CN207624043U - Commodity clearance equipment and purchase system - Google Patents
Commodity clearance equipment and purchase system Download PDFInfo
- Publication number
- CN207624043U CN207624043U CN201721885962.5U CN201721885962U CN207624043U CN 207624043 U CN207624043 U CN 207624043U CN 201721885962 U CN201721885962 U CN 201721885962U CN 207624043 U CN207624043 U CN 207624043U
- Authority
- CN
- China
- Prior art keywords
- commodity
- clearance
- tested
- product features
- image
- 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.)
- Active
Links
Landscapes
- Image Analysis (AREA)
Abstract
A kind of commodity clearance equipment and purchase system.The commodity clearance equipment includes video generation device, product features extraction element, commodity confirmation device and veritifies device.The purchase system includes the commodity clearance equipment, server and electronic fare payment system, and the commodity clearance equipment receives commodity transaction inventory by electronic fare payment system and/or server.By the commodity clearance equipment and purchase system, consumer can be made quickly to be veritified in autonomous after closing, can further greatly reduce the time that consumer waits in line.Further, it is possible to reduce input of the businessman to human cost, the transaction cost of businessman is thus reduced.
Description
Technical field
The utility model is related to the commodity clearance equipments in a kind of purchase system and a kind of purchase system.
Background technology
Currently, in the retail business such as department store, each commodity are identified with codes such as bar code, Quick Response Codes, using such as
The scanning systems such as POS machine are settled accounts, and consumer must it is expected that the commodity of purchase are presented to cashier one by one in cashier,
Entire service scan piece by piece by cashier, accumulative total price, consumer then could by cash or swipe the card etc. modes into
Row clearing, carry out commodity clearance.However during above-mentioned commodity clearance, consumer needs the knot that takes an undesirably long time
Account.
In particular, when customer count is more, workload is huge in such a way that cashier manually verifies commodity one by one,
And it will inevitably generate that a large amount of consumers are lined up crowding phenomenons or even thus individual consumer abandons this shopping.
In recent years, with the development of internet, the relevant technologies propose automatic/autonomous reckoning system, relative to tradition
Checkout efficiency have certain raising.However in previous automatic reckoning system, people is still needed to for check-out commodity
Work is verified, and the demand that people quickly check out is cannot be satisfied.
Therefore, how to promote efficiency of account settling, reduce transaction cost as this field urgent problem to be solved.
Invention content
In view of this, a kind of commodity clearance equipment of the utility model offer and purchase system, to solve prior art presence
The above problem.
Specifically, the utility model provides a kind of commodity clearance equipment, which is characterized in that including:
Video generation device, for generating the image for including commodity to be tested;
Product features extraction element, for extracting one or more commodity from the image that described image generating means generate
Feature;
Commodity confirm device, product features to be tested for extracting the product features extraction element with prestore it is each right
It is compared according to the product features of commodity, confirms the type and/or quantity of commodity to be tested;And
Veritify device, type of merchandize and/or quantity for veritifying the commodity to be tested that the commodity determining device is confirmed
With in commodity transaction inventory type of merchandize and/or quantity whether be consistent, and export veritification result.
Preferably, type of merchandize and/or quantity to be tested in commodity transaction inventory type of merchandize and/or quantity be consistent and be
Refer to the two matching degree and reaches preset matching degree threshold value or more.For example, 70% or more;75% or more;80% or more;
85% or more;90% or more;95 or more %;98 or more %;99% or more.
Preferably, the preset matching degree threshold value includes but not limited to:Type of merchandize matching degree threshold value, commodity number
Flux matched degree threshold value and/or goods amount matching degree threshold value etc..
Preferably, in above-mentioned commodity clearance equipment, the commodity confirm that device includes computing unit and matching unit, described
Computing unit is used for the product features to be tested that the product features extraction element extracts and each commodity for compareing commodity to prestore
Feature is compared, and calculates similarity;The matching unit is used to judge that the calculated similarity of the computing unit institute to be
It is no be more than defined similarity threshold, when calculated similarity be more than defined similarity threshold when, confirm commodity to be tested with
Control commodity are consistent.
Preferably, the commodity confirm that device includes deep learning unit and commodity confirmation unit, the deep learning list
Member carries out model training to the product features of each control commodity to prestore, and generates the priori frame of characterization product features;It is described true
Recognize the kind that unit confirms commodity to be tested using the product features to be tested and the priori frame of product features extraction element extraction
Class and/or quantity.
