CN108510677A - A kind of cash register system and data processing method - Google Patents
A kind of cash register system and data processing method Download PDFInfo
- Publication number
- CN108510677A CN108510677A CN201810235977.XA CN201810235977A CN108510677A CN 108510677 A CN108510677 A CN 108510677A CN 201810235977 A CN201810235977 A CN 201810235977A CN 108510677 A CN108510677 A CN 108510677A
- Authority
- CN
- China
- Prior art keywords
- server
- picture
- menu name
- correspondence
- target vegetable
- 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.)
- Withdrawn
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 19
- 235000013311 vegetables Nutrition 0.000 claims abstract description 89
- 239000011159 matrix material Substances 0.000 claims description 40
- 238000000605 extraction Methods 0.000 claims description 29
- 238000000034 method Methods 0.000 claims description 11
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 241000208340 Araliaceae Species 0.000 claims 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 1
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 1
- 235000008434 ginseng Nutrition 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 6
- 230000003340 mental effect Effects 0.000 abstract description 5
- 238000012545 processing Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 3
- 235000012149 noodles Nutrition 0.000 description 3
- 235000014347 soups Nutrition 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 235000012054 meals Nutrition 0.000 description 1
- 235000011888 snacks Nutrition 0.000 description 1
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07G—REGISTERING THE RECEIPT OF CASH, VALUABLES, OR TOKENS
- G07G1/00—Cash registers
- G07G1/12—Cash registers electronically operated
- G07G1/14—Systems including one or more distant stations co-operating with a central processing unit
- G07G1/145—PLU-management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/12—Hotels or restaurants
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Biomedical Technology (AREA)
- General Business, Economics & Management (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of cash register system and data processing method, which includes:At least one video capture device, terminal device, server and at least one display screen, wherein, at least one video capture device is connect with terminal device, terminal device is connect with server and at least one display screen, in embodiments of the present invention, due to remembering the unit price of each vegetable without cashier, and total price is obtained by mental arithmetic or calculator without cashier, the duration consumed when to advantageously reduce clearing, and then be conducive to improve settlement efficiency, simultaneously, (artificial clearing) compared with the existing technology, the embodiment of the present invention advantageously reduces the probability for mistake occur.
Description
Technical field
The present invention relates to fields of communication technology, and in particular to a kind of cash register system and data processing method.
Background technology
In the prior art, client when snack bar is settled accounts, settle accounts by the unit price of the memory based each vegetable of cashier,
In clearing, cashier obtains total price by mental arithmetic or calculator, but above-mentioned settlement method is time-consuming longer, and clearing is caused to be imitated
Rate is relatively low.
Invention content
The purpose of the present invention is to provide a kind of cash register system and data processing methods, to solve to settle accounts in the prior art
Less efficient problem.
To achieve the above object, a kind of technical solution of the invention is to provide a kind of cash register system, the cash register system packet
It includes:At least one video capture device, terminal device, server and at least one display screen, wherein at least one video
Collecting device is connect with the terminal device, and the terminal device is connect with the server and at least one display screen;
The video capture device, for acquire include carrier and the target vegetable in the carrier video
Data;And for the video data to be sent to the terminal device;
The terminal device obtains the corresponding picture of designated frame in the video data;And for the picture to be sent out
Give the server;
The server, for according to the picture, determining the target vegetable in the picture in the carrier
On location parameter and the target vegetable menu name;And for the location parameter and the menu name to be sent out
Give the terminal device;
The terminal device is additionally operable to, according to the menu name and preset first correspondence, determine the vegetable
The corresponding monovalent information of title, wherein first correspondence is the correspondence of menu name and monovalent information;And
For according to the monovalent information, obtaining the corresponding total price information of the target vegetable;And for being joined according to the position
Number, marks the menu name, the monovalent information in the picture on corresponding position, and by the total price information mark
Note is on the designated position of the picture;And for the picture after mark to be output to the display screen, so that institute
State the picture after display screen display mark.
