CN109034837A - Multi-code is traced to the source anti-fake method and system - Google Patents
Multi-code is traced to the source anti-fake method and system Download PDFInfo
- Publication number
- CN109034837A CN109034837A CN201810698368.8A CN201810698368A CN109034837A CN 109034837 A CN109034837 A CN 109034837A CN 201810698368 A CN201810698368 A CN 201810698368A CN 109034837 A CN109034837 A CN 109034837A
- Authority
- CN
- China
- Prior art keywords
- product
- code
- checked
- information
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
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
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
- G06Q30/0185—Product, service or business identity fraud
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1408—Methods for optical code recognition the method being specifically adapted for the type of code
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Electromagnetism (AREA)
- Toxicology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Trace to the source this application discloses a kind of multi-code anti-fake method and system, this method comprises the following steps: based on the pattern identification code in scanning product packaging, identification obtains first identifier information；In the case where identification obtains first identifier information, query interface is shown；Based on query interface, the character mark code on the product surface to be checked of user's input is obtained；According to the character mark code on product surface to be checked, the product traceability information of product to be checked is exported；Wherein, product traceability information includes: production information of the product to be checked in entire production process, preconfigured pattern identification code corresponding with character mark code, wherein, preconfigured pattern identification code corresponding with character mark code is used to verify the true and false of the pattern identification code in product packaging.The application, in conjunction with come the entire production process that traces product, is realized the purpose of product traceability, reduces the probability that product is forged by the identification code on the identification code and product surface in product packaging.
Description
Technical field
It traces to the source anti-fake method and system this application involves technical field of food safety more particularly to a kind of multi-code.
Background technique
Food safety affects everyone daily life and health.As the living standard of the people in recent years increasingly mentions
Height, then have enough nor being satisfied with, but to be eaten safer healthier.Recently the meat eggs quality of the aquaculture exposed is asked
Topic obtained society extensive concern, meat egg cultivation and slaughtering process opacification result in the public to meat egg twist and accidentally
Meeting, food-safety problem is unquestionable to become focus concerned by people.The product for how allowing consumer to buy oneself
Tracking is traced to the source, it is ensured that product quality, producer take various positive anti-counterfeit measures of tracing to the source mostly.
It is anti-fake in order to achieve the purpose that trace to the source, anti-false trademark, Antiforge bar code, laser anti-counterfeit are mainly used in the prior art
Label etc. helps consumer to tell truth from falsehood, but as illegal businessman fakes horizontal raising, existing anti-counterfeit measures exist
Biggish security risk.
It should be noted that above content belongs to the technology category of cognition of inventor, the prior art is not necessarily constituted.
Summary of the invention
To solve the above-mentioned problems, it traces to the source present applicant proposes a kind of multi-code anti-fake method, this method includes following step
Rapid: based on the pattern identification code in scanning product packaging, identification obtains first identifier information；First identifier information is obtained in identification
In the case where, show query interface；Wherein, query interface is packed on any one interior product surface for user's input product
Character mark code, to inquire corresponding product traceability information；Based on query interface, the product surface to be checked of user's input is obtained
On character mark code；According to the character mark code on product surface to be checked, the product traceability information of product to be checked is exported；
Wherein, product traceability information include: production information of the product to be checked in entire production process, it is preconfigured with character mark
Know the corresponding pattern identification code of code, wherein preconfigured pattern identification code corresponding with character mark code is for verifying product
The true and false of pattern identification code in packaging.
In one example, according to the character mark code on product surface to be checked, the product of product to be checked is exported
After information of tracing to the source, method further include: based on preconfigured pattern identification code corresponding with character mark code is scanned, identify
To third identification information；By comparing third identification information and first identifier information, to verify the pattern identification in product packaging
The true and false of code.
In one example, the form of pattern identification code be it is following any one: bar code, two dimensional code.
In one example, it is based on query interface, obtains the character mark code on the product surface to be checked of user's input,
It include: to detect whether to receive shooting instruction based on query interface, wherein shooting instruction is that starting shoots product table to be checked
The instruction of the image in face；In the case where receiving shooting instruction, the image of product surface to be checked is acquired, wherein to be checked
It include character mark code in the image of product surface；Based on convolutional neural networks identification model trained in advance, extract to be checked
The characteristic for the character mark code for including in the image of product surface；Characteristic is input to recurrent neural network classifier
In, and it is based on recurrent neural network classifier, according to characteristic, the output data of last moment recurrent neural network classifier
And the vector data that the character mark code that identifies of last moment recurrent neural network classifier is converted to, it is sequentially output character
The recognition result of identification code.
In one example, the forward algorithm that recurrent neural network classifier uses are as follows:
Wherein, b0=0；
Wherein, D is the dimension of input vector, and H is the number of the neuron of hidden layer, and K is of the neuron of output layer
Number, x are the characteristic that convolutional neural networks extract,For hidden layer neuron in current time recurrent neural network
Input,For the output of hidden layer neuron in current time recurrent neural network, θ () isIt arrivesFunction；wih、wh'hPoint
It is notCorresponding weight coefficient,For the input of current time recurrent neural network output layer neuron；wh'hIt is defeated
The corresponding weight of each neuron of layer out,For the output of current time recurrent neural network output layer neuron,It is general for one
Rate value, characterization current time correspond to the opposite ratio with the adduction of all neuritis output valves of output layer of neuron output value.
In one example, the image of collected product surface to be checked is pre-processed, wherein pretreatment includes
It is at least one following: gray processing processing, denoising, correction process.
In one example, it is carried out at gray processing by image of the following algorithm to collected product surface to be checked
Reason: I=0.3B+0.59G+0.11R；Wherein, I is the gray value of each pixel, and B is each pixel in original image in channel B
Component, G are component of each pixel in the channel G, component of the R for each pixel in original image in the channel R in original image.
In one example, the image of collected noise-containing product surface to be checked is passed through by following algorithm
Field average treatment is crossed, the image after being denoised:
Wherein, P for each neighborhood pixels in taken field coordinate, Q be field in include neighborhood pixels number, f (x,
It y) is collected noise-containing image, g (x, y) is the image after denoising.
