of every step (will also rotate the angle of the setting of commodity 20), control filming apparatus 110 is shot once, thus
To the image at a visual angle of commodity more than 20.
Filming apparatus 110 may include camera, photographic light sources etc., and filming apparatus 110 is arranged above rotary-tray 140
And side, so as to shoot the image at 20 top of commodity or side, in some embodiments, commodity 20 can be with difference
Posture be placed on rotary-tray 140, so as to obtain the image of bottom or other positions.The commodity of the present embodiment are believed
Breath extraction element 100 can arrange multiple filming apparatus 110, and position can be configured as needed.
Weighing device 120 can be set in the lower section of rotary-tray 140, to weigh to above-mentioned commodity 20, obtain quotient
The weight of product 20.
Controller 130 includes memory 131 and processor 132, and wherein memory 131 is stored with computer program, meter
Calculation machine program can be handled above-mentioned image and weight when being executed by processor 132, execute the commodity letter of the present embodiment
Cease extracting method.Controller 130 can be used with certain network capacity and data storage, the industry control of processing capacity
Machine, embedded controller 130 or other electronic controls are to realize.
In addition the merchandise news extraction element 100 of the present embodiment configures input/output unit with being also an option that property, such as
Keyboard, display etc. obtain so that staff importation information, such as price etc. can not be identified from the label of commodity
Information.
Below in conjunction with the merchandise news extracting method of the present embodiment, above-mentioned product information extracting device to be introduced.It should
Merchandise news extracting method can be executed by the controller 130 of the information extracting device of above-described embodiment, and information is cooperated to mention
The various sensing elements (filming apparatus 110, weighing device 120) of device and the motor of rotary-tray 140 etc. are taken to complete its function
Energy.Fig. 3 is the schematic diagram of merchandise news extracting method according to an embodiment of the invention.The merchandise news extracting method is general
May include: to property
Step S302 obtains the weight of the image of the commodity 20 of information to be extracted and the commodity 20 of information to be extracted, obtains
The image taken needs to meet the requirement for the label at least reflecting that the appearance of commodity 20 and commodity 20 are packed, a kind of optional image
Acquisition modes are as follows: control filming apparatus 110 shoots commodity 20;Feature extraction, and root are carried out to the image that shooting obtains
Judge whether the obtained image of shooting meets the image request of commodity identification model according to the characteristics of image of extraction;Reservation meets image
It is required that image, in the case where the obtained image of shooting does not meet the image request of commodity identification model further include: adjustment is clapped
The acquisition parameters for taking the photograph device 110 re-start shooting, and re-start feature extraction.
Above-mentioned image request may include the integrality of image, clarity, if the image etc. comprising desired zone, it can be with
The acquisition parameters for adjusting filming apparatus 110 include: focal length, intensity of light, focus circle, exposure intensity etc..
The process that control filming apparatus 110 shoots commodity 20 may include: the pallet 140 of control carrying commodity 20
It gradually rotates, and after the rotation of every step, control filming apparatus 110 is shot once, to obtain the image at a visual angle of commodity more than 20;
The image of the label of the appearance that can reflect commodity 20 and the packaging of commodity 20 is selected from the image at multiple visual angles.Each angle
The shooting process image request that whether commodity identification model can be met to the obtained image of shooting judge.
Step S304 carries out image recognition to the label packed comprising commodity 20, to identify the label information of commodity 20;
The label of commodity packaging generally describes the information such as manufacturer, batch, specification, brand, the title of commodity 20, identification code, example
Such as the commodity 20 of food Special Category, packaging label also needs to meet certain mandatory standard.By to commodity 20
The label of packaging carries out image recognition, the relevant information of available commodity 20.Manual typing artificial compared to traditional utilization,
Workload is substantially reduced, typing mistake is avoided, improves data input efficiency.
Identify the label information of commodity 20 after can also obtain merchandising 20 needed for information type;According to
Whether the label information that information type judgement identifies is complete, if imperfect, obtains the supplemental information of typing to label information
It is supplemented.Such as the selling price and other information that can not be read from the label that commodity 20 are packed of input commodity 20.
Image is inputted preset commodity identification model as training sample and is trained by step S306, and is training knot
Fruit distribution sale mark;Commodity identification model is trained using the appearance images of commodity 20 as sample, and machine learning is passed through
It obtains.Decision tree, multilayer convolutional neural networks etc. can be used in specific learning algorithm, and finally obtained model can use defeated
The image data entered identifies corresponding commodity 20.Mark of the sale mark as the commodity 20 for being extracted information, utilizes the pin
Selling mark can determine unique corresponding commodity 20.
