CN107944860A - A kind of bakery identification and cash register system and method based on neutral net - Google Patents
A kind of bakery identification and cash register system and method based on neutral net Download PDFInfo
- Publication number
- CN107944860A CN107944860A CN201711132359.4A CN201711132359A CN107944860A CN 107944860 A CN107944860 A CN 107944860A CN 201711132359 A CN201711132359 A CN 201711132359A CN 107944860 A CN107944860 A CN 107944860A
- Authority
- CN
- China
- Prior art keywords
- cash register
- bakery
- recognition module
- identification
- picture
- 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
- 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
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/08—Payment architectures
- G06Q20/20—Point-of-sale [POS] network systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07G—REGISTERING THE RECEIPT OF CASH, VALUABLES, OR TOKENS
- G07G1/00—Cash registers
- G07G1/12—Cash registers electronically operated
Abstract
The present invention relates to a kind of bakery identification based on neutral net and cash register system and method, the system includes placing the cash register area of cash register pallet, the CCD camera being fixed on directly over cash register area, the picture recognition module being connected with CCD camera and the cash register module being connected with picture recognition module, and picture recognition module is equipped with the neural network image detection unit for being used for identifying image；The method includes：1) image taking：Taken pictures using CCD camera to the bakery in cash register pallet, the picture of shooting is passed to picture recognition module；2) image recognition：Picture recognition module is detected image identification, the bakery classification and the quantity of every kind of classification included in statistical picture, and by statistical result real-time Transmission to cash register module；3) bill generates：Cash register module is according to default bakery calculation of price bill, and after billing settlement, billing data charges to cash register database.Compared with prior art, the present invention has the advantages that quick and convenient, cost is low etc..
Description
Technical field
The present invention relates to dining room bakery cash register field, knows more particularly, to a kind of bakery based on neutral net
Not with cash register system and method.
Background technology
At present in similar bakery, western-style cake shop etc. sell the StoreFront of baked food, check over goods with cash register by manually checking,
It is manually entered product quantity, classification and price.The characteristics of bakeing based food is same day production, no packaging, therefore inconvenience use
Bar code and FRID patch electronic tags.Use the packed bread product of bar code, it is also desirable to which staff is scanned successively manually
Bar code.During this causes cash register and checks over goods, manual labor is big, and time-consuming, and bakes and banks up with earth businessman for big flow client
Say, human cost is high, is taken when checking over goods cash register by hand also longer, and inconvenience is brought for customer.
Though convergence is ripe for the system of checking over goods based on FRID, the additional electron label on commodity is needed.And businessman is daily
Need the electronic tag amount that consumes big, cause high cost；It is inconvenient for use and electronic tag has certain thickness.
The content of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide a kind of convenient and efficient, accurate
Rate high bakery identification and cash register system and method based on neutral net.
The purpose of the present invention can be achieved through the following technical solutions：
A kind of bakery identification and cash register system based on neutral net, the system include the cash register for holding bakery
Pallet, the cash register area for placing cash register pallet, the CCD camera being fixed on directly over cash register area, the image recognition being connected with CCD camera
Module and the cash register module being connected with picture recognition module, the picture recognition module are equipped with the nerve for being used for identifying image
Network image detection unit, the neural network image detection unit are connected with CCD camera.
Preferably, the resolution ratio of the CCD camera is 1920*1080.
Preferably, the picture recognition module further includes the image data base for storing bakery characteristic information,
The image data base is connected with neural network image detection unit.
Preferably, the cash register module includes the cash register database being connected with picture recognition module and cash register database
The calculation of price unit of connection, and the POS machine being connected with calculation of price unit.
Preferably, the picture recognition module is PC, and the PC is connected with CCD camera by USB connecting lines.
A kind of identification of bakery and cash method, using the bakery identification based on neutral net and cash register system,
This method comprises the following steps：
1) image taking：Taken pictures using the CCD camera being fixed on directly over cash register area to the bakery in cash register pallet,
The Three Channel Color picture of shooting is passed to picture recognition module；
2) image recognition：Picture recognition module is detected identification, the baking included in statistical picture to incoming image
The quantity of food classification and each classification, and statistical result is transmitted to cash register module in real time；
3) bill generates：Cash register module is according to default bakery calculation of price bill, after billing settlement, billing data
Charge to cash register database.
Preferably, the particular content of the step 2) is：
21) the neutral net frame in neural network image detection unit is inputted using Three Channel Color picture as input data
Frame, into next step；
22) neural network framework uses depth convolution net, and convolution feature extraction is carried out to the Three Channel Color image of input,
Obtain convolution characteristic pattern；
23) suggest that net obtains region of interest, the corresponding feature vector of extraction region of interest on convolution characteristic pattern using region；
24) feature vector is handled using two full articulamentums, obtains the feature vector of fixed dimension, and it is defeated respectively
Enter to grader and return device；
25) grader processing feature vector, obtains the targeted species included in region of interest；Device processing feature vector is returned,
Obtain the position of the target included in region of interest in the picture.
Compared with prior art, the present invention has the following advantages：
First, it is quick and convenient：The present invention takes pictures the bakery in cash register pallet using high-resolution CCD camera, and
The image of shooting is handled by the picture recognition module with neural network image detection unit, can be disposable in real time
Detection identifies included bakery type and quantity, by the default bakery price of cash register module, realizes real-time
Express checkout；
2nd, discrimination is high：Under the conditions of ensureing that imaging clearly is complete, the present invention is known using the image based on neutral net
Other method, the identification to hundred class bakeries, accuracy rate is more than 98%；
3rd, cost is low：The present invention uses high definition CCD camera of the resolution ratio for 1920*1080, which only needs one
Independent display cards of the CPU or video memory of intel i5 ranks more than 2G can reach real-time detection, compared to traditional complicated receipts
Silver-colored system, greatly reduces cost.
Brief description of the drawings
Fig. 1 is a kind of bakery identification based on neutral net and the structure diagram of cash register system；
Fig. 2 is a kind of identification of bakery and the flow chart of cash method；
Fig. 3 is a kind of bakery identification and the image detection identification process figure of cash method in the embodiment of the present invention；
Fig. 4 is a kind of bakery identification and the neural network image detection framework stream of cash method in the embodiment of the present invention
Cheng Tu.
In Fig. 1 shown in label：
1st, CCD camera, 2, picture recognition module, 3, cash register module, 4, cash register pallet, 5, bakery, 6, cash register area.
Embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.
Embodiment
The present invention relates to a kind of bakery identification based on neutral net and cash register system and method, wherein：
A kind of bakery identification based on neutral net includes cash register area 6, CCD camera 1, image recognition with cash register system
Module 2 and cash register module 3, as shown in Figure 1.CCD camera 1 is fixed on directly over cash register area 6, picture recognition module 2 and CCD camera
1 connection, cash register module 3 are connected with picture recognition module 2.Bakery 5 is placed on cash register pallet 4, and cash register pallet 4 is placed on
In cash register area 6.
The resolution ratio of CCD camera 1 is 1920*1080, which only needs the CPU or aobvious of an intel i5 rank
Deposit the independent display card more than 2G and can reach real-time detection.CCD camera 1 is fixed on directly over cash register area 6, for obtaining cash register area
It is interior completely to face image.
Picture recognition module 2 is equipped with neural network image detection unit and image data base, and neural network image detection is single
One end of member is connected with CCD camera, and the other end is connected with image data base.Image data base stores bakery feature letter
Breath.Picture recognition module 2 is PC, it is connected with CCD camera 1 by USB connecting lines.
Cash register module 3 includes cash register database, the calculation of price unit being attached thereto, the Gu being connected with calculation of price unit
Display screen display screen, the printer being connected with calculation of price unit and POS machine.It is corresponding that calculation of price unit is equipped with all bakeries
Pricing information；Cash register database is responsible for storing the bakery classification after picture recognition module identifies and quantity information；Price meter
Unit is calculated to be used to calculate bakery total value.Calculation of price total value is shown by client display screen, and prints account by printer
Single, POS machine is used for customer payment bill.Client display screen can use lcd display screens or led display screens.