In one example, by following algorithm by the image of collected product surface to be checked along each predetermined inclination
Angle carries out Radon transformation, calculates the sum of the gradient absolute value of the corresponding projecting integral's figure of each predetermined inclination angle, and will be terraced
The maximum tilt angle of accumulated value of degree absolute value is determined as the tilt angle of original image, according to determining tilt angle to original image
It is corrected, the image after being corrected:
Rφ(x')=∫ f (x'cos φ-y'sin φ, x'sin φ-y'cos φ) dy'；
Wherein, φ indicates predetermined inclination angle, Rφ() indicates to carry out Radon transformation along the direction φ, and f (x, y) is to collect
The inclined image of character mark code.
In one example, based on convolutional neural networks identification model trained in advance, product surface to be checked is extracted
Image in include character mark code characteristic before, method further include: product table to be checked is extracted using filter
The feature in eight directions of image in face；Using the image of product surface to be checked and the image extracted by filter characteristic as volume
Product neural network recognization model input, wherein convolutional neural networks identification model be include two layers of convolutional layer and one layer of multireel
The neural network of lamination；The highest convolutional neural networks model of recognition correct rate will be tested to be determined as extracting product surface to be checked
Image in comprising character recognition code characteristic convolutional neural networks identification model.
In one example, filter is Gabor filter, and Gabor filter extracts the formula of feature are as follows:
Wherein, (x, y) identifies location of pixels, and M is direction number,Indicate direction, σ identifier space scale factor.
In one example, product traceability information includes at least one following: cultivation factory's information, pouity dwelling place information, poultry letter
Breath, poultry growth information, poultry materials information, poultry vaccine information, breeding environment information.
In one example, it is being based on pattern identification code, to set corresponding word with each birds, beasts and eggs of a collection of poultry production
After according with identification code, method further include: building block chain network, wherein block node in block chain network include such as down toward
Few one kind；Poultry seedling supplier, supply of forage quotient, vaccine supplier, poultry farming factory, birds, beasts and eggs retailer, consumer；It is based on
The product traceability information of each birds, beasts and eggs is stored into block chain network each block node by the character mark code of each birds, beasts and eggs
On block chain.
On the other hand, the application also proposed a kind of multi-code and trace to the source anti-fake system, comprising: consumer end, for sweeping
The pattern identification code in product packaging is retouched, identification obtains first identifier information；Cloud Server is communicated with consumer end, is used for
In the case where consumer end identifies to obtain first identifier information, query interface is shown, and user is obtained based on query interface
Character mark code on the product surface to be checked of input exports to be checked according to the character mark code on product surface to be checked
Ask the product traceability information of product；Wherein, product traceability information includes: production letter of the product to be checked in entire production process
Breath, preconfigured pattern identification code corresponding with character mark code, wherein preconfigured figure corresponding with character mark code
Shape identification code is used to verify the true and false of the pattern identification code in product packaging.
In one example, system further include: the first stamp equipment, for generating the pattern identification code of product packaging, and
Pattern identification code is printed upon in corresponding product packaging；Second stamp equipment, for generating the word of each product in product packaging
Identification code is accorded with, and the identification code of each product is printed upon on corresponding product surface.
In one example, system further include: acquisition equipment is communicated with Cloud Server, to acquire the product of each product
It traces to the source information；Wherein, character mark code of the Cloud Server based on each product, by the product traceability information of each product store to
In block chain network.
In one example, product is birds, beasts and eggs, wherein system further include: raiser's terminal is communicated with Cloud Server, is used
In the cultivation information for remotely monitoring at least one cultivation factory.
The multi-code anti-fake mode of tracing to the source proposed by the application can be brought the following benefits:
1. being arranged on each product surface in the pattern identification code and the product packaging by being arranged in product packaging
Character mark code traces the product traceability information of each product, has ensured the safety of product；
2. being tested according to the pattern identification code that character mark code returns the pattern identification in product packaging by comparing
Card, reduces the probability of fake product；
3. storing the information of tracing to the source of each product using block chain, it can be ensured that the uniqueness of data, it can not tamper.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen
Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings:
Fig. 1 is that a kind of multi-code provided by the embodiments of the present application is traced to the source anti-fake system structure diagram；
Fig. 2 is a kind of optional cultivation factory's schematic diagram provided by the embodiments of the present application；
Fig. 3 is that a kind of optional birds, beasts and eggs provided by the embodiments of the present application pack schematic diagram；
Fig. 4 is a kind of query interface schematic diagram provided by the embodiments of the present application；
Fig. 5 is that a kind of multi-code provided by the embodiments of the present application is traced to the source the flow diagram of anti-fake method.
Specific embodiment
For the clearer general idea for illustrating the application, carry out in an illustrative manner with reference to the accompanying drawings of the specification detailed
It describes in detail bright.
It should be noted that multi-code provided by the present application is traced to the source, anti-fake scheme can be applied but be not limited to various meats, egg
Class animal products, by the identification code on the identification code and product surface in product packaging in conjunction with come the entire production that traces product
Process is realized since the internal relation between the identification code on the identification code and product surface in product packaging only has product
Producer is known, the probability that product is forged is thus greatly reduced.
In addition, it should also be noted that, the said goods can be any one poultry (including but not limited to geese and ducks, chicken,
Quail etc.) birds, beasts and eggs (including but not limited to goose egg, duck's egg, egg, quail egg etc.), as an alternative embodiment, under
The application each embodiment in face is illustrated by taking egg as an example.
Embodiments herein discloses a kind of multi-code traceability anti-fake system, as shown in Figure 1, the system includes: cloud service
(Fig. 1 shows n, is icon 1-1, icon 1- respectively for device 2, raiser's terminal 3, consumer end 4 and at least one cultivation factory
2 ... icon 1-n), wherein there are multiple henhouses (to illustrate only four in Fig. 1, that is, cultivate the henhouse of factory 1-1 in each cultivation factory
1-1-1, henhouse 1-1-2, henhouse 1-1-3, henhouse 1-1-4；Cultivate the henhouse 1-2-1 of factory 1-2, henhouse 1-2-2, henhouse 1-2-3,
Henhouse 1-2-4；Cultivate henhouse 1-n-1, henhouse 1-n-2, the henhouse 1-n-3, henhouse 1-n-4 of factory 1-n).By in each cultivation
Factory department affixes one's name to the internet-of-things terminal communicated with Cloud Server 2 and (i.e. the internet-of-things terminal 5-1 of cultivation factory 1-1, cultivates factory 1-2's
Internet-of-things terminal 5-2, the internet-of-things terminal 5-n for cultivating factory 1-n), to be acquired in each cultivation factory by various sensors
To cultivation information be sent to Cloud Server 2.The cultivation information that each cultivation factory is managed by Cloud Server, may be implemented cloud
Cultivation.