Step S308 saves label information, 20 weight of commodity and sale mark matching, to obtain going out for commodity 20
Sell characteristic.
After matching preservation obtains the sale characteristic of commodity, these sell characteristics can be by automatic vending machine
300 and automatic vending network server device 200 for being settled accounts to the commodity of sale, such as commodity identification model is for mentioning
Supply automatic vending machine 300 and/or automatic vending network server device 200, for the selling images to automatic vending machine 300 into
Row identification, to determine sold commodity.Commodity weight is for being supplied to automatic vending machine 300 and/or automatic vending network clothes
Business equipment 200, it is true by the weight of reduction for being verified to the commodity sold determined using commodity identification model
Whether the commodity for determining the reduction that image recognition goes out are correct, and can verify whether consumer has damaged commodity.Label information (including
The information of supplement) after be supplied to automatic vending machine 300 and/or automatic vending network server device 200, it can be to being sold
Commodity carry out settlement of transactions, form transaction record.
The merchandise news extracting method of the present embodiment learns commodity image using machine learning algorithm, trains
Commodity identification model, by identifying that the label image of commodity packaging obtains the label information of commodity, by weighing to obtain to commodity
The weight of commodity.Utilize the sale characteristic that label information, commodity weight and sale mark matching are saved to obtained commodity
According to, can for automatic vending machine 300 in vending articles by the sale image of shooting, determine the commodity that user takes out, utilize
Commodity weight is verified to article is picked and placed, and carries out Automatic-settlement accordingly, and whole process carries out automatically, high-efficient, can be with
Merchandise sales information needed for being quickly obtained automatic vending machine 300.
The present embodiment additionally provides a kind of automatically vending system, and Fig. 4 is automatic vending according to an embodiment of the invention
The schematic architectural diagram of system 10, the automatic harvesting system 10 generally may include: above-mentioned merchandise news extraction element
100, automatic vending machine 300, network server device 200, wherein above-mentioned merchandise news extraction element 100 is for providing merchandise sales
Information, automatic vending machine 300 determine the commodity sold, net using the sale characteristic that merchandise news extraction element 100 obtains
Network service equipment 200 is used to carry out data management to merchandise sales information and automatic vending machine 300, and completes the commodity sold
It is settled accounts, data management, the functions such as information processing.User can complete identification authentication and settlement operations by terminal 400.
Below in conjunction with automatically vending system 10 progress merchandise sales an example to merchandise news in above-described embodiment
The effect of merchandise sales information obtained in extraction element 100 is introduced.In the commodity or commodity for needing restocking New raxa
Packaging or specification when changing, new product can be placed on the rotary-tray 140 of merchandise news extraction element 100.
Merchandise news extraction element 100 learns product image, and obtains relevant weight information, label information, supplemental information,
It obtains selling required sale characteristic.
Network server device 200 can the sale characteristic to the commodity in all sales be managed.Network service
Equipment 200 can be realized by network equipments such as server, cloud device, fictitious host computers, due to the specific framework of this kind of equipment
Belong to known by those of ordinary skill in the art, this will not be repeated here, the network clothes in the automatically vending system 10 of the present embodiment
Business equipment 200 is for realizing data management, Automatic-settlement function using above-mentioned framework.
The automatic vending machine 300 includes cabinet, door body.Box house can define one or more chambers, in chamber
At least one area Shou Wu can be set, for placing commodity to be vended.The construction in the area Shou Wu can according to need setting.?
In some alternative embodiments, cabinet can be set to the heat insulating box for being similar to refrigerator or incubator, thus realize heat preservation or
Different storing temperature can be set in cold storage function, such as different chamber, thus meet the needs of customer consumption difference commodity,
In this kind of embodiment, temperature control system, refrigeration system, heating system can also be arranged in automatic vending machine 300, to realize to selling object
The temperature in area controls.
The area Shou Wu can be separated to form by board putting things or shelf, and one or more can be set on every layer of board putting things and shelf
A weighing device weighs the weight of the commodity in the area Shou Wu.
Also filming apparatus can be set in cabinet, the position of the filming apparatus can be according to the area Shou Wu and the position of door body
It sets and is configured.Such as filming apparatus can be used for being arranged inside object area on sale, so as to for shooting commodity in the area Shou Wu
Modified-image.In addition filming apparatus also can be set at the position of door body, and the motion images of commodity are picked and placed with user.