Optionally, the server, for according to the picture, determining the target vegetable in the picture described
When the menu name of location parameter and the target vegetable on carrier, it is specifically used for:
The server by utilizing convolutional neural networks carry out convolutional calculation to the picture, are corresponded to obtaining the picture
Three-dimensional matrice;
The server handles the three-dimensional matrice, obtains location matrix by the first convolutional layer;
The server determines the location parameter according to the location matrix;
The server is handled the three-dimensional matrice by the second convolutional layer, obtains feature extraction matrix;
The server determines the menu name according to the feature extraction matrix.
Optionally, the server, for according to the feature extraction matrix, when determining the menu name, specifically using
In:
The server obtains feature vector according to the feature extraction matrix;
The server determines the menu name according to described eigenvector and pre-stored second correspondence,
Second correspondence is characterized the correspondence of vector sum menu name.
Optionally, the server, for according to described eigenvector and pre-stored second correspondence, determining institute
When stating menu name, it is specifically used for:
The server carries out mathematic interpolation to described eigenvector and pre-stored feature vector, with obtain difference to
Amount, the pre-stored feature vector are the corresponding feature vector of all vegetables;
The server obtains the absolute value of each element in the difference value vector according to the difference value vector;
The corresponding difference value vector of the absolute value is inputted full Connection Neural Network by the server, to obtain difference minimum
The corresponding feature vector of absolute value;
The server is corresponded to according to the corresponding feature vector of absolute value of difference minimum and pre-stored second
Relationship determines the menu name.
Optionally, the location parameter includes:
With the presence or absence of the probability of the target vegetable, the target vegetable center on the carrier abscissa
And ordinate, the length and width of the target vegetable.
To achieve the above object, another technical solution of the invention is to provide a kind of data processing method, and feature exists
In the method is applied in the receipts including at least one video capture device, terminal device, server and at least one display screen
In silver-colored system, the method includes:
The video capture device acquisition includes the video data of carrier and the target vegetable in the carrier;
And the video data is sent to the terminal device;
The terminal device obtains the corresponding picture of designated frame in the video data;And the picture is sent to
The server;
The server determines position of the target vegetable on the carrier in the picture according to the picture
Set the menu name of parameter and the target vegetable;And the location parameter and the menu name are sent to the end
End equipment;
The terminal device determines that the menu name corresponds to according to the menu name and preset first correspondence
Monovalent information, wherein first correspondence is the correspondence of menu name and monovalent information;And according to described
Monovalent information obtains the corresponding total price information of the target vegetable;It is right in the picture and according to the location parameter
Mark the menu name, the monovalent information on the position answered, and by the total price information labeling the picture finger
Positioning is set;And the picture after mark is output on the display screen, so that after display screen display mark
The picture.
Optionally, the server determines the target vegetable in the picture in the carrying according to the picture
The menu name of location parameter and the target vegetable on object, including:
The server by utilizing convolutional neural networks carry out convolutional calculation to the picture, are corresponded to obtaining the picture
Three-dimensional matrice;
The server handles the three-dimensional matrice, obtains location matrix by the first convolutional layer;
The server determines the location parameter according to the location matrix;
The server is handled the three-dimensional matrice by the second convolutional layer, obtains feature extraction matrix;
The server determines the menu name according to the feature extraction matrix.
Optionally, the server determines the menu name according to the feature extraction matrix, including:
The server obtains feature vector according to the feature extraction matrix;
The server determines the menu name according to described eigenvector and pre-stored second correspondence,
Second correspondence is characterized the correspondence of vector sum menu name.
Optionally, the server determines the dish according to described eigenvector and pre-stored second correspondence
The name of an article claims, including:
The server carries out mathematic interpolation to described eigenvector and pre-stored feature vector, with obtain difference to
Amount, the pre-stored feature vector are the corresponding feature vector of all vegetables;
The server obtains the absolute value of each element in the difference value vector according to the difference value vector;
The corresponding difference value vector of the absolute value is inputted full Connection Neural Network by the server, to obtain difference minimum
The corresponding feature vector of absolute value;
The server is corresponded to according to the corresponding feature vector of absolute value of difference minimum and pre-stored second
Relationship determines the menu name.