Preferably, the commodity clearance equipment further includes commodity image region recognition device, for generating described image
Image segmentation acquired in device is multiple sub-image areas, and subgraph feature, base are extracted respectively for each sub-image area
In each sub-image area of subgraph feature recognition extracted whether be commodity image region.Wherein, the product features carry
The commodity image region that device is identified based on the commodity image region recognition device is taken to extract product features.
Preferably, the commodity clearance equipment further includes clearance unit, is quotient to be tested in the veritification result for veritifying device
Kind class and/or quantity in commodity transaction inventory type of merchandize and/or in the case that quantity is consistent, the clearance unit is defeated
Go out to allow the instruction of commodity clearance.
Preferably, in the veritification result for veritifying device is type of merchandize to be tested and/or quantity and commodity transaction inventory
In the case that type of merchandize and/or quantity are not consistent, the clearance unit output abnormality prompt.
Preferably, in the case where calculated similarity is less than defined similarity threshold, the matching unit prompt
Detect background image and/or commodity edge image.
Preferably, the similarity threshold is 70-100%, such as 90%, 80% or 70%.
Preferably, the product features of the control commodity include multiple position appearance picture features of the control commodity.
Preferably, described that the indicating of commodity clearance, the abnormal prompt is allowed to include that information of voice prompt, picture cues are believed
It is any one or more in breath, luminous prompting information, text prompt message.
Preferably, the commodity clearance equipment further includes storage device, the storage device in advance by each control commodity and
Verification corresponding with each control commodity is stored in association with feature.
The utility model also provides a kind of purchase system comprising server and electronic fare payment system, which is characterized in that its
Further include above-mentioned commodity clearance equipment, the commodity clearance equipment receives the quotient by electronic fare payment system and/or server
Product schedule of dealing.
The communication connection type of above-mentioned each component is not limited, can need to select wired company according to specific communication
Connect, be wirelessly connected in one kind or combinations thereof.
Advantageous effect
By commodity clearance equipment described in the utility model and purchase system, consumer can be made quick in autonomous after closing
It is veritified, can further greatly reduce the time that consumer waits in line.Further, it is possible to reduce businessman to human cost
Input, thus reduces the transaction cost of businessman.
Description of the drawings
Fig. 1 is an exemplary structural schematic diagram of the utility model commodity clearance equipment.
Specific implementation mode
The utility model is illustrated below by way of specific implementation mode, but the utility model is not limited to
These embodiments.
As shown in Figure 1, commodity clearance equipment described in the utility model may include:
For obtaining the video generation device 1 for carrying check-out commodity image to be tested, described image generating means 1 include
CPU, ROM, RAM, one or more photographic devices (or scanner) and communication interface, the photographic device can be by ccd images
The components such as sensor or cmos sensor are constituted.When check-out commodity to be tested are placed into operation console by user, the camera shooting
Device under the control of cpu, such as with the frame rate of 30fps etc. shoots image, generates image data, and camera unit is by generation
Image data is saved in RAM;
Acquired one or more is included the image of commodity by product features extraction element 2, described image generating means 1
It is sent to product features extraction element 2.The image that the product features extraction element 2 is obtained from described image generating means 1
The one or more product features of middle extraction.For example, the bumps of the profile of the extraction of product features extraction element 2 commodity, surface
The state of one or more of situation, color etc. is as product features.
Commodity confirm that device 3, the commodity confirm that device 3 includes computing unit 4 and matching unit 5, the computing unit 4
By the product features to be tested that the product features extraction element 2 is extracted and each quotient for compareing commodity for being pre-stored in storage device 8
Product feature is compared, and calculates their similarity.
Such as calculate separately the product features extracted and the histogram that compares commodity, then calculate returning for two histograms
One changes related coefficient (Pasteur's distance, histogram intersection distance) etc..Preferably, can also feature based point it is similar to carry out image
Degree calculates or calculates similarity using CNN disaggregated models scheduling algorithm.
The utility model is not restricted the calculation of similarity.
Preferably, the product features of the control commodity include multiple position external appearance characteristics of the control commodity.
Preferably, the computing unit 4 is realized by microprocessors such as DSP.
The matching unit 5 receives 4 calculated similarity of the computing unit, judges that the computing unit 4 is calculated
Whether the similarity gone out is more than defined similarity threshold, the case where calculated similarity is more than defined similarity threshold
Under, confirm that commodity to be tested are consistent with commodity are compareed.