The image of above-mentioned filming apparatus shooting can use the commodity identification mould that the training of merchandise news extraction element 100 obtains
Type determines the type and quantity for the commodity that time user picks and places.
Door body is set to cabinet front, for closing the area Shou Wu.Cabinet front can be arranged by hinged mode in door body
Side, by being pivotably opened and closed the area Shou Wu, the quantity Matching for the chamber that the quantity of door body can be limited with box house,
To be separately turned on each chamber.
Door body can be correspondingly arranged on automatic door lock and door body detection device, and wherein automatic door lock is used to refer in reception unlock
It is unlocked after order, so that door body can be opened, various mechanical locks, electromagnetic lock etc. can be used.Door body detection device is used for
Contact switch realization, the door body as known by those skilled in the art generally can be used in the opening and closing event for detecting door body respectively
Detection device can also detect the opening and closing movement of door body with means such as hall device, magnetosensitive devices.Door body detection device can be set
It is set to multiple, setting is matched with door body.
User can pass through the two dimensional code of scanning automatic vending machine 300 when needing to buy the commodity of automatic vending machine 300
Or other modes obtain the mark of automatic vending machine 300, and combine identification information (such as the account of user, password, the people of user
Body biological information) unlocking request is sent to network server device 200 together.
Network server device 200 authenticates user according to unlocking request, generates unlock instruction and to automatic vending machine
300 provide, and automatic vending machine 300 drives door lock to unlock according to unlock instruction, hence into automatic vending process so that user opens
The area Men Congshouwu picks and places commodity.
After determining that door body is opened, the filming apparatus on automatic vending machine 300 starts, and shoots the change of commodity in the area Shou Wu
Change image and/or user picks and places the motion images of commodity, and the image obtained according to shooting is sold using the determination of commodity identification model
Commodity change conditions in object area.
After door body is closed, the commodity weight in the area Shou Wu operated is detected, before and after obtaining loading or unloading operation
Weight change amount, and judge whether weight change amount is corresponding with the commodity variation determined according to image.Such as it can be according to figure
As the value volume and range of product for the variation commodity determined is according to the determining weight for changing commodity of commodity data model;Judge weight change
It can be to judge to detect whether obtained weight change amount pass through that whether amount changes corresponding with the commodity determined according to image
In the setting error for the weight that model is determined, namely before and after switch gate storing part weight reduction determined with image recognition
The weight of commodity taken out of user be consistent;If then determining whether weight change amount is corresponding with commodity variation.
The process compared by above-mentioned weight and image recognition, on the one hand can verify the result of image recognition,
On the other hand it is also possible to prevent user and opens commodity packaging, after taking out commodity, reduction commodity packaging appearance puts back to automatic vending machine
In 300, the behavior of commodity is stolen.Such as user puts back to beverage bottle certainly after taking out the beverage in automatic vending machine 300 and drinking
In dynamic vending machine 300, the behavior for forging non-consumer lines can be identified by weight change amount.
If weight change amount is corresponding with the commodity variation determined according to image, is changed according to commodity and generate consumption clearing letter
Breath, is settled accounts with reporting to network server device 200 for it.
If above-mentioned can be with defeated in the case where weight change amount is not correspond to the commodity variation determined according to image
Abnormal prompt is settled accounts out, and user is waited to carry out confirmation operation;If settling accounts abnormal prompt in setting time (such as in 1 minute)
Not by user response, then clearing exception information is reported to network server device 200, so that network server device 200 is according to user
The processing protocol agreed in advance is settled accounts.
Such as in the case where determining reduced weight and not meeting the commodity of user's taking-up, automatic vending machine can be passed through
300 screen shows the goods for consumption inventory of user, confirms for user.Or it is shown and is consumed to user by user terminal 400
List of articles confirms for user.When there is such case, the image that user picks and places article can be saved, and be looked into as subsequent
The evidence of card.
If user carries out responding above-mentioned clearing abnormal prompt, network server device 200 is agreed in advance according to user
Processing protocol settled accounts.Such as according to a kind of more progress amount of money meters of spending amount in image recognition or weighing results
It calculates, in addition it can be recorded in customer consumption credit.
