CN109308760B - AI cloud system of paper currency ware discernment - Google Patents
AI cloud system of paper currency ware discernment Download PDFInfo
- Publication number
- CN109308760B CN109308760B CN201810949973.8A CN201810949973A CN109308760B CN 109308760 B CN109308760 B CN 109308760B CN 201810949973 A CN201810949973 A CN 201810949973A CN 109308760 B CN109308760 B CN 109308760B
- Authority
- CN
- China
- Prior art keywords
- paper money
- data
- identification
- paper
- recognition
- 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.)
- Expired - Fee Related
Links
- 238000012549 training Methods 0.000 claims abstract description 6
- 238000010801 machine learning Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 3
- 230000009747 swallowing Effects 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 2
- 238000012423 maintenance Methods 0.000 abstract description 3
- 238000000034 method Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013479 data entry Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/004—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using digital security elements, e.g. information coded on a magnetic thread or strip
- G07D7/0047—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using digital security elements, e.g. information coded on a magnetic thread or strip using checkcodes, e.g. coded numbers derived from serial number and denomination
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/20—Testing patterns thereon
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
- Inspection Of Paper Currency And Valuable Securities (AREA)
Abstract
The invention discloses an AI cloud system for improving the identification accuracy of a paper money device, which comprises an AI cloud platform and at least one paper money device connected with the AI cloud platform through a network, wherein the AI cloud platform comprises a paper money identification device, a data acquisition system, a model trainer and an operation system, the data input end of the model trainer is respectively connected with the data acquisition system and the operation system, the data output end of the model trainer is respectively connected with the paper money identification device and the paper money device, the paper money identification device is in data connection with the paper money device, the paper money device is in data connection with the data acquisition system, the model trainer performs paper money model training according to data provided by the data acquisition system and the operation system to generate a new paper money identification model, and the data of the new paper money identification device is sent to the paper money identification device and the paper money device. According to the invention, through the AI cloud, the paper currency identification rules and the false paper currency blacklist can be updated in real time, the identification accuracy of the paper currency device is improved, and the operation and maintenance cost is reduced.
Description
Technical Field
The invention relates to the field of paper money devices, in particular to an AI cloud system for identifying a paper money device.
Background
The paper currency device is a very key tool on a vending machine and an ATM, and the identification accuracy of the paper currency device is also very key. However, the existing paper money devices are all closed systems, have no machine learning function, and are difficult to follow the new trend of the paper money in real time so as to accurately identify the paper money in time.
Aiming at the characteristic that the existing paper money devices are closed, the AI cloud is connected with the paper money devices and is responsible for tracking the identification data of the paper money devices and learning the data to form a new identification model, and the identification model is synchronized to each paper money device in real time, so that the identification accuracy of the paper money devices is improved in time.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an AI cloud system for identifying paper currency devices.
The technical scheme of the invention is as follows:
an AI cloud system for identifying paper money devices comprises an AI cloud platform and at least one paper money device which is in network connection with the AI cloud platform, wherein the AI cloud platform comprises a paper money identification device, a data acquisition system, a model trainer and an operation system, the data input end of the model trainer is respectively connected with the data acquisition system and the operation system, and the data output end of the model trainer is respectively connected with the paper money identification device and the paper money device;
the paper money recognition device is in data connection with the paper money device, and is used for recognizing paper money to be detected by the paper money device in real time or recognizing difficult paper money and feeding back a recognition result to the paper money device in real time;
the paper money device is in data connection with the data acquisition system, and the data acquisition system is used for acquiring the data of paper money recognized by the paper money device and sending the data to the model trainer;
the operation system is used for marking labels on the reported identification data or inputting wrong indexes of paper money by operators and sending the index information data to the model trainer;
and the model trainer performs paper money model training according to the data provided by the data acquisition system and the operation system to generate a new paper money recognition model and sends the data of the new paper money recognition device to the paper money recognition device and the paper money device.
In the above technical solution, the data collected by the data collection system for identifying paper money by the paper money collector includes a serial number of the paper money collector, a serial number of the paper money, identification time, collected paper money information, and an identification result.
In the above technical solution, the collected data further includes front and back images of the paper money.
In the technical scheme, the model trainer forms different characteristics according to different positions of paper money based on collected data and then learns by using a corresponding machine learning algorithm.
In the technical scheme, the error indexes of the paper money input by the operator comprise a counterfeit money blacklist and error identification label data.
In the technical scheme, the paper money device has a paper money identification function, and collects paper money data through a hardware device and identifies the data.
In the technical scheme, the paper money collector can be used for spitting or swallowing paper money according to the identification result.
In the technical scheme, the paper money device reports the unreported paper money identification data to the data acquisition system at regular time.