Conversely, in the case that calculated similarity is less than defined similarity threshold, the prompt inspection of the matching unit 5
The image measured is commodity edge image or background image.In the case, commodity clearance equipment continues next product features
Judge.
Preferably, the matching unit 5 is realized by programmable arrays such as the microprocessors such as DSP or FPGA.
Preferably, the commodity confirm that device 3 includes deep learning unit and commodity confirmation unit, the deep learning unit
Model training is carried out for the product features to each control commodity to prestore, and generates the priori frame of characterization product features;One
In embodiment, the machine training pattern of the more hidden layers of deep learning cell formation, the depth of model structure includes 5 layers, 6 layers, even
The hidden node of 10 multilayers.Deep learning unit generates priori frame based on full convolutional neural networks.
The confirmation unit is used for the product features to be tested using product features extraction element extraction and the priori
Frame confirms the type and/or quantity of commodity to be tested.
Preferably, the deep learning unit is realized by multilayer perceptron or autocoder, the confirmation unit
For example, by using all kinds of microprocessors such as CPU, DSP.
It veritifies device 6 and preferably includes CPU, ROM, RAM and communication interface.Confirm device institute really for veritifying the commodity
The type of merchandize and/or quantity to be tested recognized in commodity transaction inventory type of merchandize and/or quantity whether be consistent, and export core
Test result.
Preferably, the device 6 of veritifying veritifies the type of merchandize and/or quantity and quotient that the commodity confirm that device is confirmed
Whether type of merchandize and/or quantity in product schedule of dealing are consistent.Wherein, type of merchandize and/or quantity to be tested and commodity transaction
It refers to that the two matching degree reaches preset matching degree threshold value or more that type of merchandize and/or quantity in inventory, which are consistent,.For example,
70% or more;75% or more;80% or more;85% or more;90% or more;95 or more %;98 or more %;99% or more.
Preferably, the preset matching degree threshold value includes type of merchandize matching degree threshold value, commodity amount matching degree
Threshold value and/or goods amount matching degree threshold value.
Preferably, the commodity clearance equipment further includes clearance unit, if the veritification device 6 veritifies result phase
Symbol, the clearance unit output allow the instruction of commodity clearance.
Preferably, if the veritification result for veritifying device 6 is not consistent, the clearance unit output abnormality prompt.
Preferably, described that the indicating of commodity clearance, the abnormal prompt is allowed to include that information of voice prompt, picture cues are believed
It is any one or more in breath, luminous prompting information, text prompt message.
Preferably, the similarity threshold is set as 70-100%, such as 90%, 80% or 70%.
Preferably, the veritification device 6 exists the inventory to be measured of veritification, and/or commodity transaction inventory real-time display
On display unit 7.
Preferably, it is shown on the display unit 7 when the veritification device 6 will veritify fructufy.
Preferably, the commodity clearance equipment further includes storage device 8, and the storage device 8 is in advance by each control commodity
Information and product features corresponding with each control commodity store in association.
In a variation example, the commodity clearance equipment of the utility model further includes:
Commodity image region recognition device 9, for being multiple sons by the image segmentation acquired in described image generating means
Image-region extracts subgraph feature respectively for each sub-image area, each based on the subgraph feature recognition extracted
Whether sub-image area is commodity image region;
For example, commodity image region recognition device 9 using pattern match technology etc. identify each sub-image area whether be
A part for commodity or commodity.Specifically, commodity image region recognition device 9 utilizes the image zooming-out wheel of binary image data
Profile utilizes edge detection algorithm or difference algorithm detection commodity image region.
Wherein, the commodity figure that the product features extraction element 2 is identified based on the commodity image region recognition device 9
Product features are extracted as region.
In addition, the utility model additionally provides a kind of purchase system, including server and electronic fare payment system, further include
Above-mentioned commodity clearance equipment, user generate commodity transaction inventory by electronic fare payment system, and by the commodity transaction list synchronization
It is sent to server, the commodity clearance equipment receives the commodity transaction inventory by electronic fare payment system and/or server.
As more specific embodiment, the electronic fare payment system is the smart mobile phone equipped with shopping APP, the intelligence
Mobile phone has camera.
In a specific purchase system embodiment, customer logs in APP using phone number.Pass through cell-phone camera in APP
Commodity are put into shopping cart by head items scanning bar code after identifying commodity.Click shopping vehicle confirms commodity to be purchased after scanning through commodity
Inventory.The selection means of payment (such as wechat, Alipay) is paid after clicking payment.Payment is completed to generate commodity transaction inventory,
The commodity transaction inventory is sent to server simultaneously.