When output clearing abnormal prompt, it can both show that the value volume and range of product for the variation commodity determined and detection obtained
The weight change amount, with to user report settle accounts abnormal cause, can also further show interface on solution party be provided
Case, such as offer customer service contact method, offer require to reanalyse image, weigh again, require unlock enabling extract again
Option of operation such as product etc..
It can be seen that commodity pin obtained in the merchandise news extraction element 100 using the present embodiment by above-mentioned introduction
Information is sold, information management and the Automatic-settlement of automatic vending network can be completed, greatly improve the efficiency while avoiding user
It is largely operated when being consumed, and meets the demand that user voluntarily selects cargo.To be automatic vending machine 300
Large-scale promotion provides technical support.
So far, although those skilled in the art will appreciate that present invention has been shown and described in detail herein multiple shows
Example property embodiment still without departing from the spirit and scope of the present invention, still can according to the present disclosure directly
Determine or deduce out many other variations or modifications consistent with the principles of the invention.Therefore, the scope of the present invention is understood that and recognizes
It is set to and covers all such other variations or modifications.
Claims (10)
1. a kind of merchandise news extracting method, comprising:
The image of the commodity of information to be extracted is obtained, described image at least reflects the appearance of the commodity and the mark of commodity packaging
Label;
Image recognition is carried out to the label comprising the commodity packaging, to identify the label information of the commodity;
Obtain the weight of the commodity of the information to be extracted;
Described image is inputted preset commodity identification model as training sample to be trained, and is training result distribution sale
Mark;
The label information, the commodity weight and the sale are identified matching to save, to obtain going out for the commodity
Sell characteristic.
2. according to the method described in claim 1, wherein, the step of obtaining the image of the commodity of the information to be extracted, includes:
Control filming apparatus shoots the commodity;
Feature extraction is carried out to the image that shooting obtains, and whether is accorded with according to the image that the characteristics of image of extraction judges that shooting obtains
Close the image request of the commodity identification model;
Retain the image for meeting described image requirement.
3. according to the method described in claim 2, wherein, not meeting the figure of the commodity identification model in the image that shooting obtains
In the case where requiring further include:
The acquisition parameters for adjusting the filming apparatus re-start shooting, and re-start feature extraction.
4. according to the method described in claim 2, wherein, the process that control filming apparatus shoots the commodity includes:
The pallet that control carries the commodity gradually rotates, and after the rotation of every step, controls the filming apparatus shooting once, from
And obtain the image at the multiple visual angles of the commodity;
The image of the label of the appearance and commodity packaging that can reflect the commodity is selected from the image at the multiple visual angle.
5. according to the method described in claim 1, wherein, also being wrapped after the step of identifying the label information of the commodity
It includes:
Obtain information type needed for selling the commodity;
Whether the label information identified according to information type judgement is complete, if imperfect, obtains the supplemental information of typing
To supplement the label information.
6. the method according to any one of claims 1 to 5, wherein the step of saving the sale characteristic it
Afterwards further include:
The commodity identification model is provided to automatic vending machine and/or automatic vending network server device, for described in its utilization
Commodity identification model identifies the selling images of the automatic vending machine, to determine sold commodity.
7. according to the method described in claim 6, wherein, after the step of saving the sale characteristic further include:
The commodity weight is provided to automatic vending machine and/or automatic vending network server device, to utilize the commodity weight
The commodity sold determined using the commodity identification model are verified.
8. according to the method described in claim 6, wherein, after the step of saving the sale characteristic further include:
The label information is provided to automatic vending machine and/or automatic vending network server device, to utilize the label information
Settlement of transactions is carried out to the commodity sold.
9. a kind of merchandise news extraction element, comprising:
Rotary-tray, the commodity for support information to be extracted;
At least one filming apparatus is set to the periphery of the rotary-tray, for shooting to the commodity;
Weighing device is set to the lower section of rotary-tray, to weigh to the commodity;And
Controller, the controller includes memory and processor, wherein the memory is stored with computer program, it is described
Computer program is able to carry out merchandise news extraction side described in any item of the claim 1 to 8 when being executed by the processor
Method.