Optionally, the location parameter includes:
With the presence or absence of the probability of the target vegetable, the target vegetable center on the carrier abscissa
And ordinate, the length and width of the target vegetable.
The invention has the advantages that:
In embodiments of the present invention, since the video capture device in cash register system can acquire carrier and positioned at carrying
The video data of target vegetable in object, and the video is sent to terminal device, terminal device can obtain the video data
The corresponding picture of middle designated frame, then terminal device the picture is sent to server, which can be true according to the picture
The menu name of location parameter and the target vegetable of the target vegetable on carrier in the fixed picture, and server should
Location parameter and menu name are sent to terminal device, terminal device according to the menu name and preset first correspondence,
Determine the corresponding monovalent information of the menu name and the corresponding total price information of target vegetable, then terminal device grace according to
The location parameter marks menu name, monovalent information in picture on corresponding position, and by total price information labeling in picture
Designated position on, and the picture after mark is output on display screen, after using above-mentioned processing mode, due to being not necessarily to cash register
Member remembers the unit price of each vegetable, and obtains total price by mental arithmetic or calculator without cashier, to advantageously reduce knot
The duration consumed when calculation, and then be conducive to improve settlement efficiency, meanwhile, compared with the existing technology (artificial clearing), the present invention is real
It applies example and advantageously reduces the probability for mistake occur.
Description of the drawings
Fig. 1 is a kind of flow chart of data processing method provided in an embodiment of the present invention;
Fig. 2 is the flow chart of another data processing method provided in an embodiment of the present invention;
Fig. 3 is the flow chart of another data processing method provided in an embodiment of the present invention;
Fig. 4 is the flow chart of another data processing method provided in an embodiment of the present invention;
Fig. 5 is a kind of structural schematic diagram of cash register system provided in an embodiment of the present invention.
Specific implementation mode
The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention..
Embodiment 1
Fig. 1 is a kind of flow chart of data processing method provided in an embodiment of the present invention, as shown in Figure 1, this method application
In the cash register system including at least one video capture device, terminal device, server and at least one display screen, this method
Include the following steps:
101, video capture device acquisition includes the video data of carrier and the target vegetable in carrier.
Specifically, when the personnel of having dinner choose vegetable clearing, which can be put in a carrying by the vegetable chosen
On object (such as service plate), and the carrier is put on designated position, such as the carrier is positioned over video capture device energy
Enough at the position of acquisition video data, then the video capture device acquires video data.
It should be noted that the video capture device includes picture pick-up device, such as camera, about video capture device
Specific type is not specifically limited herein.
102, video data is sent to terminal device by video capture device.
103, terminal device obtains the corresponding picture of designated frame in video data.
It should be noted that specifically obtaining the corresponding picture of which frame can be set according to actual needs, herein not
It is specifically limited.
104, picture is sent to server by terminal device.
Specifically, server can carry out data interaction with multiple terminal devices, when a server correspond to it is multiple
When terminal device, the data processing pressure of server can be reduced by obtaining picture by terminal device, i.e., terminal device is for service
Can play the role of load balancing for device, to advantageously reduce the burden of server, simultaneously because terminal device be by
The corresponding picture of one designated frame is sent to server, to advantageously reduce the requirement to network speed.
105, server determines location parameter and target vegetable of the target vegetable in picture on carrier according to picture
Menu name.
It should be noted that the target vegetable is all vegetables on carrier.
106, the menu name of location parameter and target vegetable of the target vegetable on carrier.
107, terminal device determines the corresponding unit price letter of menu name according to menu name and preset first correspondence
Breath, wherein the first correspondence is the correspondence of menu name and monovalent information.
108, terminal device obtains the corresponding total price information of target vegetable according to monovalent information.
Specifically, after the monovalent information for obtaining each target vegetable, terminal device can sum to the unit price information
Operation, to obtain the corresponding total price information of each target vegetable.
