CN110009836A - The system and method for deep learning based on EO-1 hyperion photography technology - Google Patents
The system and method for deep learning based on EO-1 hyperion photography technology Download PDFInfo
- Publication number
- CN110009836A CN110009836A CN201910255288.XA CN201910255288A CN110009836A CN 110009836 A CN110009836 A CN 110009836A CN 201910255288 A CN201910255288 A CN 201910255288A CN 110009836 A CN110009836 A CN 110009836A
- Authority
- CN
- China
- Prior art keywords
- shopper
- shopping
- door lock
- movement track
- facial information
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/30—Individual registration on entry or exit not involving the use of a pass
- G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
- G07C9/37—Individual registration on entry or exit not involving the use of a pass in combination with an identity check using biometric data, e.g. fingerprints, iris scans or voice recognition
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/30—Individual registration on entry or exit not involving the use of a pass
- G07C9/38—Individual registration on entry or exit not involving the use of a pass with central registration
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07G—REGISTERING THE RECEIPT OF CASH, VALUABLES, OR TOKENS
- G07G1/00—Cash registers
- G07G1/0036—Checkout procedures
- G07G1/0045—Checkout procedures with a code reader for reading of an identifying code of the article to be registered, e.g. barcode reader or radio-frequency identity [RFID] reader
Abstract
The present invention provides a kind of deep learning system and method based on EO-1 hyperion photography technology, this method comprises: the facial information of storage user and identity information corresponding with the facial information；Face shots are carried out to shopper using recognition of face camera and obtain the facial information of shopper；It is compared with the facial information of the user of storage, when there are when the facial information of the shopper, open shopping door lock on server；The movement track of captured in real-time shopper, and the movement track image of shopper is uploaded to server and is stored；The weight amount of variation is sent to server, and reports the position of the shelf；The shopper for being currently at shelf location is obtained according to the movement track image comparison of the position of shelf and shopper；The unit price correspondence for transferring each clearing article is settled accounts together；When detecting the check-out request for completing shopper, shop door lock is opened out.The present invention realizes the accurate monitoring management to the Shopping Behaviors in unmanned supermarket.
Description
Technical field
This application involves the technical field of data processing more particularly to a kind of deep learnings based on EO-1 hyperion photography technology
System and method.
Background technique
With the development of society, the progress of science and technology, industrialization or intelligent industry are gradually substituting artificial industry, either
On Industrial Efficiency, or in the accuracy of control, management and production, it is exclusive that industrialization or intelligent industry suffer from its
Advantage.Supermarket also becomes the industrialization or intelligent industry as the essential industry place of modern life naturally
One of research emphasis, unmanned supermarket, no shop assistant supermarket, also known as unmanned supermarket, the not instead of shop assistant of responsible cash register, one
A payment devices, and present unmanned supermarket, not only unmanned cash register more develop towards directions such as unattended, monitoring.
Unmanned supermarket is the new development trend of the following retail trade, and by the terminal inside supermarket, user can carry out quotient
The selection of product, merchandise news check, self-help settlement etc..According to the needs of function, unmanned supermarket system needs to apply journey with multiple
Sequence/system carries out network docking, such as inventory management system, payment system, subscriber authentication system, delivery system.Currently, in nothing
During the operation of people supermarket, due to unattended, the user in supermarket, can not be constantly to the feelings in supermarket when buying goods
Condition is analyzed, and action message of the user in supermarket also can not be accurately obtained, and is not easy to supermarket's maintenance and the visitor in later period in this way
Family hobby analysis cannot also carry out big data analysis to supermarket user, and then preferably provide hobby commodity push for people
Etc. business.Moreover, can not be in time or from root only by activity of the camera record purchaser in supermarket in supermarket
Prevent some illegal activities on this, cannot accurately verify the shopping type and quantity of shopper, some unnecessary entangle can be caused
Confusingly, it is unfavorable for the control management of supermarket, also increases the operation cost of unmanned supermarket, is unfavorable for the operation pipe of unmanned supermarket
Reason.
Therefore, how to provide the flight scenario that one kind is highly reliable, controls with multiple destination flight communications is this field
Technical problem urgently to be resolved.
Summary of the invention
The application's is designed to provide a kind of deep learning system and method based on EO-1 hyperion photography technology, solves existing
There are unmanned supermarket's monitoring management low efficiency in technology, and the technical problem that accuracy is not high.
In order to achieve the above objectives, the application provides a kind of method of deep learning based on EO-1 hyperion photography technology, comprising:
The facial information of storage user and identity information corresponding with the facial information on the server；
When detecting that shopper enters unmanned supermarket, face shots are carried out to the shopper using recognition of face camera
Obtain the facial information of the shopper；The facial information of the shopper is uploaded to the server, the user with storage
Facial information comparison, when on the server exist when the facial information of the shopper, open do shopping door lock；
Start the movement track of shopper described in infrared ray tracking motor driven CCD dynamic camera head captured in real-time, and will
The movement track image of the shopper is uploaded to the server and stores；
When the weight sensor being arranged on the shelf has detected weight change, the weight amount of variation is sent
The extremely server, and report the position of the shelf；The server is according to the type and single item of the article on the shelf
Weight calculates the type of goods picked and placed and quantity, according to the movement track image of the position of the shelf and the shopper
Comparison obtains the shopper for being currently at the shelf location；
When receiving check-out request, believed according to the identity that the check-out request analyzes the shopper of request checkout
Breath, the type of goods and quantity that will the shopper of request checkout the shopping goods during shopping and request clearing
It is checked, when confirmation, the unit price correspondence for transferring each clearing article is settled accounts together；
When detecting the check-out request for completing the shopper, shop door lock is opened out.
Optionally, wherein this method further include:
Shopping door lock is opened after preset time, and/or detects the shopper for completing clearing from the shopping
After door lock passes through, the shopping door lock is automatically closed.
Shop gate is opened out after preset time, and/or detect complete clearing the shopper from it is described go out shop
After door lock passes through, the shop door lock out is automatically closed.
Optionally, wherein this method further include:
The static movement track image of the shopper is combined into image path, and is uploaded to the server and carries out
Storage；
The image path for analyzing all shopping of the shopper, obtains shopper's purchase number, and by institute
It states purchase number to sort from large to small, the purchase of predetermined quantity is chosen according to the purchase number collating sequence
Hobby article as the shopper.
Optionally, wherein the facial information of the shopper is uploaded to the server, the face with the user of storage
Information comparison, when on the server exist when the facial information of the shopper, open do shopping door lock, are as follows:
The facial information of the shopper is uploaded to the server, is compared with the facial information of the user of storage, when
There are when the facial information of the shopper, open shopping door lock on the server；
When the facial information of the shopper is not present on the server, automatically generates facial information acquisition message and show
Show；
Receive the facial information and identity that acquire the shopper that message uploads in client according to the facial information
Information, and the facial information of the shopper received and identity information are stored on the server.