Optionally, each cultivation factory communicated with Cloud Server can be acquired in each cultivation factory by various acquisition equipment
Cultivation information.Wherein, acquisition equipment can be various sensors or input equipment, to acquire the product traceability letter of each product
Breath；Wherein, character mark code of the Cloud Server based on each egg surface stores the product traceability information of each egg to area
In block chain network.It specifically, can be by each baby chick supplier, supply of forage quotient, vaccine supplier, breeding layer chicken factory, egg
The node of retailer and consumer as block chain network constructs block chain network, unique character mark based on each egg
Code, by the product traceability of each egg (including from baby chick, feed, the vaccine that laying hen is injected, the cultivation of laying hen, the sale of egg
To the entire supply chain of ultimate consumer) information is stored in each node in block network, it can be ensured that and data can not be usurped
It is modified.
For cultivating factory 1-1, Fig. 2 is a kind of optional cultivation factory's schematic diagram provided by the embodiments of the present application, such as Fig. 2 institute
Show, icon 7-1-1, icon 7-1-2, icon 7-1-3, icon 7-1-4, icon 7-1-5, icon 7-1-6, icon 7-1-7 difference
For temperature sensor, humidity sensor, air velocity transducer, carbon dioxide sensor, optical sensor, ammonia gas sensor, PM2.5
Sensor.Temperature, humidity, wind speed, carbon dioxide, illumination, ammonia and the PM2.5 of each cultivation factory are acquired by these sensors
Equal breeding environments information, and collected information is sent to Cloud Server 2 by internet-of-things terminal 5-1.
In order to realize intelligent cultivation, each cultivation factory can dispose intelligence cultivation robot, for example, the intelligence of cultivation factory 1-1
Robot 6-1 can be cultivated.Intelligence cultivation robot is communicated with the various sensors for cultivating administration of factory department, so that intelligence cultivates robot
According to the collected information of sensor, to cultivation, factory carries out intelligent management, for example, when optical sensor 7-1-5 detect it is feeding
When growing illumination deficiency in factory, intelligence cultivates robot 6-1 and can open the window for cultivating factory, or opens dedicated light compensating lamp
(LED bulb that the light compensating lamp can be the characteristic spectrum of livestock-raising use), carries out light filling.
Consumer end 4 shown in Fig. 1 refers to the terminal device that consumer-user uses when buying product, can be
But it is not limited to various forms of mobile phones.Pattern identification code (the example in the scanning product packaging of consumer end 4 is utilized in purchase user
Such as, the two dimensional code 8 on egg packaging case shown in Fig. 3), first identifier information is obtained when consumer end 4 is based on scanning recognition
In the case where, the Cloud Server communicated with consumer end 4 shows query interface by consumer end, and in consumer-user
By the character mark code on any one product surface to be checked in the query interface input product packing case (for example, Fig. 3 institute
Character 9 in the egg packaging case shown in egg surface) in the case where, due to storing tracing back for each product on Cloud Server 2
Source information, thus, the character mark code on the product surface that Cloud Server 2 receives user's input by above-mentioned query interface
Afterwards, the information of tracing to the source of the product can be inquired according to the character mark code, wherein include that product to be checked exists in information of tracing to the source
Production information and preconfigured pattern identification code corresponding with the character mark code in entire production process, so as to root
According to by comparing the pattern identification code on the pattern identification code and product packaging that the character mark code returns, and then verifying product packet
The true and false for the pattern identification code loaded onto.
Due to the character mark by being manually entered on product surface, user experience will affect, thus, it can as another kind
The embodiment of choosing, in the case that consumer end 4 obtains first identifier information based on scanning recognition, on consumer end 4
The query interface of display can also support image identification function.As shown in figure 4, i.e. consumer is without being manually entered on product surface
Character mark code (for example, WC20170122), but click and take pictures inquiry, consumer end 4 detects shooting instruction, then beats
Camera is opened, and is attached with the egg surface of character mark code " WC20170122 " by shooting, to identify in egg surface
Character mark code.
When user is when shooting the image of product surface, due to the influence of ambient enviroment, the image that may cause shooting is rushed
There are noises, in addition, the angle of shooting is different, character mark in the image of the product surface of shooting may be made to exist centainly
Tilt angle first has to as a result, when carrying out image recognition to the image taken to collected product surface to be checked
Image is pre-processed, wherein pretreatment includes at least one following: gray processing processing, denoising, correction process.
When identifying character image, first have to collected color image being converted to gray level image, as one
The optional embodiment of kind, can be converted to grayscale image for acquired image by following algorithm:
I=0.3B+0.59G+0.11R；
Wherein, I is the gray value of each pixel, and B is that for each pixel in the component of channel B, G is in original image in original image
Component of each pixel in the channel G, component of the R for each pixel in original image in the channel R；
Further, in order to enhance the signal-to-noise ratio of image, can by following algorithm by it is collected it is noise-containing to
The image for inquiring product surface, by field average treatment, image after being denoised:
Wherein, P for each neighborhood pixels in taken field coordinate, Q be field in include neighborhood pixels number, f (x,
It y) is collected noise-containing image, g (x, y) is the image after denoising；
In addition, when the image to collected product surface to be checked is corrected, it is thus necessary to determine that character mark in image
Know the tilt angle of code, and then correction process is carried out to acquired image according to determining tilt angle.As a kind of optional
Embodiment, can by following algorithm by acquired image along each predetermined inclination angle carry out Radon transformation, calculate
The sum of the gradient absolute value of the corresponding projecting integral's figure of each predetermined inclination angle, and the accumulated value of gradient absolute value is maximum
Tilt angle is determined as the tilt angle of original image:
Rφ(x')=∫ f (x'cos φ-y'sin φ, x'sin φ-y'cos φ) dy'；
Wherein, φ indicates predetermined inclination angle, Rφ() indicates to carry out Radon transformation along the direction φ, and f (x, y) is to collect
The inclined image of character mark code.