109, terminal device marks menu name, monovalent information according to location parameter in picture on corresponding position, with
And by total price information labeling on the designated position of picture.
Specifically, after marking menu name and monovalent information on corresponding position in picture, it can be true according to the picture
The fixed different corresponding unit price of vegetable, and, it can be determined that mark in the picture is put with each vegetable in actual bearer object
Whether unanimously, so as to prevent mistakes and omissions it the case where, holds consultation convenient for the personnel of having dinner and cashier, to be conducive to carry
Accuracy rate when height clearing.
1010, the picture after mark is output on display screen by terminal device, so that the picture after display screen display mark.
Specifically, the picture that personnel and cashier can show according to display screen of having dinner is settled accounts, conveniently have dinner personnel
Negotiation when being settled accounts with cashier.
It should be noted that the display screen number can be two, one towards the personnel of having dinner, one towards cash register
Member, the number of certain display screen may be one, as soon as when the number of display screen is also, which can be arranged
On the position that meal personnel and cashier are able to observe that simultaneously, so that the personnel of having dinner and cashier hold consultation in clearing, close
It can be configured in the specific number of display screen, be not specifically limited herein according to actual needs.
In embodiments of the present invention, after using processing method as shown in Figure 1, remember each vegetable due to being not necessarily to cashier
Unit price, and obtain total price by mental arithmetic or calculator without cashier, consumed when to advantageously reduce clearing when
It is long, and then be conducive to improve settlement efficiency, meanwhile, compared with the existing technology (artificial clearing), the embodiment of the present invention is conducive to drop
The low probability for mistake occur.
It should be noted that the function and workflow of server and terminal device can integrate in one apparatus in fact
It is existing, it can also be implemented separately, specific set-up mode is not specifically limited herein.For example, being in single machine work in cash register system
When state, i.e., when cash register system is not connect with internet, terminal device also takes into account the function of server, i.e. terminal device also has
The function of above-mentioned server, and data being handled according to the workflow of above-mentioned server, processing mode with it is above-mentioned
The processing mode of server is identical.
Embodiment 2
Fig. 2 is the flow chart of another data processing method provided in an embodiment of the present invention, as shown in Fig. 2, in server
The target vegetable in picture is determined in the menu name of location parameter and target vegetable on carrier according to picture, it is specific to wrap
Include following steps:
201, server by utilizing convolutional neural networks carry out convolutional calculation to picture, to obtain the corresponding three-dimensional square of picture
Battle array.
202, server is handled three-dimensional matrice by the first convolutional layer, obtains location matrix.
Specifically, making target vegetable not identify simultaneously to solve the problem of mutually to block, for example, in same position
Upper there are two target vegetables, which sets each other, such as a target vegetable in two target vegetables
For noodle soup, another target vegetable is spiced egg, and the spiced egg is put in noodle soup, so that spiced egg blocks noodle soup formation,
Location matrix can be changed to the set-up mode such as 5*n, every 5 unit lengths correspond to a target vegetable, pass through above-mentioned design
The corresponding location matrix of two target vegetables can be made to distinguish, to avoid due to mutually blocking the problem.
203, server determines location parameter according to location matrix.
Optionally, location parameter includes:With the presence or absence of the probability of target vegetable, target vegetable center in carrier
Upper abscissa and ordinate, the length and width of target vegetable.
Specifically, after determining location parameter, so that it may to determine that different target vegetables is being held according to the location parameter
Position in loading in order to which position subsequently corresponding with picture is proofreaded, and determines target vegetable on picture
Different location.
204, server is handled three-dimensional matrice by the second convolutional layer, obtains feature extraction matrix.
It should be noted that the first convolutional layer and the second convolutional layer are two different convolutional layers, about the first convolutional layer
It is not specifically limited herein with the specific type of the second convolutional layer.
205, server determines menu name according to feature extraction matrix.
Specifically, different feature extraction matrixes corresponds to different menu names, i.e., different menu names has difference
Feature extraction matrix, therefore can determine menu name according to feature extraction matrix.
In embodiments of the present invention, after using processing mode as shown in Figure 2, can be conducive to improve the position determined
Set parameter and the accuracy of menu name.
Embodiment 3
On the basis of Fig. 2 corresponding data processing methods, an embodiment of the present invention provides another data processing methods
Flow chart, as shown in figure 3, when determining menu name, may comprise steps of according to feature extraction matrix in server:
301, server obtains feature vector according to feature extraction matrix.
302, server determines menu name according to feature vector and pre-stored second correspondence, and second corresponds to
Relationship is characterized the correspondence of vector sum menu name.
Specifically, since the feature vector for including in different feature extraction matrixes is different, and due to different vegetable names
Claim corresponding different feature vector, therefore can be determined to correspond to according to the feature vector of extraction after using above-mentioned processing method
Menu name, to be conducive to improve the accuracy of menu name determined.
Embodiment 4
On the basis of Fig. 3 corresponding data processing methods, an embodiment of the present invention provides another data processing methods
Flow chart, as shown in figure 4, determining menu name according to feature vector and pre-stored second correspondence in server
When, it may comprise steps of:
401, server carries out mathematic interpolation to feature vector and pre-stored feature vector, to obtain difference value vector,
Pre-stored feature vector is the corresponding feature vector of all vegetables.
402, server obtains the absolute value of each element in difference value vector according to difference value vector.
403, the corresponding difference value vector of absolute value is inputted full Connection Neural Network by server, to obtain the exhausted of difference minimum
To being worth corresponding feature vector.
404, absolute value corresponding feature vector and pre-stored second correspondence of the server according to difference minimum,
Determine menu name.
In embodiments of the present invention, in order to further increase the accuracy for the menu name determined, in order to reduce by
One menu name determines the probability at another similar menu name, needs to the feature vector of acquisition and pre-stored
Feature vector carries out mathematic interpolation, and obtains the absolute value of each element in difference value vector, then by the corresponding difference of absolute value
Vector inputs full Connection Neural Network, to obtain the corresponding feature vector of absolute value of difference minimum, due to different vegetable names
Claim corresponding different feature vector, therefore the difference of the absolute value of a feature vector and corresponding menu name is minimum,
Therefore the accuracy for the menu name determined according to the corresponding feature vector of absolute value of difference minimum is relatively high, to have
Conducive to further increasing the accuracy for the menu name determined.
Embodiment 5
Fig. 5 is a kind of structural schematic diagram of cash register system provided in an embodiment of the present invention, as shown in figure 5, cash register system packet
It includes:At least one video capture device 51, terminal device 52, server 53 and at least one display screen 54, wherein at least one
Video capture device 51 is connect with terminal device 52, and terminal device 52 is connect with server 53 and at least one display screen 54;
Video capture device 51, for acquire include carrier and the target vegetable in carrier video data;
And for video data to be sent to terminal device 52;
Terminal device 52 obtains the corresponding picture of designated frame in video data;And for picture to be sent to server
53;
Server 53, for according to picture, determining location parameter and target of the target vegetable in picture on carrier
The menu name of vegetable;And for location parameter and menu name to be sent to terminal device 52;
Terminal device 52 is additionally operable to, according to menu name and preset first correspondence, determine that menu name is corresponding
Monovalent information, wherein the first correspondence is the correspondence of menu name and monovalent information;And for being believed according to unit price
Breath obtains the corresponding total price letter of target vegetable;And for according to location parameter, dish to be marked on corresponding position in picture
The name of an article claims, monovalent information, and by total price information labeling on the designated position of picture;And for the picture after marking
It is output on display screen 54, so that the picture after the display mark of display screen 54.