In the technical scheme, the time reported by the paper money counter at regular time is 5 minutes.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the method, the data of the paper money is collected in real time, model training is carried out according to the label data, the recognition accuracy of the paper money is improved, and meanwhile, the accuracy of unknown paper money recognition is improved by real-time model judgment on difficult paper money;
2. the new counterfeit money is rejected in a wind control mode in time through a real-time counterfeit money blacklist, so that loss is reduced;
3. through model distribution, new paper money identification models and machine learning models are distributed in real time, the operation and maintenance cost is reduced, and the updating speed of paper money identification rules is increased.
Drawings
FIG. 1 is a functional block diagram of the present invention;
FIG. 2 is a flow chart of banknote machine and AI cloud interaction;
FIG. 3 is a flow chart of AI cloud internal learning;
fig. 4 is a flow chart of operator data entry.
Detailed Description
The invention is described in detail below with reference to the figures and the specific embodiments.
Referring to fig. 1 to 4, the AI cloud system for identifying paper money devices provided by the present invention includes an AI cloud platform and at least one paper money device connected to the AI cloud platform via a network, wherein the AI cloud platform includes a paper money identification device, a data acquisition system, a model trainer and an operation system, a data input end of the model trainer is respectively connected to the data acquisition system and the operation system, and a data output end of the model trainer is respectively connected to the paper money identification device and the paper money device.
The paper money identification device is in data connection with the paper money device and is used for identifying paper money to be detected by the paper money device in real time or identifying difficult paper money and feeding back an identification result to the paper money device in real time.
The paper currency recognition device recognizes paper currency through the exceptional recognition rule trained by the model trainer through data fed back by the data acquisition system and the operation system.
The paper money device is in data connection with the data acquisition system, and the data acquisition system is used for acquiring the data of the paper money recognized by the paper money device and sending the data to the model trainer.
The data acquisition system acquires the data of the paper money recognized by the paper money recognizer, wherein the data comprises the serial number of the paper money recognizer, the serial number of the paper money, recognition time, the acquired paper money information (such as fluorescence information, magnetic information, infrared penetration information, laser information and anti-pinching information, the acquired information depends on the detection principle of the paper money recognizer), recognition results (true and false, denomination and country) and front and back pictures (optional) of the paper money.
And the operation system is used for marking the reported identification data or inputting the wrong index of the paper money by an operator and sending the index information data to the model trainer. The operator records the error indexes of the paper money, including a fake-money blacklist (the blacklist is used for vivid fake-money identification and comprises denomination, issuing time and serial number) and error identification label data (paper money device serial number and paper money serial number).
And the model trainer performs paper money model training according to the data provided by the data acquisition system and the operation system to generate a new paper money recognition model and sends the data of the new paper money recognition device to the paper money recognition device and the paper money device. Through AI cloud, real-time update paper currency identification rule, false paper money blacklist improve the discernment degree of accuracy of paper currency ware, reduce fortune dimension cost.
The model trainer forms different characteristics according to different positions of paper money based on collected data, and then learns by using a corresponding machine learning algorithm, such as Bayesian classification, decision tree and the like.
Reducing unidentified banknotes: the collected paper money pictures and the paper money information are labeled by operators, and the machine learning is utilized to become a new identification model.
And distributing a fake banknote blacklist recorded by the operator.
Based on unsupervised clustering learning, feature clustering is carried out on the paper money data reported in real time, clustering abnormity of some exceptional feature (key) features is alarmed in real time, operators are informed to observe in real time, and a large amount of new counterfeit paper money is prevented from being identified.
As shown in fig. 2, the flow illustrates:
1. the paper money device collects paper money data according to a hardware device of the paper money device, identifies the paper money data, and directly identifies the paper money if a model owned by the paper money device can identify the paper money; if the identification is not possible, the bill identifying device is requested to carry out further real-time identification.
2. The paper currency device obtains the recognition result and further acts as follows: and (4) paper money spitting (rejecting) and paper money swallowing (receiving).
3. The paper money machine is used for timing (every 5 minutes), and the paper money identification data which are accumulated and not reported currently are reported to the data acquisition system.
4. The data acquisition system carries out real-time cluster analysis (unsupervised) to find abnormality, and carries out real-time statistics and report display on unidentified paper money.