In order to realize quick identification, without carrying out cutting to the character in character image, the application uses convolutional Neural net
Network (CNN) and recurrent neural network (RNN) are come the continuous knowledge to the image progress word sequence for containing entire character mark code
Not.Specifically, it is primarily based on convolutional neural networks identification model trained in advance, extracts and is wrapped in the image of product surface to be checked
Then characteristic will be input in recurrent neural network classifier by the characteristic of the character mark code contained, and be based on passing
Return neural network classifier, according to characteristic, the output data of last moment recurrent neural network classifier and upper a period of time
The vector data that the character mark code that recurrent neural network classifier identifies is converted to is carved, the knowledge of character mark code is sequentially output
Other result.
As an alternative embodiment, the forward algorithm formula that uses of above-mentioned recurrent neural network classifier can be with
Are as follows:
Wherein, b0=0；
Wherein, D is the dimension of input vector, and H is the number of the neuron of hidden layer, and K is of the neuron of output layer
Number, x are the characteristic that convolutional neural networks extract,For hidden layer neuron in current time recurrent neural network
Input,For the output of hidden layer neuron in current time recurrent neural network, θ () isIt arrivesFunction；wih、wh'hPoint
It is notCorresponding weight coefficient, in a forward algorithm transmittance process, each moment wih、wh'hIt is total across timing
Enjoy, across timing it is shared refer to recurrent neural network in signal forward direction transmittance process, each moment wih、wh'hValue it is identical, no
In the same timeValue it is identical, reduce the complexity of model parameter, also avoid the line of model complexity
Property increase caused by over-fitting.For the input of current time recurrent neural network output layer neuron；wh'hFor each mind of output layer
Through the corresponding weight of member,For the output of current time recurrent neural network output layer neuron,For a probability value, characterization
Current time corresponds to the opposite ratio with the adduction of all neuritis output valves of output layer of neuron output value.
As can be seen from the above formula that in recurrent neural network used in this application hidden layer neuron input data
Including the training sample feature that CNN is extracted, the output data of last moment recurrent neural network hidden layer.Therefore, this Shen
The recurrent neural network that please be used both had relied on the feature of image, had been also relied on when predicting word (word) at current time
The feature (language model) of last moment output, improves recognition efficiency and accuracy rate.
It should be noted that extracting product surface to be checked based on convolutional neural networks identification model trained in advance
Image in include character mark code characteristic before, it is also necessary to training convolutional neural networks identification model.In order to mention
The accuracy of high convolutional neural networks identification model can carry out training convolutional neural networks identification model by the following method.Tool
Body extracts the Gabor characteristic in image eight directions of product surface to be checked using Gabor filter first, then will be to
The image of product surface and the image Jing Guo Gabor filter feature extraction are inquired as the defeated of convolutional neural networks identification model
Enter, will will finally test the highest convolutional neural networks model of recognition correct rate and be determined as extracting the image of product surface to be checked
In comprising character recognition code characteristic convolutional neural networks identification model.It is easy it is noted that the volume that the application uses
Product neural network recognization model is the neural network for including two layers of convolutional layer and one layer of multireel lamination.
As an alternative embodiment, the formula for the Gabor filter that the application uses are as follows:
Wherein, (x, y) identifies location of pixels, and M is direction number,Indicate direction, σ identifier space scale factor.
As shown in Fig. 2, above system can also include: the first stamp equipment 10 and the second stamp equipment 11, will pass through
First scanning device 10 generates the pattern identification code of product packaging, and pattern identification code is printed upon in corresponding product packaging；It is logical
The second stamp equipment 11 is crossed, generates the character mark code of each product in product packaging, and the identification code of each product is printed
On corresponding product surface.
Optionally, by taking egg as an example, be for the pattern identification code (for example, two dimensional code) with a batch of chicken it is identical,
Different character mark codes is set on the egg of certain a batch of chicken production, i.e., corresponds to multiple words with a batch of pattern identification code
Accord with identification code.
As a kind of optional embodiment, above system can also include: raiser's terminal 3, communicate with Cloud Server, use
In the cultivation information for remotely monitoring at least one cultivation factory.Optionally, raiser's terminal include but is not limited to it is following any one:
Computer, mobile phone, tablet computer, laptop etc..
In addition, it should also be noted that, the application by Cloud Server 2 and it is each cultivation factory internet-of-things terminal (for example,
The internet-of-things terminal 5-1 for cultivating factory 1-1, the internet-of-things terminal 5-2 for cultivating factory 1-2, the internet-of-things terminal 5-n for cultivating factory 1-n) it is logical
Letter can provide cloud aquaculture management, i.e., acquire each henhouse by the various sensors disposed in each cultivation each henhouse of factory
Breeding environment data in the growth information of interior baby chick and each henhouse, it is whole by the Internet of Things disposed in each cultivation factory
End is uploaded to Cloud Server 2, so that a large amount of cloud cultivation data are stored on Cloud Server 2, by analyzing these data,
Can with the henhouse of each cultivation factory of scientific management, to the growth of every a collection of chicken, environment, feeding, vaccinate, information of laying eggs carries out
Digitization analysis.
A kind of optional embodiment, each cultivation factory that cultivation expert can also be monitored by Cloud Server 2 is supported
Video data is grown, to obtain the whole process of raiser's cultivation, intuitive analysis cultivation data, and cultivation is provided for raiser and builds
View, to improve breeding efficiency.
Optionally, Cloud Server 2 can use artificial intelligence, and intelligentized cloud aquaculture management is provided by machine learning
Scheme, including but not limited to it is following any one: (1) data are cultivated (for example, materials data, growth according to collected history
The environmental datas such as situation, temperature and humidity) prediction next period cultivation data, it is preferable that different regions, difference can also be directed to
Weather provide differentiation adjustment；(2) data that the diagnostic data of cultivation factory, raiser are putd question to according to each expert of collection
Deng prediction epidemic situation, to remind raiser to prevent in time；(3) a large amount of cultivation data and city's number of fields of the offer of acquisition internet are provided
According to intellectual analysis predicts most suitable breeding way or cultivation data.