In embodiments of the present invention, since the video capture device 51 in cash register system can acquire carrier and positioned at holding
The video data of target vegetable in loading, and the video is sent to terminal device 52, terminal device 52 can obtain this and regard
Frequency according in the corresponding picture of designated frame, then terminal device 52 picture is sent to server 53, which can
The menu name of target the vegetable location parameter on carrier and the target vegetable in the picture is determined according to the picture, and
And the location parameter and menu name are sent to terminal device 52 by server 53, terminal device 52 is according to the menu name and in advance
If the first correspondence, determine the corresponding monovalent information of menu name and the corresponding total price information of target vegetable, then eventually
End equipment 52 can mark menu name, monovalent information according to the location parameter in picture on corresponding position, and by total price
Information labeling is output on the designated position of picture, and by the picture after mark on display screen 54, is using above-mentioned processing side
After formula, total price is obtained by mental arithmetic or calculator due to remembering the unit price of each vegetable without cashier, and without cashier,
The duration consumed when to advantageously reduce clearing, and then be conducive to improve settlement efficiency, meanwhile, (people compared with the existing technology
Work is settled accounts), the embodiment of the present invention advantageously reduces the probability for mistake occur.
Optionally, server 53, for according to picture, determining location parameter of the target vegetable on carrier in picture
When with the menu name of target vegetable, it is specifically used for:Server 53 utilizes convolutional neural networks, and convolutional calculation is carried out to picture,
To obtain the corresponding three-dimensional matrice of picture;Server 53 is handled three-dimensional matrice by the first convolutional layer, obtains position square
Battle array;Server 53 determines location parameter according to location matrix;Server 53 is by the second convolutional layer, at three-dimensional matrice
Reason, obtains feature extraction matrix;Server 53 determines menu name according to feature extraction matrix.
Optionally, server 53, for according to feature extraction matrix, when determining menu name, being specifically used for:Server 53
According to feature extraction matrix, feature vector is obtained;Server 53 is according to feature vector and pre-stored second correspondence, really
Determine menu name, the second correspondence is characterized the correspondence of vector sum menu name.
Optionally, server 53, for according to feature vector and pre-stored second correspondence, determining menu name
When, it is specifically used for:Server 53 carries out mathematic interpolation to feature vector and pre-stored feature vector, with obtain difference to
Amount, pre-stored feature vector are the corresponding feature vector of all vegetables;Server 53 according to difference value vector, obtain difference to
The absolute value of each element in amount;The corresponding difference value vector of absolute value is inputted full Connection Neural Network by server 53, to obtain
The corresponding feature vector of absolute value of difference minimum;Server 53 is according to the corresponding feature vector of absolute value of difference minimum and pre-
The second correspondence first stored, determines menu name.
Optionally, location parameter includes:With the presence or absence of the probability of target vegetable, target vegetable center in carrier
Upper abscissa and ordinate, the length and width of target vegetable.
It should be noted that can refer to embodiment 1 to embodiment 4 about being discussed in detail for cash register system, herein no longer in detail
Carefully repeat.
Although above having used general explanation and specific embodiment, the present invention is described in detail, at this
On the basis of invention, it can be made some modifications or improvements, this will be apparent to those skilled in the art.Therefore,
These modifications or improvements without departing from theon the basis of the spirit of the present invention belong to the scope of protection of present invention.
Claims (10)
1. a kind of cash register system, which is characterized in that the cash register system includes:At least one video capture device, terminal device,
Server and at least one display screen, wherein at least one video capture device is connect with the terminal device, the end
End equipment is connect with the server and at least one display screen;
The video capture device, for acquire include carrier and the target vegetable in the carrier video counts
According to;And for the video data to be sent to the terminal device;
The terminal device obtains the corresponding picture of designated frame in the video data;And for the picture to be sent to
The server;
The server, for according to the picture, determining the target vegetable in the picture on the carrier
The menu name of location parameter and the target vegetable;And for the location parameter and the menu name to be sent to
The terminal device;
The terminal device is additionally operable to, according to the menu name and preset first correspondence, determine the menu name
Corresponding unit price information, wherein first correspondence is the correspondence of menu name and monovalent information;And it is used for
According to the monovalent information, the corresponding total price information of the target vegetable is obtained;And for according to the location parameter,
The menu name, the monovalent information are marked in the picture on corresponding position, and the total price information labeling is existed
On the designated position of the picture;And for the picture after mark to be output to the display screen, so that described aobvious
The picture after display screen display mark.