10. a kind of automatically vending system, comprising:
Merchandise news extraction element according to claim 9;
Automatic vending machine is configured with the sale characteristic that the merchandise news extraction element obtains and determines the quotient sold
Product;And
Network server device distinguishes data connection with the merchandise news extraction element and automatic vending machine, for sale
Commodity settled accounts.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810150043.6A CN110164033A (en) | 2018-02-13 | 2018-02-13 | Merchandise news extracting method, merchandise news extraction element and automatically vending system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810150043.6A CN110164033A (en) | 2018-02-13 | 2018-02-13 | Merchandise news extracting method, merchandise news extraction element and automatically vending system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110164033A true CN110164033A (en) | 2019-08-23 |
Family
ID=67635313
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810150043.6A Pending CN110164033A (en) | 2018-02-13 | 2018-02-13 | Merchandise news extracting method, merchandise news extraction element and automatically vending system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110164033A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110751780A (en) * | 2019-12-10 | 2020-02-04 | 重庆康伲乐医药科技有限公司 | Goods detection system based on RFID and detection method thereof |
CN110827487A (en) * | 2019-11-04 | 2020-02-21 | 北京三快在线科技有限公司 | Article image data acquisition method and device, storage medium and electronic equipment |
CN110987140A (en) * | 2019-11-28 | 2020-04-10 | 浙江由由科技有限公司 | Commodity weighing auxiliary equipment and weighing equipment |
CN112233344A (en) * | 2020-09-10 | 2021-01-15 | 苏宁云计算有限公司 | Commodity information acquisition device |
CN112419015A (en) * | 2021-01-25 | 2021-02-26 | 北京每日优鲜电子商务有限公司 | Article information pushing method and device, electronic equipment and computer readable medium |
CN113178035A (en) * | 2021-04-19 | 2021-07-27 | 上海云央网络科技有限公司 | Method for AI to identify shipment information in automatic vending device and application |
CN113450505A (en) * | 2020-10-22 | 2021-09-28 | 齐喝彩(上海)管理咨询有限公司 | Temperature control method and device of self-service vending machine |
CN115100780A (en) * | 2022-05-30 | 2022-09-23 | 河南云迹智能技术有限公司 | Intelligent container capable of automatically replenishing goods |
CN117115571A (en) * | 2023-10-25 | 2023-11-24 | 成都阿加犀智能科技有限公司 | Fine-grained intelligent commodity identification method, device, equipment and medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN201243844Y (en) * | 2008-09-04 | 2009-05-27 | 郑锐 | Rotating show table with regulated speed |
CN104424578A (en) * | 2013-09-10 | 2015-03-18 | 上海民通电子商务有限公司 | Device for demonstrating merchandises in electronic commerce |
CN105989512A (en) * | 2015-02-10 | 2016-10-05 | 黄金富知识产权咨询(深圳)有限公司 | Commodity sale system and corresponding method |
CN106529427A (en) * | 2016-10-24 | 2017-03-22 | 合肥华凌股份有限公司 | Method and system of identifying target object in image, and method and system of identifying food in refrigerator |
CN107167221A (en) * | 2017-05-12 | 2017-09-15 | 智锐达仪器科技南通有限公司 | The agricultural product source tracing method and intelligent electronic-scale of a kind of utilization intelligent electronic-scale |
CN107292248A (en) * | 2017-06-05 | 2017-10-24 | 广州诚予国际市场信息研究有限公司 | A kind of merchandise control method and system based on image recognition technology |
CN107679850A (en) * | 2017-09-15 | 2018-02-09 | 苏衍杰 | A kind of commodity settlement method, apparatus and system |
-
2018
- 2018-02-13 CN CN201810150043.6A patent/CN110164033A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN201243844Y (en) * | 2008-09-04 | 2009-05-27 | 郑锐 | Rotating show table with regulated speed |
CN104424578A (en) * | 2013-09-10 | 2015-03-18 | 上海民通电子商务有限公司 | Device for demonstrating merchandises in electronic commerce |
CN105989512A (en) * | 2015-02-10 | 2016-10-05 | 黄金富知识产权咨询(深圳)有限公司 | Commodity sale system and corresponding method |
CN106529427A (en) * | 2016-10-24 | 2017-03-22 | 合肥华凌股份有限公司 | Method and system of identifying target object in image, and method and system of identifying food in refrigerator |
CN107167221A (en) * | 2017-05-12 | 2017-09-15 | 智锐达仪器科技南通有限公司 | The agricultural product source tracing method and intelligent electronic-scale of a kind of utilization intelligent electronic-scale |