Designated lane is paid in APP, commodity clearance equipment generates the check-out commodity image to be tested of user, will be based on to be tested
The product features that commodity image extracts are compared with the product features of the control commodity to prestore, and calculate phase between the two
Confirm the type and/or number of commodity to be tested in the case where calculated similarity is more than defined similarity threshold like degree
Amount, however veritify confirmed type of merchandize and/or quantity in commodity transaction inventory commodity and/or quantity whether be consistent.
For example, whether the type of commodity, quantity and/or transaction amount reach preset matching degree threshold value or more.
If veritifying result to be consistent, customer speeds passage through customs.
In another embodiment, it is not inconsistent if veritifying result, commodity clearance equipment carries out abnormal prompt.User is optional
It selects and supplement payment is carried out to non-payment for merchandise by smart mobile phone, after completing supplement payment, clearance equipment enters through state.
Claims (13)
1. a kind of commodity clearance equipment, which is characterized in that including:
Video generation device, for generating the image for including commodity to be tested;
Product features extraction element, it is special for extracting one or more commodity from the image that described image generating means generate
Sign;
Commodity confirm device, and the product features to be tested for extracting the product features extraction element respectively compare quotient with what is prestored
The product features of product are compared, and confirm the type and/or quantity of commodity to be tested;And
Device is veritified, for veritifying the type of merchandize and/or quantity to be tested and commodity transaction that the commodity confirm that device is confirmed
Whether type of merchandize and/or quantity in inventory are consistent, and export veritification result.
2. commodity clearance equipment according to claim 1, which is characterized in that
Type of merchandize and/or quantity to be tested in commodity transaction inventory type of merchandize and/or quantity to be consistent both refer to matching
Degree reaches preset matching degree threshold value or more.
3. commodity clearance equipment according to claim 2, which is characterized in that
The preset matching degree threshold value includes type of merchandize matching degree threshold value, commodity amount matching degree threshold value and/or quotient
Product amount of money matching degree threshold value.
4. commodity clearance equipment described in any one of claim 1 to 3, which is characterized in that the commodity confirm device
Including computing unit and matching unit, the commodity to be tested that the computing unit is used to extract the product features extraction element are special
It levies and is compared with the product features of each control commodity to prestore, and calculate similarity;The matching unit is described for judging
Whether the calculated similarity of computing unit institute is more than defined similarity threshold, when calculated similarity is more than defined phase
When like degree threshold value, confirm that commodity to be tested are consistent with commodity are compareed.
5. commodity clearance equipment described in any one of claim 1 to 3, which is characterized in that the commodity confirm device
Including deep learning unit and commodity confirmation unit, the deep learning unit is used for special to the commodity of each control commodity to prestore
Sign carries out model training, and generates the priori frame of characterization product features;The confirmation unit using the product features for being carried
Product features to be tested that device extracts and the priori frame are taken to confirm the type and/or quantity of commodity to be tested.
6. commodity clearance equipment described in any one of claim 1 to 3, which is characterized in that further include:
Commodity image region recognition device, for being multiple sub-image regions by the image segmentation acquired in described image generating means
Subgraph feature is extracted in domain respectively for each sub-image area, based on each sub-image regions of subgraph feature recognition extracted
Whether domain is commodity image region;
Wherein, the commodity image region that the product features extraction element is identified based on the commodity image region recognition device
To extract product features.
7. commodity clearance equipment described in any one of claim 1 to 3, which is characterized in that further include clearance unit,
The veritification result for veritifying device is type of merchandize to be tested and/or quantity and the type of merchandize and/or quantity in commodity transaction inventory
In the case of being consistent, the clearance unit output allows the instruction of commodity clearance;It is quotient to be tested in the veritification result for veritifying device
Kind class and/or quantity in commodity transaction inventory type of merchandize and/or in the case that quantity is not consistent, the clearance unit
Output abnormality prompts.
8. commodity clearance equipment according to claim 4, which is characterized in that be less than defined phase in calculated similarity
In the case of like degree threshold value, the matching unit prompt detects background image and/or commodity edge image.
9. commodity clearance equipment according to claim 4, which is characterized in that the similarity threshold is set as 70%-
100%.
10. commodity clearance equipment described in any one of claim 1 to 3, which is characterized in that the quotient of the control commodity
Product feature includes multiple position external appearance characteristics of the control commodity.