2. cash register system as described in claim 1, which is characterized in that the server, for according to the picture, determining institute
The target vegetable in picture is stated in the menu name of location parameter and the target vegetable on the carrier, specifically
For:
The server by utilizing convolutional neural networks carry out convolutional calculation, to obtain the picture corresponding three to the picture
Tie up matrix;
The server handles the three-dimensional matrice, obtains location matrix by the first convolutional layer;
The server determines the location parameter according to the location matrix;
The server is handled the three-dimensional matrice by the second convolutional layer, obtains feature extraction matrix;
The server determines the menu name according to the feature extraction matrix.
3. cash register system as claimed in claim 2, which is characterized in that the server, for according to the feature extraction square
Battle array, when determining the menu name, is specifically used for:
The server obtains feature vector according to the feature extraction matrix;
The server determines the menu name according to described eigenvector and pre-stored second correspondence, described
Second correspondence is characterized the correspondence of vector sum menu name.
4. cash register system as claimed in claim 3, which is characterized in that the server, for according to described eigenvector and
Pre-stored second correspondence, when determining the menu name, is specifically used for:
The server carries out mathematic interpolation to described eigenvector and pre-stored feature vector, to obtain difference value vector,
The pre-stored feature vector is the corresponding feature vector of all vegetables;
The server obtains the absolute value of each element in the difference value vector according to the difference value vector;
The corresponding difference value vector of the absolute value is inputted full Connection Neural Network by the server, to obtain the exhausted of difference minimum
To being worth corresponding feature vector;
Absolute value corresponding feature vector and pre-stored second correspondence of the server according to difference minimum,
Determine the menu name.
5. cash register system as described in claim 1, which is characterized in that the location parameter includes:
With the presence or absence of the probability of the target vegetable, the target vegetable center on the carrier abscissa and vertical
Coordinate, the length and width of the target vegetable.
6. a kind of data processing method, which is characterized in that the method is applied including at least one video capture device, terminal
In the cash register system of equipment, server and at least one display screen, the method includes:
The video capture device acquisition includes the video data of carrier and the target vegetable in the carrier;With
And the video data is sent to the terminal device;
The terminal device obtains the corresponding picture of designated frame in the video data;And the picture is sent to described
Server;
The server determines position ginseng of the target vegetable in the picture on the carrier according to the picture
The menu name of number and the target vegetable;And the location parameter and the menu name are sent to the terminal and set
It is standby;
The terminal device determines the corresponding list of the menu name according to the menu name and preset first correspondence
Valence information, wherein first correspondence is the correspondence of menu name and monovalent information;And according to the unit price
Information obtains the corresponding total price information of the target vegetable;It is corresponding in the picture and according to the location parameter
Mark the menu name, the monovalent information on position, and by the total price information labeling the picture specific bit
It sets;And the picture after mark is output on the display screen, so that described after display screen display mark
Picture.
7. method as claimed in claim 6, which is characterized in that the server is determined according to the picture in the picture
Location parameter and the target vegetable of the target vegetable on the carrier menu name, including:
The server by utilizing convolutional neural networks carry out convolutional calculation, to obtain the picture corresponding three to the picture
Tie up matrix;
The server handles the three-dimensional matrice, obtains location matrix by the first convolutional layer;
The server determines the location parameter according to the location matrix;
The server is handled the three-dimensional matrice by the second convolutional layer, obtains feature extraction matrix;
The server determines the menu name according to the feature extraction matrix.
8. the method for claim 7, which is characterized in that the server determines institute according to the feature extraction matrix
Menu name is stated, including:
The server obtains feature vector according to the feature extraction matrix;
The server determines the menu name according to described eigenvector and pre-stored second correspondence, described
Second correspondence is characterized the correspondence of vector sum menu name.