Optionally, wherein the row of shopper described in starting infrared ray tracking motor driven CCD dynamic camera head captured in real-time
Dynamic rail mark, and the movement track image of the shopper is uploaded to the server and is stored, are as follows:
It is analyzed to obtain the average value of shopper's action rhythm according to the movement track that the history of the shopper is done shopping；
According to the rhythm and frequency of preset shopper action rhythm and infrared ray tracking motor driven CCD dynamic camera head captured in real-time
Corresponding relationship obtains the rhythm and frequency of the captured in real-time of the shopper；
Start infrared ray and track motor driven CCD dynamic camera head, according to the rhythm of the captured in real-time of the shopper and
Frequency drives the movement track of shopper described in the infrared ray tracking motor driven CCD dynamic camera head captured in real-time；
If when the movement track that the history of the no shopper is done shopping, starting infrared ray tracking motor driven CCD dynamic is taken the photograph
As head, the infrared ray is driven to track motor driven CCD dynamic camera head captured in real-time according to preset standard cadence and frequency
The movement track of the shopper；
And the movement track image of the shopper is uploaded to the server and is stored.
On the other hand, the present invention also provides a kind of systems of deep learning based on EO-1 hyperion photography technology, comprising: identity
Information setter, shopping door lock controller, shopper's movement track collector, shopping article logger, shopping are settled accounts device and are gone out
Shop door brake controller；Wherein,
The identity information setter is connected, for storing user on the server with the shopping door lock controller
Facial information and identity information corresponding with the facial information；
The shopping door lock controller, is connected with the identity information setter and shopper's movement track collector,
When for detecting that shopper enters unmanned supermarket, face shots being carried out to the shopper using recognition of face camera and are obtained
The facial information of the shopper；The facial information of the shopper is uploaded to the server, the face with the user of storage
Portion's information comparison, when there are when the facial information of the shopper, open shopping door lock on the server；
Shopper's movement track collector is connected with the shopping door lock controller and shopping article logger,
For starting the movement track of shopper described in infrared ray tracking motor driven CCD dynamic camera head captured in real-time, and will be described
The movement track image of shopper is uploaded to the server and stores；
The shopping article logger is connected with shopper's movement track collector and shopping clearing device, is used for
When the weight sensor being arranged on the shelf has detected weight change, the weight amount of variation is sent to the clothes
Business device, and report the position of the shelf；The server is calculated according to the type of article on the shelf and single item weight
The type of goods and quantity picked and placed out is obtained according to the position of the shelf and the movement track image comparison of the shopper
It is currently at the shopper of the shelf location；
Device is settled accounts in the shopping, and shop door brake controller is connected with the shopping article logger and out, for that ought receive
When check-out request, the identity information of the shopper of request checkout is analyzed according to the check-out request, by request checkout
Shopping goods of the shopper during shopping are checked with the type of goods and quantity for requesting clearing, check nothing
It mistakes, the unit price correspondence for transferring each clearing article is settled accounts together；
The shop door brake controller out, is connected with shopping clearing device, detects the completion shopper for working as
Check-out request when, open out shop door lock.
Optionally, wherein the shopping door lock controller, comprising: shopper's facial information recognition unit, shopping door
Lock opening unit and shopping door lock closing unit；Wherein,
Shopper's facial information recognition unit is connected with the identity information setter and shopping door lock opening unit
It connecing, when for detecting that shopper enters unmanned supermarket, face shots being carried out to the shopper using recognition of face camera
Obtain the facial information of the shopper；
The shopping door lock opening unit, with shopper's facial information recognition unit, shopping door lock closing unit and
Shopper's movement track collector is connected, for the facial information of the shopper to be uploaded to the server, with storage
User facial information comparison, when on the server exist when the facial information of the shopper, open do shopping door lock；
The shopping door lock closing unit is connected with the shopping door lock opening unit, for opening shopping door lock warp
After crossing preset time, and/or after detecting that the shopper for completing clearing passes through from the shopping door lock, be automatically closed described in
Shopping door lock；
The shop door brake controller out, comprising: go out shop gate opening unit and out shop closing gate unit；Wherein,
The shop gate opening unit out, settles accounts device with the shopping and closing gate unit in shop is connected out, use
In when detecting the check-out request for completing the shopper, shop door lock is opened out；
The shop closing gate unit out is connected, for opening out shop gate warp with the shop gate opening unit out
After crossing preset time, and/or detect complete the shoppers of clearing from it is described go out after shop door lock passes through, be automatically closed described
Shop door lock out.
Optionally, wherein the system further include: shopper likes analyzer, with shopper's movement track collector
It is connected, is used for:
The static movement track image of the shopper is combined into image path, and is uploaded to the server and carries out
Storage；
The image path for analyzing all shopping of the shopper, obtains shopper's purchase number, and by institute
It states purchase number to sort from large to small, the purchase of predetermined quantity is chosen according to the purchase number collating sequence
Hobby article as the shopper.
Optionally, wherein the shopping door lock controller, comprising: shopping door lock control control unit and shopper information
Acquisition unit；Wherein,
The shopping door lock controls control unit, with the identity information setter, shopper information acquisition unit and purchase
Object person's movement track collector is connected, and when for detecting that shopper enters unmanned supermarket, utilizes recognition of face camera pair
The shopper carries out face shots and obtains the facial information of the shopper；The facial information of the shopper is uploaded to institute
Server is stated, is compared with the facial information of the user of storage, when, there are when the facial information of the shopper, being opened on the server
Open shopping door lock；
The shopper information acquisition unit is connected, for working as the clothes with shopping door lock control control unit
When the facial information of the shopper is not present on business device, automatically generates facial information acquisition message and show；Receive in client
The facial information and identity information of the shopper that message uploads are acquired according to the facial information, and described in receiving
The facial information and identity information of shopper is stored on the server.
Optionally, wherein shopper's movement track collector, comprising: shopper's movement track shoots rhythm analysis
Unit and shopper's movement track acquisition unit；Wherein,
Shopper's movement track shoots rhythm analytical unit, is connected with shopper's movement track acquisition unit
It connects, the movement track for being done shopping according to the history of the shopper is analyzed to obtain the average value of shopper's action rhythm；
According to the rhythm and frequency of preset shopper action rhythm and infrared ray tracking motor driven CCD dynamic camera head captured in real-time
Corresponding relationship obtains the rhythm and frequency of the captured in real-time of the shopper；
Shopper's movement track acquisition unit is saved with the shopping door lock controller, the shooting of shopper's movement track
It plays analytical unit and shopping article logger is connected, for starting infrared ray tracking motor driven CCD dynamic camera head, according to
The rhythm and frequency of the captured in real-time of the shopper drive the infrared ray tracking motor driven CCD dynamic camera head to clap in real time
Take the photograph the movement track of the shopper；
If when the movement track that the history of the no shopper is done shopping, starting infrared ray tracking motor driven CCD dynamic is taken the photograph
As head, the infrared ray is driven to track motor driven CCD dynamic camera head captured in real-time according to preset standard cadence and frequency
The movement track of the shopper；
And the movement track image of the shopper is uploaded to the server and is stored.