CN107292248A (en) * | 2017-06-05 | 2017-10-24 | 广州诚予国际市场信息研究有限公司 | A kind of merchandise control method and system based on image recognition technology |
CN107679850A (en) * | 2017-09-15 | 2018-02-09 | 苏衍杰 | A kind of commodity settlement method, apparatus and system |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110827487A (en) * | 2019-11-04 | 2020-02-21 | 北京三快在线科技有限公司 | Article image data acquisition method and device, storage medium and electronic equipment |
CN110987140A (en) * | 2019-11-28 | 2020-04-10 | 浙江由由科技有限公司 | Commodity weighing auxiliary equipment and weighing equipment |
CN110751780A (en) * | 2019-12-10 | 2020-02-04 | 重庆康伲乐医药科技有限公司 | Goods detection system based on RFID and detection method thereof |
CN110751780B (en) * | 2019-12-10 | 2023-05-16 | 重庆康伲乐医药科技有限公司 | Goods detection system based on RFID and detection method thereof |
CN112233344A (en) * | 2020-09-10 | 2021-01-15 | 苏宁云计算有限公司 | Commodity information acquisition device |
CN113450505A (en) * | 2020-10-22 | 2021-09-28 | 齐喝彩(上海)管理咨询有限公司 | Temperature control method and device of self-service vending machine |
CN113450505B (en) * | 2020-10-22 | 2023-08-11 | 齐喝彩(上海)管理咨询有限公司 | Temperature control method and device of self-service vending machine |
CN112419015A (en) * | 2021-01-25 | 2021-02-26 | 北京每日优鲜电子商务有限公司 | Article information pushing method and device, electronic equipment and computer readable medium |
CN113178035A (en) * | 2021-04-19 | 2021-07-27 | 上海云央网络科技有限公司 | Method for AI to identify shipment information in automatic vending device and application |
CN115100780A (en) * | 2022-05-30 | 2022-09-23 | 河南云迹智能技术有限公司 | Intelligent container capable of automatically replenishing goods |
CN115100780B (en) * | 2022-05-30 | 2023-12-05 | 河南云迹智能技术有限公司 | Automatic intelligent packing cupboard of moisturizing |
CN117115571A (en) * | 2023-10-25 | 2023-11-24 | 成都阿加犀智能科技有限公司 | Fine-grained intelligent commodity identification method, device, equipment and medium |
CN117115571B (en) * | 2023-10-25 | 2024-01-26 | 成都阿加犀智能科技有限公司 | Fine-grained intelligent commodity identification method, device, equipment and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110164033A (en) | Merchandise news extracting method, merchandise news extraction element and automatically vending system | |
CN206179117U (en) | Intelligence display system | |
CN107274555B (en) | Automatic vending method, apparatus and system | |
TWI732146B (en) | Intelligent vending equipment and its control method | |
CN105324714B (en) | Computer control, unattended automatic checkout retail shop | |
WO2019184646A1 (en) | Method and device for identifying merchandise, merchandise container | |
US20220114854A1 (en) | Systems and methods for vending and/or purchasing mobile phones and other electronic devices | |
US20190172041A1 (en) | Methods and apparatus for providing an easy return, exchange and purchasing experience in a retail store environment | |
CN107730168A (en) | A kind of automatic vending machine automated stock control system and method based on image recognition | |
CN106710071A (en) | Vending system with function of radio-frequency chip recognition | |
CN107833368A (en) | Automatic selling cabinet and its application method | |
CN108198054A (en) | A kind of unmanned counter that commodity is supported to subscribe and marketing system | |
CN110164032A (en) | Automatic vending machine and its data processing method and automatically vending system | |
CN108399681A (en) | A kind of intelligence vending machine | |
AU2017378617A1 (en) | An automatic in-store registration system | |
CN110164036A (en) | Automatic vending machine and its control method | |
CN204791269U (en) | Automatic sales counter | |
CN209168266U (en) | A kind of self-service selling cabinet | |
CN110197240A (en) | The control method of unmanned Weighing type sales counter based on image recognition | |
CN110164029A (en) | Automatic vending machine and its control method | |
CN208027468U (en) | Self-service equipment and self-service system | |
WO2019165895A1 (en) | Automatic vending method and system, and vending device and vending machine | |
CN110298965A (en) | A kind of intelligent selling cabinet system and method based on dynamic computer check visual identity | |
CN107657728A (en) | Self-help intelligent vending machine, selling system and its vending method | |
CN109190706A (en) | Self-service method, apparatus and system |
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 |
Application publication date: 20190823 |
|
RJ01 | Rejection of invention patent application after publication |