11. commodity clearance equipment according to claim 7, which is characterized in that it is described allow commodity clearance instruction and/or
The abnormal prompt includes any one in information of voice prompt, picture cues information, luminous prompting information, text prompt message
Item is multinomial.
12. commodity clearance equipment described in any one of claim 1 to 3, which is characterized in that further include storage device,
The storage device in advance stores each control merchandise news and product features corresponding with each control commodity in association.
13. a kind of purchase system comprising server and electronic fare payment system, which is characterized in that further include claim 1~12
Any one of them commodity clearance equipment, the commodity clearance equipment pass through described in electronic fare payment system and/or server reception
Commodity transaction inventory.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201721885962.5U CN207624043U (en) | 2017-12-28 | 2017-12-28 | Commodity clearance equipment and purchase system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201721885962.5U CN207624043U (en) | 2017-12-28 | 2017-12-28 | Commodity clearance equipment and purchase system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN207624043U true CN207624043U (en) | 2018-07-17 |
Family
ID=62830493
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201721885962.5U Active CN207624043U (en) | 2017-12-28 | 2017-12-28 | Commodity clearance equipment and purchase system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN207624043U (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109166232A (en) * | 2018-07-19 | 2019-01-08 | 深圳市赛亿科技开发有限公司 | A kind of automatic vending machine control method and system |
CN109376656A (en) * | 2018-10-24 | 2019-02-22 | 弭吉荣 | A kind of method and system for differentiating type of goods and number of computations by shooting style |
CN109741551A (en) * | 2018-12-24 | 2019-05-10 | 青岛海信智能商用系统股份有限公司 | A kind of commodity identification settlement method, apparatus and system |
-
2017
- 2017-12-28 CN CN201721885962.5U patent/CN207624043U/en active Active
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109166232A (en) * | 2018-07-19 | 2019-01-08 | 深圳市赛亿科技开发有限公司 | A kind of automatic vending machine control method and system |
CN109376656A (en) * | 2018-10-24 | 2019-02-22 | 弭吉荣 | A kind of method and system for differentiating type of goods and number of computations by shooting style |
CN109741551A (en) * | 2018-12-24 | 2019-05-10 | 青岛海信智能商用系统股份有限公司 | A kind of commodity identification settlement method, apparatus and system |
CN109741551B (en) * | 2018-12-24 | 2021-09-24 | 青岛海信智能商用系统股份有限公司 | Commodity identification settlement method, device and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20230013957A1 (en) | Non-Scan Loss Verification at Self-Checkout Terminal | |
US11514497B2 (en) | Method of using, apparatus, product, and system for a no touch point-of-sale self-checkout | |
TWI681352B (en) | Order information determination method and device | |
US10290031B2 (en) | Method and system for automated retail checkout using context recognition | |
US10846678B2 (en) | Self-service product return using computer vision and Artificial Intelligence | |
WO2015140853A1 (en) | Pos terminal device, pos system, product recognition method, and non-transient computer-readable medium having program stored thereon | |
WO2016158748A1 (en) | Payment system, payment device, program, and payment method | |
JP5704279B1 (en) | Associated program and information processing apparatus | |
CN207624043U (en) | Commodity clearance equipment and purchase system | |
CN110050284A (en) | Register system in a kind of automatic shop | |
CN109559453A (en) | Human-computer interaction device and its application for Automatic-settlement | |
US10372998B2 (en) | Object recognition for bottom of basket detection | |
US10383461B2 (en) | System of control and identification of goods in a shop | |
US20220092597A1 (en) | Frictionless and unassisted return processing | |
KR102120554B1 (en) | System for security and automatic payment in manned or unmanned store | |
CN108399693A (en) | A kind of unmanned supermarket system based on smart shopper basket/vehicle | |
US11854068B2 (en) | Frictionless inquiry processing | |
JP2023162229A (en) | Monitoring device and program | |
US11157917B2 (en) | Self-service product return using computer vision and artificial intelligence | |
JP7181766B2 (en) | Product payment system, product transport cart and product payment method | |
JP5962747B2 (en) | Associated program and information processing apparatus | |
CN113516469B (en) | Unmanned supermarket vending system and vending method based on deep learning and eye tracking | |
CN110264647A (en) | Payment amount instant playback platform | |
US20220270061A1 (en) | System and method for indicating payment method availability on a smart shopping bin | |
KR102303495B1 (en) | Unmanned management in multi-use facility and System for secure identification payment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GR01 | Patent grant | ||
GR01 | Patent grant |