As shown in Figure 1, the cultivation data of each cultivation factory are uploaded to Cloud Server by internet-of-things terminal, will pass through cloud
Server come manage it is each cultivation factory cultivation data, realize cloud aquaculture management.In cloud management system for breeding, in order to realize
Retrospect to each cultured product production process, a kind of anti-fake method of tracing to the source this application provides multi-code can be applied but not
It is limited in multi-code traceability anti-fake system shown in FIG. 1.As shown in figure 5, this method comprises the following steps:
Step S501, based on the pattern identification code in scanning product packaging, identification obtains first identifier information.
Specifically, the pattern identification code in product packaging can be but not limited to bar code, two dimensional code；First identifier information
It can be the information of characterization rule of origin.The said goods can be any one product for needing quality tracing, including but unlimited
In various meats, eggs livestock products.By taking the birds, beasts and eggs of poultry (for example, laying hen) production as an example, first identifier information can be fowl
Which give up code (for example, henhouse code), for which pouity dwelling place of the birds, beasts and eggs (for example, egg) from cultivation factory to be characterized.
Optionally, poultry can be any one in chicken, duck, goose, quail etc., and birds, beasts and eggs can be egg, duck's egg, goose
Any one in egg, quail egg, by taking egg as an example, consumer-user passes through on the terminal device scans egg packagings such as mobile phone
After pattern identification code (for example, two dimensional code), the culturing area and scale of cultivation factory can be checked, and reproduce the production of product
Journey.
Step S502 shows query interface in the case where identification obtains first identifier information；Wherein, query interface is used
Character mark code in user's input product is packed on any one product surface, to inquire corresponding product traceability information.
Specifically, after identifying to obtain the identification information of rule of origin by step S501, query interface can be exported, it should
Query interface further inquires the information of tracing to the source of any one product product packaging Nei for user, so that user can input certain
Character mark code on a product surface, to inquire the information of tracing to the source of the corresponding character mark code.
Step S504 is based on query interface, obtains the character mark code on the product surface to be checked of user's input.
Specifically, the character mark code in product packaging on any one product surface can be manually entered into and look by user
It askes in interface, " clicking shooting inquiry " button shown on query interface can also be clicked, include character mark code by shooting
Product surface image, using image recognition technology come the character mark code on automatic identification product surface, so as to according to should
Character mark code inquires corresponding information of tracing to the source.
Step S504 exports the product traceability letter of product to be checked according to the character mark code on product surface to be checked
Breath；Wherein, product traceability information includes: production information of the product to be checked in entire production process, preconfigured and word
Accord with the corresponding pattern identification code of identification code, wherein preconfigured pattern identification code corresponding with character mark code is for verifying
The true and false of pattern identification code in product packaging.
Specifically, as user, by terminal devices such as mobile phones, the input product in query interface packs some interior product surface
On character mark code after, Cloud Server can return to corresponding query result, and pass through according to the character mark code received
The query interface shown on the terminal devices such as mobile phone is shown.It is easy it is noted that query result corresponding with the character mark code
In, it not only include production information of the product in entire production process, but also preconfigured with the character mark including producer
The corresponding pattern identification code of code, will pass through pattern identification in pattern identification verifying product packaging corresponding with the character mark code
The true and false of code.
Optionally, still by taking egg as an example, consumer-user inputs the character mark on egg by terminal devices such as mobile phones
After code, the data such as growth, environment, temperature, humidity, feeding raising, the daily monitoring information of laying hen of production egg can be checked.
Preferably, figure identification code corresponding with character mark code on egg in query result is identified, it can also be according to identification
As a result the true and false of the pattern identification code on egg packaging case is verified, further to prevent from buying fake products.
Step S505, based on preconfigured pattern identification code corresponding with character mark code is scanned, identification obtains third
Identification information.
Specifically, user can be by inquiring on the terminal device scans query interface such as mobile phone and the character mark code
Corresponding pattern identification code, obtains third identification information.
Step S506, by comparing third identification information and first identifier information, to verify the figure mark in product packaging
Know the true and false of code.
Specifically, third identification information and first identifier information are compared, if the two is consistent, in the figure for showing product packaging
Shape identification code is not to forge；, whereas if the two is inconsistent, then show that the pattern identification code in product packaging is to forge.
As a kind of optional embodiment, in the case that above-mentioned product to be checked is birds, beasts and eggs, according to product to be checked
Character mark code on surface, before the product traceability information for exporting product to be checked, method further include: obtain with a collection of poultry
Cultivation factory information and pouity dwelling place information；According to the cultivation factory information and pouity dwelling place information of same a collection of poultry, corresponding figure mark is generated
Know code；Based on pattern identification code, to set corresponding character mark code with each birds, beasts and eggs of a collection of poultry production；Wherein, same
The each birds, beasts and eggs for criticizing poultry production have unique character mark code.Optionally, poultry can be laying hen, and birds, beasts and eggs can be chicken
Egg, pouity dwelling place can be henhouse.
Optionally, the said goods information of tracing to the source includes at least one following: cultivation factory's information, pouity dwelling place information, poultry information,
Poultry growth information, poultry materials information, poultry vaccine information, breeding environment information.
In order to improve the safety of data, as a kind of optional embodiment, it is being based on pattern identification code, for a collection of family
After each birds, beasts and eggs of fowl production set corresponding character mark code, the above method can also include the following steps: to construct block
Chain network, wherein the block node in block chain network includes following at least one；Poultry seedling (for example, baby chick) supplier,
Supply of forage quotient, vaccine supplier, poultry farming factory, birds, beasts and eggs retailer, consumer；Character mark code based on each birds, beasts and eggs,
On the block chain that the product traceability information of each birds, beasts and eggs is stored into block chain network to each block node.
The information of tracing to the source of each product is stored by block chain, it is ensured that data can not tamper.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part explanation.
Professional should further appreciate that, described in conjunction with the examples disclosed in the embodiments of the present disclosure
Unit and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, hard in order to clearly demonstrate
The interchangeability of part and software generally describes each exemplary composition and step according to function in the above description.
These functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.
Professional technician can use different methods to achieve the described function each specific application, but this realization
It is not considered that exceeding scope of the present application.
The above description is only an example of the present application, is not intended to limit this application.For those skilled in the art
For, various changes and changes are possible in this application.All any modifications made within the spirit and principles of the present application are equal
Replacement, improvement etc., should be included within the scope of the claims of this application.
Claims (15)
- A kind of anti-fake method 1. multi-code is traced to the source, which is characterized in that described method includes following steps:Based on the pattern identification code in scanning product packaging, identification obtains first identifier information；In the case where identification obtains the first identifier information, query interface is shown；Wherein, the query interface is used for user The character mark code in the product packaging on any one product surface is inputted, to inquire corresponding product traceability information；Based on the query interface, the character mark code on the product surface to be checked of user's input is obtained；According to the character mark code on the product surface to be checked, the product traceability information of the product to be checked is exported；Its In, the product traceability information include: production information of the product to be checked in entire production process, it is preconfigured with The corresponding pattern identification code of the character mark code, wherein the preconfigured figure corresponding with the character mark code Identification code is used to verify the true and false of the pattern identification code in the product packaging.
- 2. the method according to claim 1, wherein according to the character mark on the product surface to be checked Yard, after the product traceability information for exporting the product to be checked, the method also includes:Based on the preconfigured pattern identification code corresponding with the character mark code is scanned, identification obtains third mark letter Breath；By comparing the third identification information and the first identifier information, to verify the pattern identification in the product packaging The true and false of code.
- 3. method according to claim 1 or 2, which is characterized in that the product to be checked is birds, beasts and eggs, and according to Character mark code on product surface to be checked, before the product traceability information for exporting the product to be checked, the method is also Include:Obtain cultivation factory information and pouity dwelling place information with a collection of poultry；According to the cultivation factory information and pouity dwelling place information with a collection of poultry, corresponding pattern identification code is generated；Based on the pattern identification code, corresponding character mark code is set for each birds, beasts and eggs with a collection of poultry production；Its In, each birds, beasts and eggs with a collection of poultry production have unique character mark code.
- 4. according to the method described in claim 3, it is characterized in that, the product traceability information includes at least one following: supporting Grow factory's information, pouity dwelling place information, poultry information, poultry growth information, poultry materials information, poultry vaccine information, breeding environment letter Breath；Wherein, it is being based on the pattern identification code, is setting corresponding character mark for each birds, beasts and eggs with a collection of poultry production After code, the method also includes:Construct block chain network, wherein the block node in the block chain network includes following at least one；Poultry seedling supplies Answer quotient, supply of forage quotient, vaccine supplier, poultry farming factory, birds, beasts and eggs retailer, consumer；Character mark code based on each birds, beasts and eggs stores the product traceability information of each birds, beasts and eggs to the block chain network In each block node block chain on.
- 5. obtaining user's input the method according to claim 1, wherein being based on the query interface Character mark code on product surface to be checked, comprising:Based on the query interface, detect whether to receive shooting instruction, wherein the shooting instruction be starting shooting it is described to Inquire the instruction of the image of product surface；In the case where receiving the shooting instruction, the image of the product surface to be checked is acquired, wherein described to be checked It include the character mark code in the image of product surface；Based on convolutional neural networks identification model trained in advance, the word for including in the image of the product surface to be checked is extracted Accord with the characteristic of identification code；The characteristic is input in recurrent neural network classifier, and is based on the recurrent neural network classifier, root According to the characteristic, the output data of last moment recurrent neural network classifier and last moment recurrent neural network point The vector data that the character mark code that class device identifies is converted to is sequentially output the recognition result of the character mark code.
- 6. according to the method described in claim 5, it is characterized in that, the forward algorithm that the recurrent neural network classifier uses Are as follows:Wherein, b0=0；Wherein, D is the dimension of input vector, and H is the number of the neuron of hidden layer, and K is the number of the neuron of output layer, x For the characteristic that convolutional neural networks extract,For in current time recurrent neural network hidden layer neuron it is defeated Enter,For the output of hidden layer neuron in current time recurrent neural network, θ () isIt arrivesFunction；wih、wh'hRespectively ForCorresponding weight coefficient,For the input of current time recurrent neural network output layer neuron；wh'hFor output layer The corresponding weight of each neuron,For the output of current time recurrent neural network output layer neuron,For a probability value, Characterization current time corresponds to the opposite ratio with the adduction of all neuritis output valves of output layer of neuron output value.
- 7. according to the method described in claim 5, it is characterized in that, to the image of the collected product surface to be checked into Row pretreatment, wherein the pretreatment includes at least one following: gray processing processing, denoising, correction process.
- 8. the method according to the description of claim 7 is characterized in that by following algorithm to the collected product to be checked The image on surface carries out gray processing processing:I=0.3B+0.59G+0.11R；Wherein, I is the gray value of each pixel, and B is that for each pixel in the component of channel B, G is each in original image in original image Component of the pixel in the channel G, component of the R for each pixel in original image in the channel R.
- 9. the method according to the description of claim 7 is characterized in that will be collected noise-containing to be checked by following algorithm The image for asking product surface, by field average treatment, image after being denoised:Wherein, P is the coordinate of each neighborhood pixels in taken field, and Q is the number for the neighborhood pixels for including in field, and f (x, y) is Collected noise-containing image, g (x, y) are the image after denoising.
- 10. the method according to the description of claim 7 is characterized in that by following algorithm by collected product table to be checked The image in face carries out Radon transformation along each predetermined inclination angle, calculates the corresponding projecting integral's figure of each predetermined inclination angle The sum of gradient absolute value, and the maximum tilt angle of the accumulated value of gradient absolute value is determined as to the tilt angle of original image, Original image is corrected according to determining tilt angle, the image after being corrected:Rφ(x')=∫ f (x'cos φ-y'sin φ, x'sin φ-y'cos φ) dy'；Wherein, φ indicates predetermined inclination angle, Rφ() indicates to carry out Radon transformation along the direction φ, and f (x, y) is collected word Accord with the inclined image of identification code.