What the system and method for the deep learning based on EO-1 hyperion photography technology of the application was realized has the beneficial effect that:
(1) system and method for the deep learning based on EO-1 hyperion photography technology of the application, design is reasonable, user
Just, by being provided with CCD dynamic camera head, infrared ray tracks being applied in combination for motor and action recognition module, realize to
The tracking at family and analysis to user action, to realize to the monitoring inside unmanned supermarket and be adopted to user's buying behavior
Collection, by being provided with spectrometer, using the EO-1 hyperion combination CCD dynamic camera head of spectrometer, improves to user action track
Capture, so that movement details will not be missed, smart structural design is low in cost, is suitble to be widely popularized.
(2) system and method for the deep learning based on EO-1 hyperion photography technology of the application is imaged real by EO-1 hyperion
Activity trajectory and details of the Shi Jilu shopper in unmanned supermarket store the shopper to server and with server memory storage
Information comparison, realizes the accurate monitoring to the Shopping Behaviors in unmanned supermarket, reduces the supervision cost of unmanned supermarket.
(3) system and method for the deep learning based on EO-1 hyperion photography technology of the application, is sensed by shelf weight
Device records the type and quantity that shopper buys article, in checkout automatically by the article and shelf weight in shopper's shopping cart
The type of merchandize and quantity of sensor record are checked, and can guarantee the shopping type and quantity for accurately settling accounts shopper,
Improve the accuracy of unmanned supermarket's supervision.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, 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
The some embodiments recorded in application can also be obtained according to these attached drawings other attached for those skilled in the art
Figure.
Fig. 1 is the process signal of the method for the first deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention
Figure；
Fig. 2 is the process signal of the method for second of deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention
Figure；
Fig. 3 is the process signal of the method for the third deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention
Figure；
Fig. 4 is the process signal of the method for the 4th kind of deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention
Figure；
Fig. 5 is the principle process schematic diagram of the method for the deep learning based on EO-1 hyperion photography technology in Fig. 4；
Fig. 6 is the process signal of the method for the 5th kind of deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention
Figure；
Fig. 7 is the structural representation of the system of the first deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention
Figure；
Fig. 8 is the structural representation of the system of second of deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention
Figure；
Fig. 9 is the structural representation of the system of the third deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention
Figure；
Figure 10 is that the structure of the system of the 4th kind of deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention is shown
It is intended to；
Figure 11 is that the structure of the system of the 5th kind of deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention is shown
It is intended to.
Specific embodiment
Below with reference to the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Ground description, it is clear that described embodiment is some embodiments of the present application, instead of all the embodiments.Based on the application
In embodiment, those skilled in the art's every other embodiment obtained without making creative work, all
Belong to the range of the application protection.
Embodiment
As shown in Figure 1, the method flow for the first deep learning based on EO-1 hyperion photography technology in the present embodiment shows
It is intended to, the facial information that record shopper is imaged by EO-1 hyperion and the movement track image in unmanned supermarket, in conjunction with shelf
Upper weight sensor obtains shopper's purchase information and completes shopping information verification and clearing verification, ensure that unmanned supermarket
Accuracy, while reducing the management cost of unmanned supermarket.This method comprises the following steps:
Step 101, the facial information for storing user on the server and identity information corresponding with the facial information.
Step 102, detect shopper enter unmanned supermarket when, using recognition of face camera to shopper carry out face
Shooting obtains the facial information of shopper；The facial information of shopper is uploaded to server, is believed with the face of the user of storage
Breath comparison, when there are when the facial information of the shopper, open shopping door lock on server.
Step 103, the movement track of starting infrared ray tracking motor driven CCD dynamic camera head captured in real-time shopper,
And the movement track image of shopper is uploaded to server and is stored.
CCD dynamic camera head is the abbreviation of Charge Coupled Device (charge-coupled device) dynamic camera head,
It is a kind of semiconductor imaging device, thus has that high sensitivity, anti-Qiang Guang, distortion is small, small in size, the service life is long, anti-vibration etc.
Advantage can be identified and be analyzed to things in shooting image convenient for backstage in the dynamic image of shooting clear under EO-1 hyperion.
Step 104, when the weight sensor being arranged on shelf has detected weight change, by the weight amount of variation
It is sent to server, and reports the position of the shelf；Server is according to the type of article on the shelf and single item weight
The type of goods picked and placed and quantity are calculated, is obtained currently according to the movement track image comparison of the position of shelf and shopper
Shopper in shelf location.
The type of article on each shelf, position, quantity, article unit price, single item weight etc. are recorded in the server
And other items information, there is article to be removed or placed into the numerical value change that Shi Douhui causes weight sensor from shelf, according to article
Characteristic information calculates the real time status information of article on available shelf, calculates without manual record, improves unmanned supermarket
Management level and accuracy.
Step 105, when receiving check-out request, according to check-out request analyze request checkout shopper identity believe
Breath carries out shopping goods of the shopper of request checkout during the shopping with the type of goods and quantity for requesting clearing
Verification, when confirmation, the unit price correspondence for transferring each clearing article is settled accounts together.
When clearing, settlement device scans through all items that shopper is chosen, and automatically generates after the time of setting
Check-out request or shopper request to carry out shopping clearing by client (APP), it might even be possible to according to shopper in clearing area
When domain residence time meets or exceeds preset settlement time threshold value, all items chosen automatically to shopper are tied
It calculates, and generates the client that statement of account is sent to shopper, realize intelligentized shopping clearing.
Step 106, when detect complete shopper check-out request when, open out shop door lock.
In some alternative embodiments, as shown in Fig. 2, for second in the present embodiment based on EO-1 hyperion photography technology
The flow diagram of the method for deep learning, unlike Fig. 1, this method further includes following steps:
Step 201 opens shopping door lock after preset time, and/or detects the shopper for completing clearing from shopping
After door lock passes through, shopping door lock is automatically closed.
Step 202 opens out shop gate after preset time, and/or detects the shopper for completing clearing from shop out
After door lock passes through, shop door lock is automatically closed out.
Optionally, as shown in figure 3, for the third deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention
The flow diagram of method, unlike Fig. 1, further includes:
The static movement track image of shopper is combined into image path, and is uploaded to server progress by step 301
Storage.
Step 302, analyze shopper all shopping image path, obtain shopper's purchase number, and by institute
Shopping product number sorts from large to small, and chooses the purchase of predetermined quantity as shopping according to purchase number collating sequence
The hobby article of person.
Optionally, as shown in Figure 4 and Figure 5, Fig. 4 is the 4th kind of depth based on EO-1 hyperion photography technology in the embodiment of the present invention
Spend the flow diagram of the method for study；Fig. 5 is the principle stream of the method for the deep learning based on EO-1 hyperion photography technology in Fig. 4
Journey schematic diagram.Unlike Fig. 1, the facial information of shopper is uploaded to server, is believed with the face of the user of storage
Breath comparison, when, there are when the facial information of the shopper, unlatching shopping door lock, is following steps on server:
The facial information of shopper is uploaded to server by step 401, is compared with the facial information of the user of storage, when
There are when the facial information of the shopper, open shopping door lock on server.
Step 402, when on server be not present the shopper facial information when, automatically generate facial information acquisition message
And it shows.
Step 403 receives the facial information and identity that acquire the shopper that message uploads in client according to facial information
Information, and on the server by the facial information of the shopper received and identity information storage.