9. method as claimed in claim 8, which is characterized in that the server is according to described eigenvector and pre-stored
Second correspondence determines the menu name, including:
The server carries out mathematic interpolation to described eigenvector and pre-stored feature vector, to obtain difference value vector,
The pre-stored feature vector is the corresponding feature vector of all vegetables;
The server obtains the absolute value of each element in the difference value vector according to the difference value vector;
The corresponding difference value vector of the absolute value is inputted full Connection Neural Network by the server, to obtain the exhausted of difference minimum
To being worth corresponding feature vector;
Absolute value corresponding feature vector and pre-stored second correspondence of the server according to difference minimum,
Determine the menu name.
10. method as claimed in claim 6, which is characterized in that the location parameter includes:
With the presence or absence of the probability of the target vegetable, the target vegetable center on the carrier abscissa and vertical
Coordinate, the length and width of the target vegetable.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810235977.XA CN108510677A (en) | 2018-03-21 | 2018-03-21 | A kind of cash register system and data processing method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810235977.XA CN108510677A (en) | 2018-03-21 | 2018-03-21 | A kind of cash register system and data processing method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108510677A true CN108510677A (en) | 2018-09-07 |
Family
ID=63377921
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810235977.XA Withdrawn CN108510677A (en) | 2018-03-21 | 2018-03-21 | A kind of cash register system and data processing method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108510677A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111597862A (en) * | 2019-02-21 | 2020-08-28 | 杭州海康威视数字技术股份有限公司 | Dish category identification method and device and electronic equipment |
US12029901B2 (en) | 2017-06-30 | 2024-07-09 | Lungpacer Medical Inc. | Devices and methods for prevention, moderation, and/or treatment of cognitive injury |
-
2018
- 2018-03-21 CN CN201810235977.XA patent/CN108510677A/en not_active Withdrawn
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US12029901B2 (en) | 2017-06-30 | 2024-07-09 | Lungpacer Medical Inc. | Devices and methods for prevention, moderation, and/or treatment of cognitive injury |
CN111597862A (en) * | 2019-02-21 | 2020-08-28 | 杭州海康威视数字技术股份有限公司 | Dish category identification method and device and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109508681B (en) | Method and device for generating human body key point detection model | |
CN107194398B (en) | Vehicle damages recognition methods and the system at position | |
CN107992871A (en) | The automatic accounting method in dining room based on image recognition | |
CN107784332A (en) | Electronics price tag information processing system and method | |
CN104239062B (en) | Client-based polytype Shipping Options Page display methods and its system | |
CN108510677A (en) | A kind of cash register system and data processing method | |
CN103389978B (en) | A kind of method and system being obtained information by augmented reality | |
CN103561086A (en) | Method for presenting schedule prompting message, terminal device and cloud server | |
CN108986731A (en) | A kind of display panel and its compensation method, display device | |
CN107480737A (en) | A kind of production management system and method for detection PCBA bar codes | |
CN110059715A (en) | Floristic recognition methods and device, storage medium, computer equipment | |
CN109977762A (en) | A kind of text positioning method and device, text recognition method and device | |
CN109255562A (en) | Object processing method and device | |
CN116681364A (en) | Multi-role online instant messaging method, device, equipment and storage medium | |
CN207993178U (en) | A kind of cash register system | |
CN106533528B (en) | The two-way time compares compression and the calculation method of modem transmitted data amount | |
CN104573518B (en) | File scanning method, device, server and system | |
CN109426935A (en) | Communication information distribution method and device, communication method and device, equipment and medium | |
CN116883717A (en) | Platen checking method and device, computer readable storage medium and computer equipment | |
CN105117891A (en) | Intelligent signing method and device | |
CN103839064A (en) | Method and system for extracting electronic stamp account information | |
CN111797922B (en) | Text image classification method and device | |
CN111429641B (en) | Order calling method, device and system | |
CN107895246A (en) | A kind of materials storage management method and system | |
CN104898958B (en) | A kind of data transfering method, device and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180907 |