In conclusion, the method has the advantages that the data of the paper money is collected in real time, the model training is carried out according to the label data, the recognition accuracy of the paper money is improved, and meanwhile, the accuracy of the unknown paper money recognition is improved through real-time model judgment on the difficult paper money; the new counterfeit money is rejected in a wind control mode in time through a real-time counterfeit money blacklist, so that loss is reduced; through model distribution, new paper money identification models and machine learning models are distributed in real time, the operation and maintenance cost is reduced, and the updating speed of paper money identification rules is increased.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. An AI cloud system of paper currency ware discernment which characterized in that: the paper money recognition system comprises an AI cloud platform and at least one paper money device which is in network connection with the AI cloud platform, wherein the AI cloud platform comprises a paper money recognition device, a data acquisition system, a model trainer and an operation system, the data input end of the model trainer is respectively connected to the data acquisition system and the operation system, and the data output end of the model trainer is respectively connected with the paper money recognition device and the paper money device;
the paper money recognition device is in data connection with the paper money device, and is used for recognizing paper money to be detected by the paper money device in real time or recognizing difficult paper money and feeding back a recognition result to the paper money device in real time; the paper money identification is completed by a paper money device and a paper money identification device together, the paper money device collects paper money data according to a hardware device of the paper money device and identifies the paper money data, and if a model owned by the paper money device can identify the paper money, the paper money device directly identifies the paper money; if the paper money can not be identified, the paper money identification device is requested to carry out further real-time identification;
the paper money device is in data connection with the data acquisition system, and the data acquisition system is used for acquiring the data of paper money recognized by the paper money device and sending the data to the model trainer;
the operation system is used for marking labels on the reported identification data or inputting wrong indexes of paper money by operators and sending information data of the indexes to the model trainer;
the model trainer performs paper money model training according to the data provided by the data acquisition system and the operation system to generate a new paper money recognition model and sends the data of the new paper money recognition device to the paper money recognition device and the paper money device; the model trainer forms different characteristics according to different positions of paper money based on the acquired data and then utilizes a corresponding machine learning algorithm to learn; the machine learning algorithm comprises Bayesian classification and decision tree.
2. The AI cloud system for banknote validator identification of claim 1, wherein: the data acquisition system acquires the data of the paper money recognized by the paper money recognizer, wherein the data comprises the serial number of the paper money recognizer, the serial number of the paper money, recognition time, the acquired paper money information and a recognition result.
3. The AI cloud system for banknote validator identification of claim 2, wherein: the collected data also comprises paper money front and back pictures.
4. The AI cloud system for banknote validator identification of claim 1, wherein: the operator enters the error indexes of the paper money, including a fake paper money blacklist and error identification label data.
5. The AI cloud system for banknote validator identification of claim 1, wherein: the paper money device has a paper money identification function, and collects paper money data through a hardware device and identifies the paper money data.
6. The AI cloud system for banknote validator identification of claim 1, wherein: and the paper money device carries out paper money spitting or paper money swallowing according to the identification result.
7. The AI cloud system for banknote validator identification of claim 1, wherein: and the paper money device reports the unreported paper money identification data to the data acquisition system at regular time.
8. The AI cloud system for banknote validator identification of claim 7, wherein: the time reported by the paper money device at regular time is 5 minutes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810949973.8A CN109308760B (en) | 2018-08-20 | 2018-08-20 | AI cloud system of paper currency ware discernment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810949973.8A CN109308760B (en) | 2018-08-20 | 2018-08-20 | AI cloud system of paper currency ware discernment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109308760A CN109308760A (en) | 2019-02-05 |
| CN109308760B true CN109308760B (en) | 2021-06-29 |
Family
ID=65224011
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810949973.8A Expired - Fee Related CN109308760B (en) | 2018-08-20 | 2018-08-20 | AI cloud system of paper currency ware discernment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109308760B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111723815B (en) * | 2020-06-23 | 2023-06-30 | 中国工商银行股份有限公司 | Model training method, image processing device, computer system and medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1271914A (en) * | 1999-04-26 | 2000-11-01 | 罗烈尔银行机器股份有限公司 | Coin distinguish equipment |
| TW200407804A (en) * | 2002-11-01 | 2004-05-16 | Star News Network Co Ltd | Pattern identification system |
| JP2006018589A (en) * | 2004-07-01 | 2006-01-19 | Canon Inc | Operation module, cad apparatus and note entry method |
| CN101635072A (en) * | 2009-01-07 | 2010-01-27 | 高新现代智能系统股份有限公司 | Coin treatment device and coin treatment method for automatic ticketing |
| CN204856519U (en) * | 2015-07-14 | 2015-12-09 | 赵忠义 | Parcel bar code recognition device |
| CN106355740A (en) * | 2016-08-19 | 2017-01-25 | 深圳怡化电脑股份有限公司 | Currency detection information processing device and method |
| CN106408746A (en) * | 2016-08-25 | 2017-02-15 | 深圳怡化电脑股份有限公司 | Safety thread identification method and apparatus |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7997484B2 (en) * | 2006-09-13 | 2011-08-16 | Crane Merchandising Systems, Inc. | Rich content management and display for use in remote field assets |
| CN101206743A (en) * | 2006-12-20 | 2008-06-25 | 金瓯 | Method for monitoring and keeping away counterfeit money based on vendor cluster network |
| EP3179454B1 (en) * | 2008-07-28 | 2019-01-02 | Glory Ltd. | Banknote handling system |
| CN102473328B (en) * | 2009-07-06 | 2015-08-12 | 光荣株式会社 | The program installation system of money discriminator and program installation method |
| JP5602674B2 (en) * | 2011-05-16 | 2014-10-08 | 日立オムロンターミナルソリューションズ株式会社 | Paper sheet processing apparatus and identification algorithm update program |
| CN104620288A (en) * | 2012-05-22 | 2015-05-13 | 博彩伙伴国际有限公司 | Total money management system |
| JP6119374B2 (en) * | 2013-03-29 | 2017-04-26 | 沖電気工業株式会社 | Medium processing apparatus and medium transaction apparatus |
| KR101452330B1 (en) * | 2013-06-21 | 2014-10-22 | 기산전자 주식회사 | Apparatus for optimizing bill validator using web calibration program and the method thereof |
| CN104599114B (en) * | 2015-02-13 | 2018-05-11 | 湖南金码智能设备制造有限公司 | A kind of automatically vending system and method based on fingerprint recognition Zhi Fuyu cloud services |
| CN105118152B (en) * | 2015-09-02 | 2018-01-12 | 昆山古鳌电子机械有限公司 | A kind of more new procedures of bill handling mechanism and its bank note discrimination parameter |
| CN108320390A (en) * | 2018-04-18 | 2018-07-24 | 深圳市国投创新科技有限公司 | A kind of charging pile with automatic vending function |
-
2018
- 2018-08-20 CN CN201810949973.8A patent/CN109308760B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1271914A (en) * | 1999-04-26 | 2000-11-01 | 罗烈尔银行机器股份有限公司 | Coin distinguish equipment |
| TW200407804A (en) * | 2002-11-01 | 2004-05-16 | Star News Network Co Ltd | Pattern identification system |
| JP2006018589A (en) * | 2004-07-01 | 2006-01-19 | Canon Inc | Operation module, cad apparatus and note entry method |
| CN101635072A (en) * | 2009-01-07 | 2010-01-27 | 高新现代智能系统股份有限公司 | Coin treatment device and coin treatment method for automatic ticketing |
| CN204856519U (en) * | 2015-07-14 | 2015-12-09 | 赵忠义 | Parcel bar code recognition device |
| CN106355740A (en) * | 2016-08-19 | 2017-01-25 | 深圳怡化电脑股份有限公司 | Currency detection information processing device and method |
| CN106408746A (en) * | 2016-08-25 | 2017-02-15 | 深圳怡化电脑股份有限公司 | Safety thread identification method and apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109308760A (en) | 2019-02-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105957238B (en) | A kind of paper currency management method and its system | |
| CN109389161B (en) | Garbage identification evolutionary learning method, device, system and medium based on deep learning | |
| KR101414440B1 (en) | Paper notes management system, paper notes identification apparatus, paper notes management apparatus, method for managing paper notes, and computer readable medium in which program for managing paper notes is stored | |
| CN101303779B (en) | Automatic teller machine and tracking method of paper currency | |
| CN105045780B (en) | A kind of recognition methods of invoice brief note semantic information and device | |
| CN106952419B (en) | A kind of examination paper automatic distributing recycling machine | |
| CN103761795A (en) | Method for identifying counterfeit money of paper money anti-counterfeit feature database | |
| CN104380352B (en) | Bank note managing device, paper money processing system and bill handling method | |
| KR20130050562A (en) | Method of banknote management with banknote serial number recognition in atm and the atm the method applied thereto | |
| CN101504781A (en) | Valuable document recognition method and apparatus | |
| CN107204068B (en) | A kind of paper money management system, method and Paper Currency Identification | |
| CN109308760B (en) | AI cloud system of paper currency ware discernment | |
| CN106898078B (en) | Port currency version identification method and device | |
| CN103886673A (en) | Comprehensive information analysis platform for money-counting machine | |
| CN107657764A (en) | Finance self-help terminal and deposit, continuous deposit, withdraw the money, clear method | |
| CN110309737A (en) | A kind of information processing method applied to cigarette sales counter, apparatus and system | |
| CN203849797U (en) | An information comprehensive analysis platform used for a currency count machine | |
| CN115497218A (en) | Intelligent system and method for cash sorting, bundling and article integrated storage | |
| CN107305716A (en) | A kind of banknote discriminating method and device | |
| CN201465141U (en) | Device for distinguishing and recording serial member of RMB (Renminbi) | |
| CN102810223A (en) | Paper money verification system and method thereof | |
| CN113879724B (en) | Garbage classification system based on crowdsourcing intelligence | |
| CN115271926A (en) | Financial big data automatic intelligent analysis control system and method based on cloud computing | |
| Sun et al. | The recognition of new and old banknotes based on SVM | |
| CN107133747A (en) | A kind of method for evaluating quality 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210629 |