It is in Fig. 5, as follows based on unmanned supermarket's control principle based on EO-1 hyperion photography technology, it is solved with system structure
It releases, comprising: shop door 501, recognition of face camera 502, APP identification module 503, door lock 504, CCD dynamic camera head 505, light
Spectrometer 506, action recognition module 507, infrared ray tracking motor 508, shelf weight sensor 509 and accounting device 510, face
Identify that camera 502 and APP identification module 503 and door lock 504 are electrically connected, door lock 504 is at least provided with one group, and CCD is dynamically
Camera 505 is arranged on wall and shelf, and spectrometer 506 and CCD dynamic camera head 505 is arranged in a one-to-one correspondence, action recognition
Module 507 is arranged on the ceiling right above shelf, and action recognition module 507 tracks motor 508 by infrared ray and drives CCD
Dynamic camera head 505 rotates, and the setting of shelf weight sensor 509 electrically connects in shelf surface, accounting device 510 with door lock 504
It connects.Wherein, accounting device 504 includes POS machine, Paper currency identifier, Coin validator and two dimensional code payment code, accounting device 510
It is arranged on 504 top of door lock.
Corresponding control method includes: to push shop door open, and recognition of face camera shoots the face data of user, and
It uploads onto the server and stores and compared with server data, after the failure of user's face data comparison, user can also open hand
Machine audits customer data in such a way that APP identification module identifies customer accounting code information；
Customer data after the approval, open by door lock, and client, which enters, carries out commodity shopping inside unmanned supermarket, infrared ray chases after
Track motor driven CCD dynamic camera head follows the movement of client and rotates；
When client removes the product on shelf, action recognition module identification client takes active work, while shelf weight
Sensor detects weight change, and shelf weight sensor corresponding with commodity counts, in order to carry out core when client settles accounts
It is right；
When user walks about in unmanned supermarket, spectrometer is constantly irradiated user, and CCD dynamic camera head acquires user
Action trail, and multiple still photos are combined into photo track, in order to analyze user behavior and purchase hobby；
Commodity are placed on the merchandise news scanning means in door lock and carry out commodity scanning, commodity number by user in checkout
The data that amount and type are monitored with shelf weight sensor compare, and after correlation data is errorless, are carried out by accounting device
Checkout.
As shown in fig. 6, for the method for the 5th kind of deep learning based on EO-1 hyperion photography technology in the embodiment of the present invention
Flow diagram, unlike Fig. 1, starting infrared ray tracks motor driven CCD dynamic camera head captured in real-time shopper
Movement track, and the movement track image of shopper is uploaded to server and is stored, be following steps:
Step 601 is analyzed to obtain the average value of shopper's action rhythm according to the movement track that the history of shopper is done shopping；
According to the rhythm and frequency of preset shopper action rhythm and infrared ray tracking motor driven CCD dynamic camera head captured in real-time
Corresponding relationship obtains the rhythm and frequency of the captured in real-time of shopper.
Each shopper's action rhythm is also different, and then needs filming frequency corresponding with its rhythm of taking action
The clear shopping action that shopper can be taken, obtains the movement track of shopper to image analysis convenient for subsequent, further
Improve the accuracy of unmanned supermarket's monitoring management.
Step 602, starting infrared ray track motor driven CCD dynamic camera head, according to the section of the captured in real-time of shopper
Play and frequency driving infrared ray tracking motor driven CCD dynamic camera head captured in real-time shopper movement track.
If when the movement track that step 603, the history without shopper are done shopping, starting infrared ray tracking motor driven CCD is dynamic
State camera tracks motor driven CCD dynamic camera head captured in real-time according to preset standard cadence and frequency driving infrared ray
The movement track of shopper.
The movement track image of shopper is uploaded to server and stores by step 604.
Fig. 7 is a kind of structural schematic diagram of the system 700 of the deep learning based on EO-1 hyperion photography technology in the present embodiment,
The method that the system is used to implement the above-mentioned deep learning based on EO-1 hyperion photography technology, which includes: that identity information is set
Set device 701, shopping door lock controller 702, shopper's movement track collector 703, shopping article logger 704, shopping clearing
Device 705 and out shop door brake controller 706.
Wherein, identity information setter 701 is connected with shopping door lock controller 702, uses for storage on the server
The facial information at family and identity information corresponding with the facial information.
Shopping door lock controller 702, is connected with identity information setter 701 and shopper's movement track collector 703,
When for detecting that shopper enters unmanned supermarket, face shots being carried out to shopper using recognition of face camera and are done shopping
The facial information of person；The facial information of shopper is uploaded to server, is compared with the facial information of the user of storage, works as service
There are when the facial information of the shopper, open shopping door lock on device.
Shopper's movement track collector 703 is connected with shopping door lock controller 702 and shopping article logger 704,
For starting the movement track of infrared ray tracking motor driven CCD dynamic camera head captured in real-time shopper, and by shopper's
Movement track image is uploaded to server and is stored.
Shopping article logger 704 is connected with shopper's movement track collector 703 and shopping clearing device 705, is used for
When the weight sensor being arranged on shelf has detected weight change, the weight amount of variation is sent to server, and
Report the position of the shelf；Server calculates the object picked and placed according to the type of article on the shelf and single item weight
Kind class and quantity obtain the purchase for being currently at shelf location according to the movement track image comparison of the position of shelf and shopper
Object person.
Shopping clearing device 705, and shopping article logger 704 and out shop door brake controller 706 is connected, for when receiving
When check-out request, the identity information of the shopper of request checkout is analyzed according to check-out request, and the shopper of request checkout is existed
Shopping goods during the shopping are checked with the type of goods and quantity for requesting clearing, when confirmation, are transferred each
The unit price correspondence of a clearing article is settled accounts together.
Shop door brake controller 706 out is connected, for when the checkout for detecting completion shopper with shopping clearing device 705
When request, shop door lock is opened out.
In some optionally embodiments, as shown in figure 8, being second in this implementation based on EO-1 hyperion photography technology
The structural schematic diagram of the system 800 of deep learning, unlike Fig. 7, door lock controller 702 of doing shopping, comprising: shopper's face
Information identificating unit 721, shopping door lock opening unit 722 and shopping door lock closing unit 723.
Wherein, shopper's facial information recognition unit 721, with identity information setter 701 and shopping door lock opening unit
722 are connected, and when for detecting that shopper enters unmanned supermarket, carry out facial bat to shopper using recognition of face camera
It takes the photograph to obtain the facial information of shopper.
Do shopping door lock opening unit 722, with shopper's facial information recognition unit 721, shopping door lock closing unit 723 and
Shopper's movement track collector 703 is connected, for the facial information of shopper to be uploaded to server, the user with storage
Facial information comparison, when on server exist when the facial information of the shopper, open do shopping door lock.
Shopping door lock closing unit 723 is connected with shopping door lock opening unit 722, for opening shopping door lock process
After preset time, and/or detect complete clearing shopper from shopping door lock pass through after, shopping door lock is automatically closed.
Shop door brake controller 706 out, comprising: go out shop gate opening unit 761 and out shop closing gate unit 762.