- 11. according to the method described in claim 5, it is characterized in that, identifying mould based on convolutional neural networks trained in advance Type, before the characteristic for extracting the character mark code for including in the image of the product surface to be checked, the method is also wrapped It includes:The feature in eight directions of image of the product surface to be checked is extracted using filter；Using the image of the product surface to be checked and the image extracted by the filter characteristic as convolutional neural networks The input of identification model, wherein the convolutional neural networks identification model be include two layers of convolutional layer and one layer of multireel lamination Neural network；The highest convolutional neural networks model of recognition correct rate will be tested and be determined as the extraction product surface to be checked The convolutional neural networks identification model of characteristic in image comprising character recognition code.
- 12. according to the method for claim 11, which is characterized in that the filter is Gabor filter, the Gabor The formula of filter extraction feature are as follows:Wherein, (x, y) identifies location of pixels, and M is direction number,Indicate direction, σ identifier space scale factor.
- The anti-fake system 13. a kind of multi-code is traced to the source characterized by comprisingConsumer end, for scanning the pattern identification code in product packaging, identification obtains first identifier information；Cloud Server is communicated with the consumer end, obtains the first identifier letter for identifying in the consumer end In the case where breath, query interface is shown, and the word on the product surface to be checked of user's input is obtained based on the query interface Identification code is accorded with, according to the character mark code on the product surface to be checked, exports the product traceability letter of the product to be checked Breath；Wherein, the product traceability information includes: production information of the product to be checked in entire production process, matches in advance The pattern identification code corresponding with the character mark code set, wherein described preconfigured corresponding with the character mark code Pattern identification code be used to verify the true and false of pattern identification code in the product packaging.
- 14. system according to claim 13, which is characterized in that the system also includes:First stamp equipment is printed upon corresponding production for generating the pattern identification code of product packaging, and by the pattern identification code In product packaging；Second stamp equipment, for generating the character mark code of each product in the product packaging, and by the mark of each product Know code to be printed upon on corresponding product surface.
- 15. system according to claim 13, which is characterized in that the system also includes:Equipment is acquired, is communicated with the Cloud Server, to acquire the product traceability information of each product；Wherein, the cloud service Character mark code of the device based on each product, the product traceability information of each product is stored into block chain network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810698368.8A CN109034837B (en) | 2018-06-29 | 2018-06-29 | Multi-code tracing anti-counterfeiting method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810698368.8A CN109034837B (en) | 2018-06-29 | 2018-06-29 | Multi-code tracing anti-counterfeiting method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109034837A true CN109034837A (en) | 2018-12-18 |
| CN109034837B CN109034837B (en) | 2020-12-29 |
Family
ID=65522014
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810698368.8A Active CN109034837B (en) | 2018-06-29 | 2018-06-29 | Multi-code tracing anti-counterfeiting method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109034837B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109784951A (en) * | 2019-01-29 | 2019-05-21 | 浙江甲骨文超级码科技股份有限公司 | A kind of anti-fake traceability system of product whole process based on block chain technology |
| CN110490305A (en) * | 2019-08-22 | 2019-11-22 | 腾讯科技（深圳）有限公司 | Machine learning model processing method and node based on block chain network |
| CN110782265A (en) * | 2019-11-12 | 2020-02-11 | 北京海益同展信息科技有限公司 | Information processing method, device, system and computer readable storage medium |
| CN111369261A (en) * | 2018-12-24 | 2020-07-03 | 阿里巴巴集团控股有限公司 | Product tracing method and system and product tracing information processing method |
| WO2020207216A1 (en) * | 2019-04-09 | 2020-10-15 | 新立讯科技股份有限公司 | Method and apparatus for generating and querying tracing code of commodity |
| WO2020223905A1 (en) * | 2019-05-07 | 2020-11-12 | 林晖 | Method for tracking product history |
| CN112016535A (en) * | 2020-10-26 | 2020-12-01 | 成都合能创越软件有限公司 | Vehicle-mounted garbage traceability method and system based on edge calculation and block chain |
| CN112488109A (en) * | 2020-12-10 | 2021-03-12 | 深圳市云辉牧联科技有限公司 | Identification method and device of livestock and poultry identification codes and computer readable storage medium |
| CN114241248A (en) * | 2022-02-24 | 2022-03-25 | 北京市农林科学院信息技术研究中心 | River crab origin tracing method and system |
| WO2022077135A1 (en) * | 2020-10-15 | 2022-04-21 | 江苏图码信息科技有限公司 | Anti-counterfeiting tracing system and application component for associated customized graphic code |
| CN116132107A (en) * | 2022-12-16 | 2023-05-16 | 苏州可米可酷食品有限公司 | Full life cycle quality data traceability management system based on data cloud processing product |
| CN116911883A (en) * | 2023-09-14 | 2023-10-20 | 新立讯科技股份有限公司 | Agricultural product anti-counterfeiting tracing method and cloud platform based on AI (advanced technology) authentication technology and tracing quantification |
| CN118228941A (en) * | 2024-05-24 | 2024-06-21 | 南京龙芯源智能科技有限公司 | Processing method and system based on industrial Internet identification |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105654135A (en) * | 2015-12-30 | 2016-06-08 | 成都数联铭品科技有限公司 | Image character sequence recognition system based on recurrent neural network |
| CN105678292A (en) * | 2015-12-30 | 2016-06-15 | 成都数联铭品科技有限公司 | Complex optical text sequence identification system based on convolution and recurrent neural network |
| US20160196476A1 (en) * | 2014-06-13 | 2016-07-07 | Grg Banking Equipment Co., Ltd. | Multi-cue fusion based ticket positioning recognition method and system |
| CN107330581A (en) * | 2017-06-08 | 2017-11-07 | 上海交通大学 | Agricultural product quality information system based on block chain |