A kind of identification of bakery and cash method, using the bakery identification based on neutral net and cash register system,
The flow of this method as shown in Fig. 2, including：
1) utilization is fixed on the CCD camera directly over cash register area and takes pictures to the bakery in cash register pallet, by shooting
Three Channel Color picture is passed to picture recognition module.
2) picture recognition module is detected identification to incoming image, the bakery classification that is included in statistical picture and
The quantity of each classification, and statistical result is transmitted to cash register module in real time.
3) cash register module is according to default bakery calculation of price bill, and after billing settlement, billing data charges to cash register
Database.
In step 2), picture recognition module carries out image recognition by neural network image detection unit, as shown in figure 3,
The step specifically includes：
21) the neutral net frame in neural network image detection unit is inputted using Three Channel Color picture as input data
Frame, into next step；
22) neural network framework is using 13 convolutional layers and the depth convolution net of 5 down-sampling layers, to the triple channel of input
Coloured image carries out convolution feature extraction, obtains convolution characteristic pattern；
23) suggest that net obtains region of interest, the corresponding feature vector of extraction region of interest on convolution characteristic pattern using region；
24) it is vectorial using two full articulamentum processing features, the feature vector of fixed dimension is obtained, and be separately input into point
Class device and recurrence device；
25) grader processing feature vector, obtains the targeted species included in region of interest；Device processing feature vector is returned,
Obtain the position of the target included in region of interest in the picture.
After step 25), species and the destination number of each species appearance that cash register module statistics obtains.
In step 21), CCD camera inputs the RGB color image of 1920 × 1080 resolution ratio, and scales the images to
1056 × 640 resolution ratio.
In step 22), using with 13 convolutional layers, the depth convolution net pair based on VGG16 structures of 5 down-sampling layers
Zoomed image is handled.Wherein, the size of each convolutional layer median filter is 3 × 3, step-length 1, is made after each convolution
With ReLU function pair convolution characteristic patterns into line activating.The wave filter size of each of which down-sampling layer is 2 × 2, and step-length is
2, and using maximum pond method, 1 times of down-sampling is carried out to the convolution characteristic pattern result of input so that characteristic pattern length and width ruler
Very little diminution half.Depth convolution net finally obtains the convolution characteristic pattern of 33 × 20 sizes.
In step 23), region suggests that the convolution characteristic pattern of 33 × 20 sizes is carried out region of interest extraction by net, and obtaining may
After the corresponding feature vector of region of interest comprising bakery target, input to two full articulamentums.
In step 24), after two full articulamentum processing, the feature vector of 4096 dimensions of region of interest is obtained, respectively
Input grader and return device.
In step 25), grader is classified using 4096 dimensional feature vector of Softmax function pairs, takes probable value most
High classification output, is neural network prediction bakery classification；Return device and mesh is solved using Smooth L1 loss functions
Cursor position corrects 4 parameters (2 translations, 2 scalings), can obtain neural network prediction bakery picture position.
By repeatedly attempt, using the method for the present invention to the recognition accuracy of hundred class bakeries more than 98%, reach
Preferable effect.
The above description is merely a specific embodiment, but protection scope of the present invention is not limited thereto, any
Be familiar with the staff of the art the invention discloses technical scope in, various equivalent modifications can be readily occurred in or replaced
Change, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection domain subject to.
Claims (7)
1. a kind of bakery identification and cash register system based on neutral net, it is characterised in that the system includes holding baking
The cash register pallet of food, the cash register area for placing cash register pallet, the CCD camera being fixed on directly over cash register area, be connected with CCD camera
Picture recognition module and the cash register module that is connected with picture recognition module, the picture recognition module, which is equipped with, to be used to identify
The neural network image detection unit of image, the neural network image detection unit are connected with CCD camera.
2. a kind of bakery identification and cash register system based on neutral net according to claim 1, it is characterised in that
The resolution ratio of the CCD camera is 1920*1080.
3. a kind of bakery identification and cash register system based on neutral net according to claim 1, it is characterised in that
The picture recognition module further includes the image data base for storing bakery characteristic information, the image data base and god
Connected through network image detection unit.
4. a kind of bakery identification and cash register system based on neutral net according to claim 1, it is characterised in that
The cash register module includes the cash register database being connected with picture recognition module, the calculation of price list being connected with cash register database
Member, and the POS machine being connected with calculation of price unit.
5. a kind of bakery identification and cash register system based on neutral net according to claim 1, it is characterised in that
The picture recognition module is PC, and the PC is connected with CCD camera by USB connecting lines.
6. a kind of apply such as bakery identification of the claim 1-5 any one of them based on neutral net and cash register system
Bakery identifies and cash method, it is characterised in that this method comprises the following steps：
1) image taking：Taken pictures, will be clapped to the bakery in cash register pallet using the CCD camera being fixed on directly over cash register area
The Three Channel Color picture taken the photograph is passed to picture recognition module；
2) image recognition：Picture recognition module is detected identification, the bakery included in statistical picture to incoming image
The quantity of classification and each classification, and statistical result is transmitted to cash register module in real time；
3) bill generates：Cash register module is according to default bakery calculation of price bill, and after billing settlement, billing data is charged to
Cash register database.
7. a kind of bakery identification according to claim 6 and cash method, it is characterised in that the step 2)
Particular content is：
21) neural network framework in neural network image detection unit is inputted using Three Channel Color picture as input data,
Into in next step；
22) neural network framework uses depth convolution net, carries out convolution feature extraction to the Three Channel Color image of input, obtains
Convolution characteristic pattern；
23) suggest that net obtains region of interest, the corresponding feature vector of extraction region of interest on convolution characteristic pattern using region；
24) feature vector is handled using two full articulamentums, obtains the feature vector of fixed dimension, and be separately input into
Grader and recurrence device；
25) grader processing feature vector, obtains the targeted species included in region of interest；Device processing feature vector is returned, is obtained
The position of the target included in region of interest in the picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711132359.4A CN107944860A (en) | 2017-11-15 | 2017-11-15 | A kind of bakery identification and cash register system and method based on neutral net |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711132359.4A CN107944860A (en) | 2017-11-15 | 2017-11-15 | A kind of bakery identification and cash register system and method based on neutral net |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107944860A true CN107944860A (en) | 2018-04-20 |
Family
ID=61932358
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711132359.4A Pending CN107944860A (en) | 2017-11-15 | 2017-11-15 | A kind of bakery identification and cash register system and method based on neutral net |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107944860A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886112A (en) * | 2019-01-16 | 2019-06-14 | 创新奇智(合肥)科技有限公司 | A kind of object identification method based on image, commodity self-service cash register system and electronic equipment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102760335A (en) * | 2012-07-18 | 2012-10-31 | 朱磊 | Intelligent wireless mobile cash register system and implementation method thereof |
| CN104077842A (en) * | 2014-07-02 | 2014-10-01 | 浙江大学 | Freestyle restaurant self-service payment device based on image identification and application method of device |
| CN106096932A (en) * | 2016-06-06 | 2016-11-09 | 杭州汇萃智能科技有限公司 | The pricing method of vegetable automatic recognition system based on tableware shape |
| CN106372571A (en) * | 2016-08-18 | 2017-02-01 | 宁波傲视智绘光电科技有限公司 | Road traffic sign detection and identification method |
| CN106504233A (en) * | 2016-10-18 | 2017-03-15 | 国网山东省电力公司电力科学研究院 | Image electric power widget recognition methodss and system are patrolled and examined based on the unmanned plane of Faster R CNN |
| EP3147799A1 (en) * | 2015-09-22 | 2017-03-29 | Xerox Corporation | Similarity-based detection of prominent objects using deep cnn pooling layers as features |
| CN106599827A (en) * | 2016-12-09 | 2017-04-26 | 浙江工商大学 | Small target rapid detection method based on deep convolution neural network |
| CN107122730A (en) * | 2017-04-24 | 2017-09-01 | 乐金伟 | Free dining room automatic price method |
| CN107273836A (en) * | 2017-06-07 | 2017-10-20 | 深圳市深网视界科技有限公司 | A kind of pedestrian detection recognition methods, device, model and medium |
-
2017
- 2017-11-15 CN CN201711132359.4A patent/CN107944860A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102760335A (en) * | 2012-07-18 | 2012-10-31 | 朱磊 | Intelligent wireless mobile cash register system and implementation method thereof |
| CN104077842A (en) * | 2014-07-02 | 2014-10-01 | 浙江大学 | Freestyle restaurant self-service payment device based on image identification and application method of device |
| EP3147799A1 (en) * | 2015-09-22 | 2017-03-29 | Xerox Corporation | Similarity-based detection of prominent objects using deep cnn pooling layers as features |
| CN106096932A (en) * | 2016-06-06 | 2016-11-09 | 杭州汇萃智能科技有限公司 | The pricing method of vegetable automatic recognition system based on tableware shape |
| CN106372571A (en) * | 2016-08-18 | 2017-02-01 | 宁波傲视智绘光电科技有限公司 | Road traffic sign detection and identification method |
| CN106504233A (en) * | 2016-10-18 | 2017-03-15 | 国网山东省电力公司电力科学研究院 | Image electric power widget recognition methodss and system are patrolled and examined based on the unmanned plane of Faster R CNN |
| CN106599827A (en) * | 2016-12-09 | 2017-04-26 | 浙江工商大学 | Small target rapid detection method based on deep convolution neural network |
| CN107122730A (en) * | 2017-04-24 | 2017-09-01 | 乐金伟 | Free dining room automatic price method |
| CN107273836A (en) * | 2017-06-07 | 2017-10-20 | 深圳市深网视界科技有限公司 | A kind of pedestrian detection recognition methods, device, model and medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886112A (en) * | 2019-01-16 | 2019-06-14 | 创新奇智(合肥)科技有限公司 | A kind of object identification method based on image, commodity self-service cash register system and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107844790A (en) | A kind of vegetable identification and POS and method based on image recognition | |
| US10540551B2 (en) | Generation of two-dimensional and three-dimensional images of items for visual recognition in checkout apparatus | |
| CN104954736B (en) | Retention analytical equipment, Retention analysis system and analysis method | |
| WO2022111059A1 (en) | Intelligent weighing apparatus based on computer vision technology, and intelligent weighing method | |
| CN108320404A (en) | Commodity recognition method, device, self-service cashier based on neural network | |
| WO2019019291A1 (en) | Settlement method and device of image recognition technology based on convolutional neural network | |
| CN107767590A (en) | Automatic identification commercialization bar code electronic scale and Automatic identification method | |
| CN108921048A (en) | A kind of shopping settlement method, device and user terminal | |
| US10503961B2 (en) | Object recognition for bottom of basket detection using neural network | |
| JP6801676B2 (en) | Information processing equipment, information processing methods, and programs | |
| KR101960899B1 (en) | Method for recognizing products | |
| CN108596137A (en) | A kind of commodity scanning record method based on image recognition algorithm | |
| CN104992507B (en) | Intelligent apparatus and its method for integrating cash register data | |
| CN103617420A (en) | Commodity fast recognition method and system based on image feature matching | |
| JP2015125671A (en) | Action map analysis apparatus, action map analysis system, and action map analysis method | |
| CN109919211A (en) | Commodity recognition method, device, system and computer-readable medium | |
| CN111080493B (en) | Dish information identification method and device and dish self-service settlement system | |
| CN109863530B (en) | Method, system and computer-readable storage medium for determining grain weight of ears | |
| CN108663105A (en) | A kind of weighing device and corresponding weighing method | |
| CN106203225B (en) | Pictorial element based on depth is deleted | |
| CN110119915A (en) | Object enters library processing method, device and system | |
| CN107944860A (en) | A kind of bakery identification and cash register system and method based on neutral net | |
| CN206431680U (en) | A kind of rapid safety Claims Resolution system for identity code barcode scanning of being traced to the source based on product false proof | |
| CN115900910A (en) | 3D vision intelligence balance of tracing to source based on data warehouse | |
| CN109300091A (en) | Radiance bearing calibration and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180420 |