Wherein, shop gate opening unit 761 out, shop closing gate unit 762 is connected with shopping clearing device 705 and out,
For opening out shop door lock when detecting the check-out request for completing shopper.
Shop closing gate unit 762 out is connected with shop gate opening unit 761 out, passes through for opening out shop gate
After preset time, and/or detect that after shop door lock passes through out, shop door lock is automatically closed out in the shopper for completing clearing.
In some optionally embodiments, as shown in figure 9, being the third in this implementation based on EO-1 hyperion photography technology
The structural schematic diagram of the system 900 of deep learning, unlike Fig. 7, further includes: shopper likes analyzer 901, with shopping
Person's movement track collector 703 is connected, and is used for: the static movement track image of shopper is combined into image path, and
Server is uploaded to be stored；The image path for analyzing all shopping of shopper, obtains shopper's purchase number, and
Purchase number is sorted from large to small, the purchase conduct of predetermined quantity is chosen according to purchase number collating sequence
The hobby article of shopper.
It is that the 4th kind in this implementation is based on EO-1 hyperion photography technology as shown in Figure 10 in some optionally embodiments
Deep learning system 1000 structural schematic diagram, unlike Fig. 7, do shopping door lock controller 702, comprising: shopping door
Lock controls control unit 724 and shopper information acquisition unit 725；Wherein,
Door lock of doing shopping controls control unit 724, with identity information setter 701, shopper information acquisition unit 725 and purchase
Object person's movement track collector 703 is connected, and when for detecting that shopper enters unmanned supermarket, utilizes recognition of face camera
Face shots are carried out to shopper and obtain the facial information of shopper；The facial information of shopper is uploaded to server, and is deposited
The facial information of the user of storage compares, when there are when the facial information of the shopper, open shopping door lock on server.
Shopper information acquisition unit 725 is connected, for working as server with shopping door lock control control unit 724
There is no when the facial information of the shopper, automatically generate facial information acquisition message and show；Receive in client according to face
The facial information and identity information for the shopper that information collection message in portion's uploads, and by the facial information of the shopper received and
Identity information stores on the server.
It is that the 4th kind in this implementation is based on EO-1 hyperion photography technology as shown in figure 11 in some optionally embodiments
Deep learning system 1100 structural schematic diagram, unlike Fig. 7, shopper's movement track collector 703, comprising:
Shopper's movement track shoots rhythm analytical unit 731 and shopper's movement track acquisition unit 732.
Wherein, shopper's movement track shoots rhythm analytical unit 731, with 732 phase of shopper's movement track acquisition unit
Connection, the movement track for being done shopping according to the history of shopper are analyzed to obtain the average value of shopper's action rhythm；According to pre-
If shopper take action rhythm and infrared ray tracking motor driven CCD dynamic camera head captured in real-time rhythm and the corresponding pass of frequency
System, obtains the rhythm and frequency of the captured in real-time of shopper.
Shopper's movement track acquisition unit 732 shoots rhythm with shopping door lock controller 702, shopper's movement track
Analytical unit 731 and shopping article logger 704 are connected, for starting infrared ray tracking motor driven CCD dynamic camera head,
According to the rhythm of the captured in real-time of shopper and frequency driving infrared ray tracking motor driven CCD dynamic camera head captured in real-time purchase
The movement track of object person；If when the movement track that the history of no shopper is done shopping, starting infrared ray tracking motor driven CCD dynamic
Camera, according to preset standard cadence and frequency driving infrared ray tracking motor driven CCD dynamic camera head captured in real-time purchase
The movement track of object person；And the movement track image of shopper is uploaded to server and is stored.
The system and method for the deep learning based on EO-1 hyperion photography technology in the present embodiment realizes following beneficial
Effect:
(1) system and method for the deep learning based on EO-1 hyperion photography technology in the present embodiment, design rationally, use
Convenient, by being provided with CCD dynamic camera head, infrared ray tracks motor and action recognition module is applied in combination, and realizes pair
The tracking of user and analysis to user action, to realize to the monitoring inside unmanned supermarket and be adopted to user's buying behavior
Collection, by being provided with spectrometer, using the EO-1 hyperion combination CCD dynamic camera head of spectrometer, improves to user action track
Capture, so that movement details will not be missed, smart structural design is low in cost, is suitble to be widely popularized.
(2) system and method for the deep learning based on EO-1 hyperion photography technology in the present embodiment, is taken the photograph by EO-1 hyperion
As recording activity trajectory and details of the shopper in unmanned supermarket in real time, the purchase to server and with server memory storage is stored
The comparison of object person's information, realizes the accurate monitoring to the Shopping Behaviors in unmanned supermarket, reduces the supervision cost of unmanned supermarket.
(3) system and method for the deep learning based on EO-1 hyperion photography technology in the present embodiment, passes through shelf weight
Sensor records the type and quantity that shopper buys article, in checkout automatically by the article and shelf in shopper's shopping cart
The type of merchandize and quantity of weight sensor record are checked, and can guarantee the shopping type sum number for accurately settling accounts shopper
Amount improves the accuracy of unmanned supermarket's supervision.
Although the preferred embodiment of the application has been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the application range.Obviously, those skilled in the art can be to the application
Various modification and variations are carried out without departing from spirit and scope.If in this way, these modifications and variations of the application
Belong within the scope of the claim of this application and its equivalent technologies, then the application is also intended to encompass these modification and variations and exists
It is interior.
Claims (10)
1. a kind of method of the deep learning based on EO-1 hyperion photography technology characterized by comprising
The facial information of storage user and identity information corresponding with the facial information on the server；
When detecting that shopper enters unmanned supermarket, face shots are carried out to the shopper using recognition of face camera and are obtained
The facial information of the shopper；The facial information of the shopper is uploaded to the server, the face with the user of storage
Portion's information comparison, when there are when the facial information of the shopper, open shopping door lock on the server；
Start the movement track of shopper described in infrared ray tracking motor driven CCD dynamic camera head captured in real-time, and will be described
The movement track image of shopper is uploaded to the server and stores；
When the weight sensor being arranged on the shelf has detected weight change, the weight amount of variation is sent to institute
Server is stated, and reports the position of the shelf；The server is according to the type of article on the shelf and single item weight
The type of goods picked and placed and quantity are calculated, according to the movement track image comparison of the position of the shelf and the shopper
Obtain the shopper for being currently at the shelf location；
When receiving check-out request, the identity information of the shopper of request checkout is analyzed according to the check-out request, it will
Shopping goods of the shopper of checkout during the shopping are requested to carry out with the type of goods and quantity for requesting clearing
Verification, when confirmation, the unit price correspondence for transferring each clearing article is settled accounts together；
When detecting the check-out request for completing the shopper, shop door lock is opened out.
2. the method for the deep learning according to claim 1 based on EO-1 hyperion photography technology, which is characterized in that also wrap
It includes:
Shopping door lock is opened after preset time, and/or detects the shopper for completing clearing from the shopping door lock
By rear, the shopping door lock is automatically closed；
Shop gate is opened out after preset time, and/or detect complete clearing the shopper from it is described go out shop door lock
By rear, be automatically closed it is described go out shop door lock.
3. the method for the deep learning according to claim 1 based on EO-1 hyperion photography technology, which is characterized in that also wrap
It includes:
The static movement track image of the shopper is combined into image path, and is uploaded to the server and is deposited
Storage；
The image path for analyzing all shopping of the shopper, obtains shopper's purchase number, and by the institute
Shopping product number sorts from large to small, and the purchase conduct of predetermined quantity is chosen according to the purchase number collating sequence
The hobby article of the shopper.
4. the method for the deep learning according to claim 1 based on EO-1 hyperion photography technology, which is characterized in that will be described
The facial information of shopper is uploaded to the server, compares with the facial information of the user of storage, deposits when on the server
In the facial information of the shopper, shopping door lock is opened, are as follows:
The facial information of the shopper is uploaded to the server, is compared with the facial information of the user of storage, when described
There are when the facial information of the shopper, open shopping door lock on server；
When the facial information of the shopper is not present on the server, automatically generates facial information acquisition message and show；
Receive the facial information and identity information that acquire the shopper that message uploads in client according to the facial information,
And the facial information of the shopper received and identity information are stored on the server.
5. the method for the deep learning as claimed in any of claims 1 to 4 based on EO-1 hyperion photography technology, special
Sign is, the movement track of shopper described in starting infrared ray tracking motor driven CCD dynamic camera head captured in real-time, and by institute
The movement track image for stating shopper is uploaded to the server and stores, are as follows:
It is analyzed to obtain the average value of shopper's action rhythm according to the movement track that the history of the shopper is done shopping；According to
Preset shopper's action rhythm is corresponding with the rhythm of infrared ray tracking motor driven CCD dynamic camera head captured in real-time and frequency
Relationship obtains the rhythm and frequency of the captured in real-time of the shopper；
Start infrared ray and track motor driven CCD dynamic camera head, according to the rhythm and frequency of the captured in real-time of the shopper
Drive the movement track of shopper described in the infrared ray tracking motor driven CCD dynamic camera head captured in real-time；
If when the movement track that the history of the no shopper is done shopping, starting infrared ray tracks motor driven CCD dynamic camera head,
It drives the infrared ray to track described in motor driven CCD dynamic camera head captured in real-time according to preset standard cadence and frequency to purchase
The movement track of object person；
And the movement track image of the shopper is uploaded to the server and is stored.
6. a kind of system of the deep learning based on EO-1 hyperion photography technology characterized by comprising identity information setter,
Shopping door lock controller, shopper's movement track collector, shopping article logger, shopping clearing device and shop door lock control out
Device；Wherein,
The identity information setter is connected, for storing the face of user on the server with the shopping door lock controller
Portion's information and identity information corresponding with the facial information；
The shopping door lock controller, is connected with the identity information setter and shopper's movement track collector, is used for
When detecting that shopper enters unmanned supermarket, shopper progress face shots are obtained using recognition of face camera described
The facial information of shopper；The facial information of the shopper is uploaded to the server, is believed with the face of the user of storage
Breath comparison, when there are when the facial information of the shopper, open shopping door lock on the server；
Shopper's movement track collector is connected with the shopping door lock controller and shopping article logger, is used for
Start the movement track of shopper described in infrared ray tracking motor driven CCD dynamic camera head captured in real-time, and by the shopping
The movement track image of person is uploaded to the server and stores；
The shopping article logger is connected with shopper's movement track collector and shopping clearing device, sets for working as
It sets when the weight sensor on the shelf has detected weight change, the weight amount of variation is sent to the service
Device, and report the position of the shelf；The server is calculated according to the type of article on the shelf and single item weight
The type of goods and quantity picked and placed is worked as according to the position of the shelf and the movement track image comparison of the shopper
The preceding shopper in the shelf location；
Device is settled accounts in the shopping, and shop door brake controller is connected with the shopping article logger and out, for that ought receive checkout
When request, the identity information of the shopper of request checkout is analyzed according to the check-out request, it will be described in request checkout
Shopping goods of the shopper during shopping are checked with the type of goods and quantity for requesting clearing, confirmation
When, the unit price correspondence for transferring each clearing article is settled accounts together；
The shop door brake controller out is connected with shopping clearing device, detects the knot for completing the shopper for working as
When account is requested, shop door lock is opened out.
7. the system of the deep learning according to claim 6 based on EO-1 hyperion photography technology, which is characterized in that the institute
State shopping door lock controller, comprising: shopper's facial information recognition unit, shopping door lock opening unit and shopping door lock are closed single
Member；Wherein,
Shopper's facial information recognition unit is connected with the identity information setter and shopping door lock opening unit,
When for detecting that shopper enters unmanned supermarket, face shots being carried out to the shopper using recognition of face camera and are obtained
The facial information of the shopper；
The shopping door lock opening unit, with shopper's facial information recognition unit, shopping door lock closing unit and shopping
Person's movement track collector is connected, for the facial information of the shopper to be uploaded to the server, the use with storage
The facial information at family compares, when there are when the facial information of the shopper, open shopping door lock on the server；
The shopping door lock closing unit is connected with the shopping door lock opening unit, for opening shopping door lock by pre-
If the shopping is automatically closed after the time, and/or after detecting that the shopper for completing clearing passes through from the shopping door lock
Door lock；
The shop door brake controller out, comprising: go out shop gate opening unit and out shop closing gate unit；Wherein,
The shop gate opening unit out, settles accounts device with the shopping and closing gate unit in shop is connected out, for working as
When detecting the check-out request for completing the shopper, shop door lock is opened out；
The shop closing gate unit out is connected, for opening out shop gate by pre- with the shop gate opening unit out
If after the time, and/or detect complete the shoppers of clearing from it is described go out after shop door lock passes through, be automatically closed it is described go out shop
Door lock.
8. the system of the deep learning according to claim 6 based on EO-1 hyperion photography technology, which is characterized in that also wrap
Include: shopper likes analyzer, is connected with shopper's movement track collector, is used for:
The static movement track image of the shopper is combined into image path, and is uploaded to the server and is deposited
Storage；
The image path for analyzing all shopping of the shopper, obtains shopper's purchase number, and by the institute
Shopping product number sorts from large to small, and the purchase conduct of predetermined quantity is chosen according to the purchase number collating sequence
The hobby article of the shopper.
9. the system of the deep learning according to claim 6 based on EO-1 hyperion photography technology, which is characterized in that the purchase
Object door lock controller, comprising: shopping door lock control control unit and shopper information acquisition unit；Wherein,
The shopping door lock controls control unit, with the identity information setter, shopper information acquisition unit and shopper
Movement track collector is connected, when for detecting that shopper enters unmanned supermarket, using recognition of face camera to described
Shopper carries out face shots and obtains the facial information of the shopper；The facial information of the shopper is uploaded to the clothes
Business device is compared with the facial information of the user of storage, when, there are when the facial information of the shopper, unlatching is purchased on the server
Object door lock；
The shopper information acquisition unit is connected, for working as the server with shopping door lock control control unit
It is upper that there is no when the facial information of the shopper, automatically generate facial information acquisition message and show；Receive basis in client
The facial information and identity information for the shopper that the facial information acquisition message uploads, and the shopping that will be received
The facial information and identity information of person is stored on the server.
10. the system of the deep learning according to any one of claims 6 to 9 based on EO-1 hyperion photography technology,
It is characterized in that, shopper's movement track collector, comprising: shopper's movement track shoots rhythm analytical unit and shopper
Movement track acquisition unit；Wherein,
Shopper's movement track shoots rhythm analytical unit, is connected with shopper's movement track acquisition unit, uses
It analyzes to obtain the average value of shopper's action rhythm in the movement track done shopping according to the history of the shopper；According to pre-
If shopper take action rhythm and infrared ray tracking motor driven CCD dynamic camera head captured in real-time rhythm and the corresponding pass of frequency
System, obtains the rhythm and frequency of the captured in real-time of the shopper；
Shopper's movement track acquisition unit, with the shopping door lock controller, shopper's movement track shooting rhythm point
Analysis unit and shopping article logger are connected, for starting infrared ray tracking motor driven CCD dynamic camera head, according to described
The rhythm and frequency of the captured in real-time of shopper drive the infrared ray to track motor driven CCD dynamic camera head captured in real-time institute
State the movement track of shopper；
If when the movement track that the history of the no shopper is done shopping, starting infrared ray tracks motor driven CCD dynamic camera head,
It drives the infrared ray to track described in motor driven CCD dynamic camera head captured in real-time according to preset standard cadence and frequency to purchase
The movement track of object person；
And the movement track image of the shopper is uploaded to the server and is stored.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910255288.XA CN110009836A (en) | 2019-03-29 | 2019-03-29 | The system and method for deep learning based on EO-1 hyperion photography technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910255288.XA CN110009836A (en) | 2019-03-29 | 2019-03-29 | The system and method for deep learning based on EO-1 hyperion photography technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110009836A true CN110009836A (en) | 2019-07-12 |
Family
ID=67169135
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910255288.XA Pending CN110009836A (en) | 2019-03-29 | 2019-03-29 | The system and method for deep learning based on EO-1 hyperion photography technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110009836A (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110555928A (en) * | 2019-08-15 | 2019-12-10 | 创新奇智(成都)科技有限公司 | Intelligent store entrance guard method based on face recognition and settlement method thereof |
| US10621444B1 (en) * | 2019-10-25 | 2020-04-14 | 7-Eleven, Inc. | Action detection during image tracking |
| US11003918B1 (en) | 2019-10-25 | 2021-05-11 | 7-Eleven, Inc. | Event trigger based on region-of-interest near hand-shelf interaction |
| US11023740B2 (en) | 2019-10-25 | 2021-06-01 | 7-Eleven, Inc. | System and method for providing machine-generated tickets to facilitate tracking |
| US11023741B1 (en) | 2019-10-25 | 2021-06-01 | 7-Eleven, Inc. | Draw wire encoder based homography |
| US11030756B2 (en) | 2018-10-26 | 2021-06-08 | 7-Eleven, Inc. | System and method for position tracking using edge computing |
| US11113541B2 (en) | 2019-10-25 | 2021-09-07 | 7-Eleven, Inc. | Detection of object removal and replacement from a shelf |
| US11301691B2 (en) | 2019-10-25 | 2022-04-12 | 7-Eleven, Inc. | Homography error correction using sensor locations |
| WO2022115845A1 (en) * | 2020-11-25 | 2022-06-02 | 7-Eleven, Inc. | System and method for providing machine-generated tickets to facilitate tracking |
| US11403852B2 (en) | 2019-10-25 | 2022-08-02 | 7-Eleven, Inc. | Object detection based on wrist-area region-of-interest |
| CN114973548A (en) * | 2022-05-23 | 2022-08-30 | 广东海洋大学 | Unmanned smart library management system and method based on image recognition |
| US11450011B2 (en) | 2019-10-25 | 2022-09-20 | 7-Eleven, Inc. | Adaptive item counting algorithm for weight sensor using sensitivity analysis of the weight sensor |
| US11501454B2 (en) | 2019-10-25 | 2022-11-15 | 7-Eleven, Inc. | Mapping wireless weight sensor array for item detection and identification |
| US11551454B2 (en) | 2019-10-25 | 2023-01-10 | 7-Eleven, Inc. | Homography error correction using marker locations |
| US11557124B2 (en) | 2019-10-25 | 2023-01-17 | 7-Eleven, Inc. | Homography error correction |
| US11587243B2 (en) | 2019-10-25 | 2023-02-21 | 7-Eleven, Inc. | System and method for position tracking using edge computing |
| US11674792B2 (en) | 2019-10-25 | 2023-06-13 | 7-Eleven, Inc. | Sensor array with adjustable camera positions |
| CN116471383A (en) * | 2023-06-20 | 2023-07-21 | 深圳市泽威信息科技有限公司 | Display method and device for unattended store monitoring interface |
| US11887372B2 (en) | 2019-10-25 | 2024-01-30 | 7-Eleven, Inc. | Image-based self-serve beverage detection and assignment |
| US11887337B2 (en) | 2019-10-25 | 2024-01-30 | 7-Eleven, Inc. | Reconfigurable sensor array |
| US11893757B2 (en) | 2019-10-25 | 2024-02-06 | 7-Eleven, Inc. | Self-serve beverage detection and assignment |
| US11893759B2 (en) | 2019-10-24 | 2024-02-06 | 7-Eleven, Inc. | Homography error correction using a disparity mapping |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017097536A (en) * | 2015-11-20 | 2017-06-01 | ソニー株式会社 | Image processing apparatus and method |
| CN107330684A (en) * | 2017-07-06 | 2017-11-07 | 广州联业商用机器人科技股份有限公司 | A kind of intelligent unmanned shop of management and control in high in the clouds and its automatic settlement method |
| CN107481284A (en) * | 2017-08-25 | 2017-12-15 | 京东方科技集团股份有限公司 | Method, apparatus, terminal and the system of target tracking path accuracy measurement |
| CN107705129A (en) * | 2017-09-15 | 2018-02-16 | 泾县麦蓝网络技术服务有限公司 | A kind of shopping settlement method and system applied to physical retail store |
| CN108492482A (en) * | 2017-12-18 | 2018-09-04 | 上海云拿智能科技有限公司 | Kinds of goods monitoring system and monitoring method |
| CN109409175A (en) * | 2017-08-16 | 2019-03-01 | 图灵通诺（北京）科技有限公司 | Settlement method, device and system |
-
2019
- 2019-03-29 CN CN201910255288.XA patent/CN110009836A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017097536A (en) * | 2015-11-20 | 2017-06-01 | ソニー株式会社 | Image processing apparatus and method |
| CN107330684A (en) * | 2017-07-06 | 2017-11-07 | 广州联业商用机器人科技股份有限公司 | A kind of intelligent unmanned shop of management and control in high in the clouds and its automatic settlement method |
| CN109409175A (en) * | 2017-08-16 | 2019-03-01 | 图灵通诺（北京）科技有限公司 | Settlement method, device and system |
| CN107481284A (en) * | 2017-08-25 | 2017-12-15 | 京东方科技集团股份有限公司 | Method, apparatus, terminal and the system of target tracking path accuracy measurement |
| CN107705129A (en) * | 2017-09-15 | 2018-02-16 | 泾县麦蓝网络技术服务有限公司 | A kind of shopping settlement method and system applied to physical retail store |
| CN108492482A (en) * | 2017-12-18 | 2018-09-04 | 上海云拿智能科技有限公司 | Kinds of goods monitoring system and monitoring method |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11030756B2 (en) | 2018-10-26 | 2021-06-08 | 7-Eleven, Inc. | System and method for position tracking using edge computing |
| US11501455B2 (en) | 2018-10-26 | 2022-11-15 | 7-Eleven, Inc. | System and method for position tracking using edge computing |
| CN110555928A (en) * | 2019-08-15 | 2019-12-10 | 创新奇智(成都)科技有限公司 | Intelligent store entrance guard method based on face recognition and settlement method thereof |
| US11893759B2 (en) | 2019-10-24 | 2024-02-06 | 7-Eleven, Inc. | Homography error correction using a disparity mapping |
| US11450011B2 (en) | 2019-10-25 | 2022-09-20 | 7-Eleven, Inc. | Adaptive item counting algorithm for weight sensor using sensitivity analysis of the weight sensor |
| US11551454B2 (en) | 2019-10-25 | 2023-01-10 | 7-Eleven, Inc. | Homography error correction using marker locations |
| US11023741B1 (en) | 2019-10-25 | 2021-06-01 | 7-Eleven, Inc. | Draw wire encoder based homography |
| US11003918B1 (en) | 2019-10-25 | 2021-05-11 | 7-Eleven, Inc. | Event trigger based on region-of-interest near hand-shelf interaction |
| US11113541B2 (en) | 2019-10-25 | 2021-09-07 | 7-Eleven, Inc. | Detection of object removal and replacement from a shelf |
| US11301691B2 (en) | 2019-10-25 | 2022-04-12 | 7-Eleven, Inc. | Homography error correction using sensor locations |
| US11900724B2 (en) | 2019-10-25 | 2024-02-13 | 7-Eleven, Inc. | System and method for providing machine-generated tickets to facilitate tracking |
| US11403852B2 (en) | 2019-10-25 | 2022-08-02 | 7-Eleven, Inc. | Object detection based on wrist-area region-of-interest |
| US11423657B2 (en) | 2019-10-25 | 2022-08-23 | 7-Eleven, Inc. | Event trigger based on region-of-interest near hand-shelf interaction |
| US10621444B1 (en) * | 2019-10-25 | 2020-04-14 | 7-Eleven, Inc. | Action detection during image tracking |
| US10853663B1 (en) * | 2019-10-25 | 2020-12-01 | 7-Eleven, Inc. | Action detection during image tracking |
| US11501454B2 (en) | 2019-10-25 | 2022-11-15 | 7-Eleven, Inc. | Mapping wireless weight sensor array for item detection and identification |
| US10685237B1 (en) * | 2019-10-25 | 2020-06-16 | 7-Eleven, Inc. | Action detection during image tracking |
| US11023740B2 (en) | 2019-10-25 | 2021-06-01 | 7-Eleven, Inc. | System and method for providing machine-generated tickets to facilitate tracking |
| US11557124B2 (en) | 2019-10-25 | 2023-01-17 | 7-Eleven, Inc. | Homography error correction |
| US11587243B2 (en) | 2019-10-25 | 2023-02-21 | 7-Eleven, Inc. | System and method for position tracking using edge computing |
| US11674792B2 (en) | 2019-10-25 | 2023-06-13 | 7-Eleven, Inc. | Sensor array with adjustable camera positions |
| US11893757B2 (en) | 2019-10-25 | 2024-02-06 | 7-Eleven, Inc. | Self-serve beverage detection and assignment |
| US11721029B2 (en) | 2019-10-25 | 2023-08-08 | 7-Eleven, Inc. | Draw wire encoder based homography |
| US11756213B2 (en) | 2019-10-25 | 2023-09-12 | 7-Eleven, Inc. | Object detection based on wrist-area region-of-interest |
| US11887337B2 (en) | 2019-10-25 | 2024-01-30 | 7-Eleven, Inc. | Reconfigurable sensor array |
| US11836957B2 (en) | 2019-10-25 | 2023-12-05 | 7-Eleven, Inc. | Event trigger based on region-of-interest near hand-shelf interaction |
| US11887372B2 (en) | 2019-10-25 | 2024-01-30 | 7-Eleven, Inc. | Image-based self-serve beverage detection and assignment |
| WO2022115845A1 (en) * | 2020-11-25 | 2022-06-02 | 7-Eleven, Inc. | System and method for providing machine-generated tickets to facilitate tracking |
| CN114973548A (en) * | 2022-05-23 | 2022-08-30 | 广东海洋大学 | Unmanned smart library management system and method based on image recognition |
| CN116471383B (en) * | 2023-06-20 | 2023-11-03 | 深圳市泽威信息科技有限公司 | Display method and device for unattended store monitoring interface |
| CN116471383A (en) * | 2023-06-20 | 2023-07-21 | 深圳市泽威信息科技有限公司 | Display method and device for unattended store monitoring interface |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110009836A (en) | The system and method for deep learning based on EO-1 hyperion photography technology | |
| US11790433B2 (en) | Constructing shopper carts using video surveillance | |
| CN106781014B (en) | Automatic vending machine and its operation method | |
| CN206757798U (en) | Automatic vending machine | |
| CN106408369B (en) | Method for intelligently identifying commodity information in shopping cart | |
| CN108491799B (en) | Intelligent sales counter commodity management method and system based on image recognition | |
| CN108320404B (en) | Commodity identification method and device based on neural network and self-service cash register | |
| CN108806074B (en) | Shopping information generation method and device and storage medium | |
| US11176597B2 (en) | Associating shoppers together | |
| CN108230559A (en) | Automatic vending device, operation method thereof and automatic vending system | |
| CN107464116A (en) | A kind of order settlement method and system | |
| CN108269371B (en) | Automatic commodity settlement method and device and self-service cash register | |
| CN109272347A (en) | A kind of statistical analysis technique and system of shops's volume of the flow of passengers | |
| US10252149B2 (en) | Interactive vending system and method | |
| CN109979130A (en) | A kind of commodity automatic identification and clearing sales counter, method and system | |
| CN108235771A (en) | Self-help shopping method and system, electronic equipment, program product based on high in the clouds | |
| CN111263224B (en) | Video processing method and device and electronic equipment | |
| CN108198315A (en) | A kind of auth method and authentication means | |
| CN111222870B (en) | Settlement method, device and system | |
| CN108399693A (en) | A kind of unmanned supermarket system based on smart shopper basket/vehicle | |
| US20210398097A1 (en) | Method, a device and a system for checkout | |
| AU2016403974A1 (en) | Network interactive system for selecting object for sale | |
| CN110647825A (en) | Method, device and equipment for determining unmanned supermarket articles and storage medium | |
| CN109447756A (en) | A kind of unattended supermarket shopping method and system | |
| CN111178860A (en) | Settlement method, device, equipment and storage medium for unmanned convenience store |
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: 20190712 |
|
| RJ01 | Rejection of invention patent application after publication |