-
2018
- 2018-06-29 CN CN201810698368.8A patent/CN109034837B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160196476A1 (en) * | 2014-06-13 | 2016-07-07 | Grg Banking Equipment Co., Ltd. | Multi-cue fusion based ticket positioning recognition method and system |
| CN105654135A (en) * | 2015-12-30 | 2016-06-08 | 成都数联铭品科技有限公司 | Image character sequence recognition system based on recurrent neural network |
| CN105678292A (en) * | 2015-12-30 | 2016-06-15 | 成都数联铭品科技有限公司 | Complex optical text sequence identification system based on convolution and recurrent neural network |
| CN107330581A (en) * | 2017-06-08 | 2017-11-07 | 上海交通大学 | Agricultural product quality information system based on block chain |
Non-Patent Citations (1)
| Title |
|---|
| 安和林（/USERS/MESCNOCLY2TE）: ""深度|区块链在天猫国际商品溯源中的应用"", 《HTTPS://YQ.ALIYUN.COM/ARTICLES/348787?UTM_CONTENT=M_40006》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111369261A (en) * | 2018-12-24 | 2020-07-03 | 阿里巴巴集团控股有限公司 | Product tracing method and system and product tracing information processing method |
| CN109784951A (en) * | 2019-01-29 | 2019-05-21 | 浙江甲骨文超级码科技股份有限公司 | A kind of anti-fake traceability system of product whole process based on block chain technology |
| WO2020207216A1 (en) * | 2019-04-09 | 2020-10-15 | 新立讯科技股份有限公司 | Method and apparatus for generating and querying tracing code of commodity |
| WO2020223905A1 (en) * | 2019-05-07 | 2020-11-12 | 林晖 | Method for tracking product history |
| CN110490305A (en) * | 2019-08-22 | 2019-11-22 | 腾讯科技（深圳）有限公司 | Machine learning model processing method and node based on block chain network |
| CN110782265A (en) * | 2019-11-12 | 2020-02-11 | 北京海益同展信息科技有限公司 | Information processing method, device, system and computer readable storage medium |
| WO2022077135A1 (en) * | 2020-10-15 | 2022-04-21 | 江苏图码信息科技有限公司 | Anti-counterfeiting tracing system and application component for associated customized graphic code |
| CN112016535A (en) * | 2020-10-26 | 2020-12-01 | 成都合能创越软件有限公司 | Vehicle-mounted garbage traceability method and system based on edge calculation and block chain |
| CN112488109B (en) * | 2020-12-10 | 2024-03-29 | 深圳市云辉牧联科技有限公司 | Method and device for identifying livestock and poultry identification codes and computer readable storage medium |
| CN112488109A (en) * | 2020-12-10 | 2021-03-12 | 深圳市云辉牧联科技有限公司 | Identification method and device of livestock and poultry identification codes and computer readable storage medium |
| CN114241248A (en) * | 2022-02-24 | 2022-03-25 | 北京市农林科学院信息技术研究中心 | River crab origin tracing method and system |
| CN114241248B (en) * | 2022-02-24 | 2022-07-01 | 北京市农林科学院信息技术研究中心 | River crab origin tracing method and system |
| CN116132107A (en) * | 2022-12-16 | 2023-05-16 | 苏州可米可酷食品有限公司 | Full life cycle quality data traceability management system based on data cloud processing product |
| CN116132107B (en) * | 2022-12-16 | 2024-04-12 | 苏州可米可酷食品有限公司 | Full life cycle quality data traceability management system based on data cloud processing product |
| CN116911883A (en) * | 2023-09-14 | 2023-10-20 | 新立讯科技股份有限公司 | Agricultural product anti-counterfeiting tracing method and cloud platform based on AI (advanced technology) authentication technology and tracing quantification |
| CN116911883B (en) * | 2023-09-14 | 2023-12-19 | 新立讯科技股份有限公司 | Agricultural product anti-counterfeiting tracing method and cloud platform based on AI (advanced technology) authentication technology and tracing quantification |
| CN118228941A (en) * | 2024-05-24 | 2024-06-21 | 南京龙芯源智能科技有限公司 | Processing method and system based on industrial Internet identification |
| CN118228941B (en) * | 2024-05-24 | 2024-08-30 | 南京龙芯源智能科技有限公司 | Processing method and system based on industrial Internet identification |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109034837B (en) | 2020-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109034837A (en) | Multi-code is traced to the source anti-fake method and system | |
| CN110402840B (en) | Live pig monitoring terminal and live pig monitoring system based on image recognition | |
| Zhen et al. | Impact of tree-oriented growth order in marker-controlled region growing for individual tree crown delineation using airborne laser scanner (ALS) data | |
| CN109514582A (en) | A kind of robot is funny to dote on control device and mobile robot | |
| Giese | Guidelines for people approaching breeding groups of Adélie penguins (Pygoscelis adeliae) | |
| CN109325495A (en) | A kind of crop image segmentation system and method based on deep neural network modeling | |
| CN107657459A (en) | Auth method, settlement of insurance claim method, source tracing method, storage medium and the electronic equipment of livestock | |
| CN109479746A (en) | A kind of robot is funny to dote on control method and chip | |
| CN107833145A (en) | The database building method and source tracing method of livestock, storage medium and electronic equipment | |
| JP2011248502A (en) | Information processing device, information output method and program | |
| CN114898405B (en) | Portable broiler chicken anomaly monitoring system based on edge calculation | |
| KR20210080055A (en) | System for detecting pests of shiitake mushrooms | |
| CN107423887A (en) | A kind of poultry monitoring and managing method and system based on mobile Internet | |
| CN108922621A (en) | A kind of appraisal procedure and device of the grade that disables | |
| CN109191461A (en) | A kind of Countryside Egg recognition methods and identification device based on machine vision technique | |
| JP2022114352A (en) | Estimation system by artificial intelligence (ai), learning data generator, learning device, picking object estimation device, learning system, and program | |
| JP6848998B2 (en) | Learning system, learning method and learning program | |
| Dandekar et al. | Weed plant detection from agricultural field images using yolov3 algorithm | |
| Saradopoulos et al. | Edge computing for vision-based, urban-insects traps in the context of smart cities | |
| CN111178172A (en) | Laboratory mouse sniffing action recognition method, module and system | |
| CN110766654A (en) | Live bird detection method, device and equipment based on machine learning and readable medium | |
| Lankheet et al. | Acquired versus innate prey capturing skills in super-precocial live-bearing fish | |
| CN109190663A (en) | A method of the identification cigarette case product rule based on depth learning technology | |
| CN112507770A (en) | Rice disease and insect pest identification method and system | |
| Negrão et al. | Plant phenotyping |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |