CN104143236A - Cloud computing service platform implementation method based on video stream recognition technology - Google Patents

Cloud computing service platform implementation method based on video stream recognition technology Download PDF

Info

Publication number
CN104143236A
CN104143236A CN201310164610.0A CN201310164610A CN104143236A CN 104143236 A CN104143236 A CN 104143236A CN 201310164610 A CN201310164610 A CN 201310164610A CN 104143236 A CN104143236 A CN 104143236A
Authority
CN
China
Prior art keywords
cloud computing
service platform
user
computing service
identification
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201310164610.0A
Other languages
Chinese (zh)
Inventor
金培新
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CN201310164610.0A priority Critical patent/CN104143236A/en
Publication of CN104143236A publication Critical patent/CN104143236A/en
Pending legal-status Critical Current

Links

Landscapes

  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

The invention relates to a method for recognizing streaming media to realize online payment by the cloud computing service platforms technology. The method is characterized in that a user can obviously lower down the application and development cost on the basis of the cloud computing service platform and an application product as well as ensure the safety of the user information in online payment. The application platform provided by the embodiment comprises a streaming media recognition computing server, a user data storing server and an application server; the streaming media recognition computing server is used for recognizing the surface of a credit card and processing the credit card number; the user data storing server is used for storing and managing the user information; the application server is used for transferring relation application to run in the application server according to the user access requirements.

Description

A kind of cloud computing service platform implementation method based on video flowing recognition technology
Technical field
The present invention relates to cloud computing, Signal and Information Processing, data encryption, data encoding, data show, data transmission, image recognition, data storage, and mobile phone and payment technical field, the technology that especially belongs to the technology based on image recognition and realize on-line payment based on image recognition.
Background technology
On-line payment refers to based on internet, mobile Internet, completes a kind of payment transaction mode of the serial processes such as online monetary payoff from terminal user to businessman of enterprise, fund clearing, query statistic.Business moves one of maximum obstacle of facing of application order, allows exactly the user be really a commodity payment.Using credit card or bank card to carry out in the process of on-line payment, the first step is exactly that traditional approach is generally manual input card number to the obtaining of card number; Manual input mode, needs to input continuously the numeral of 16~18, and for most users, this process is very unfriendly, and error probability is very high, causes sales payment termination of procedure.And account information is easily intercepted in transmitting procedure, by lawless person is utilized, carry out pseudo-blocking and steal account fund.A kind of cloud service platform based on video flowing recognition technology, realize the method for on-line payment and apply efficient filter characteristic extraction algorithm and rational Classing Method of on-Line, and these complex calculations process migrations have been arrived to high in the clouds completed, and user only need to obtain Media Stream image by normal domestic use computer camera or mobile phone camera, synchronously identification is also encrypted in point-to-point transmission process.Fast, conveniently, non-contacting mode accurately identifies credit card, bank card number, and then completes on-line payment safely, efficiently.Adopt cloud computing service platform can significantly reduce user's input cost, for user provides fast, the service of video flowing identification easily.
Summary of the invention
The object of the invention is to: adopt pure software mode, by family expenses camera, the first-class multimedia equipment of cell-phone camera, credit card card face is scanned, and adopt efficient character recognition technologies, automatic quick obtain under credit card number, credit card the credit card card face essential informations such as bank, and then complete payment.And these complicated mathematical operations are moved to high in the clouds complete, user profile is rationally stored, significantly reduce client's input cost, can provide real-time on-line payment service for user.
The technical scheme that the present invention takes is:
Gordian technique 1: adopt the mathematical algorithm of optimizing to ensure the real-time of software identification
In whole identifying, edge detection process belongs to intensive link; Therefore adopting the mathematical algorithm of optimizing to carry out calculated amount optimization, is the key that guarantees identification real-time; For special substance and the texture situation of credit card, I.D., we have carried out large data sample analysis to different edge detection operators; Comparison for Roberts boundary operator, Sobel boundary operator, Prewitt boundary operator, Laplacian boundary operator, Canny boundary operator scheduling algorithm, Canny operator edge detection algorithm after final selected optimization, its algorithm can elimination noise has the optimal selection method in the single order differential algorithm of rim detection that can keep the edge information characteristic; Optimize the algorithm having and can adaptively to detecting the discrimination threshold of operator, carry out analysis and regulation; Thereby effectively reduce calculated amount, guarantee the real-time detecting.
Gordian technique 2: accurately controlled identification process guarantees the accuracy of software identification
The accuracy rate of identification just depends on design and the management and control of whole identification process; Our system platform has been carried out more meticulous planning and design to identification process; Design cycle comprises intercepting video frames (VideoFrameShot), rim detection (EdgeDetection), image normalization pre-service (Normalization), card surface analysis cutting (Analysis), feature extraction (FeatureExtraction), character recognition (Recognition), aftertreatment (PostProcessing), the links such as semantic comparative analysis; For guaranteeing the accuracy of overall flow, we are optimized processing to each the subring joint in flow process, thereby controlled to whole processing and identification flow process; Large data sample is tested repeatedly, and whole recognition accuracy has reached 95%.And in Continuous optimization.
Gordian technique 3: improved minimum distance classifier recognizer is guaranteed software identification fault-tolerance
Due to the scene environment more complicated of the actual use of software, so the fault-tolerance of software algorithm is a key factor; And the link that affects fault-tolerance maximum is the identification of character; Consider that character recognition is mainly English and digital, minimum distance classifier is the sorting algorithm based on vector space model, in image recognition processing and text classification field, has unique advantage; In order to adapt to actual use scenes, we have taked a kind of new disaggregated model weighted minimum distance classifier (Weighted Minimum Distance Classifier) based on standardization Euclidean distance.
Gordian technique 4: for " cloud computing service platform " architecture design and the exploitation that customizes user
Consideration provides different differentiated services for different user's requests, computer cloud computing technology, our architecture design " video flowing identification cloud computing service platform (VRCSP) "; By using cloud computing service platform, can significantly reduce user's input cost, for user provides fast, the service of video flowing identification easily (as: credit card identification service, recognition of face service, I.D. identification serve etc.).
Embodiment
Based on a video flowing recognition technology cloud computing service platform, adopt credit card, bank card recognition technology, in conjunction with cloud computing platform, complete online transaction and pay; Consumer carries out the scanning of card face by camera (as: embedded in mobile phone camera, computer camera etc.) to credit card, the bank card of needs consumption, video flowing intercepting 3~4 pictures of scanning, software is by the video pictures of intercepting, in conjunction with user basic information packing, by Internet Transmission, arrive cloud computing center; The video pictures that cloud computing center gets performs mathematical calculations, image is processed, image recognition; The credit card number that final extraction consumer will pay, or bank card number; Cloud computing center completes the record of consumer information simultaneously; Credit card, bank card number and substantially block surface information and confirm after, transfer to banking data base; Banking data base verifies consumer's identity, as: short-message verification coding mode etc.; Consumer's identity and Transaction Information complete this transaction after confirming; Cloud computing platform records this consumption information.

Claims (10)

1. the cloud computing service platform based on video flowing recognition technology, is characterized in that, described cloud computing service platform comprises Streaming Media identification calculation server; Described Streaming Media identification calculation server, calculate in the identification and the space that mainly complete image, can complete the card face identification to credit card, and the card face identification of ordinary bank's card; And according to different user access request, call different computing applications, all or part of extracts identifying processing to credit card information.
2. user's storage server according to claim 1, it is characterized in that, described cloud computing service platform comprises storage of subscriber data server, described storage of subscriber data server, main completing user information is stored in server, and according to different user access request, call relevant user data, and these user data are reasonably returned.Storage of subscriber data server also comprises user's verify data, and user authentication data is verified user identity, and the user access request by checking is forwarded to application server.
3. cloud computing service platform according to claim 1, it is characterized in that, by using cloud computing service platform, can significantly reduce user's input cost, for user provides fast, video flowing identification service easily, including, but not limited to: credit card identification service, recognition of face service, I.D. identification are served.
4. cloud computing service platform according to claim 3, Streaming Media identification service is characterized in that, the storage mode of Streaming Media identification data, the storage mode that the data after user terminal scanning are carried out, employing rectangular configuration body surface mode is stored.
5. according to the cloud computing service platform described in claim Isosorbide-5-Nitrae, Streaming Media identification service is characterized in that, the Feature Extraction Technology of special algorithm, the Feature Extraction Technology based on similarity vector algorithm.
6. according to claim Isosorbide-5-Nitrae, the cloud computing service platform described in 5, Streaming Media identification service is characterized in that, the character recognition technologies of special algorithm, weighted minimum distance distance classification device recognizer is applied to the character recognition of credit card.
7. according to claim Isosorbide-5-Nitrae, the cloud computing service platform described in 5,6, Streaming Media identification service is characterized in that, the credit card character recognition flow process completing, the specific flow process of image recognition, pre-service, rim detection, feature extraction, character recognition.
8. according to claim 1, cloud computing service platform described in 2, it is characterized in that, the terminal device of access cloud computing service platform, it can be mobile terminal device, as: mobile phone, panel computer can be fixed terminal equipments, as: ATM, desktop computer, notebook computer etc. have the terminal device of payment function.
9. according to claim 1, cloud computing service platform described in 8, it is characterized in that, the outside payment and settlement system using, it can be the payment and settlement system of bank or Unionpay, can be Credit Card Payments settlement system, can be also the payment and settlement system that third party's payment company provides, or the payment and settlement system of prepaid card etc.
10. according to claim 1,2,3,4,5, the cloud computing service platform described in 6, is characterized in that, adopts pure software mode to complete a whole set of identification process, software realization mode can be assembly language, can be C language etc.
CN201310164610.0A 2013-05-06 2013-05-06 Cloud computing service platform implementation method based on video stream recognition technology Pending CN104143236A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310164610.0A CN104143236A (en) 2013-05-06 2013-05-06 Cloud computing service platform implementation method based on video stream recognition technology

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310164610.0A CN104143236A (en) 2013-05-06 2013-05-06 Cloud computing service platform implementation method based on video stream recognition technology

Publications (1)

Publication Number Publication Date
CN104143236A true CN104143236A (en) 2014-11-12

Family

ID=51852404

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310164610.0A Pending CN104143236A (en) 2013-05-06 2013-05-06 Cloud computing service platform implementation method based on video stream recognition technology

Country Status (1)

Country Link
CN (1) CN104143236A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107464119A (en) * 2017-08-18 2017-12-12 上海二三四五金融科技有限公司 A kind of automatic control method and device for obtaining identity information

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107464119A (en) * 2017-08-18 2017-12-12 上海二三四五金融科技有限公司 A kind of automatic control method and device for obtaining identity information

Similar Documents

Publication Publication Date Title
US20200027090A1 (en) Systems and methods for authenticating financial transactions
US10044843B2 (en) Mobile device top up apparatus
US10055734B2 (en) Systems and methods for processing customer purchase transactions using biometric data
US10284538B2 (en) System for processing an even request by determining a matching user profile based on user identifying information
US20220391913A1 (en) Techniques to improve fraud detection at financial terminals
US20210110359A1 (en) Dynamic virtual resource management system
US20200265440A1 (en) Transaction validation for plural account owners
US20210192522A1 (en) Intelligent fraud rules
US10713538B2 (en) System and method for learning from the images of raw data
US20220309387A1 (en) Computer-based systems for metadata-based anomaly detection and methods of use thereof
US20220012357A1 (en) Intelligent privacy and security enforcement tool for unstructured data
CN104102997A (en) Credit card streaming media identification technology and online payment realization method
CN102568125A (en) System and method for performing credit card consumption through fingerprint identification
CN111539807A (en) Transfer method and device
CN104143236A (en) Cloud computing service platform implementation method based on video stream recognition technology
CN113065908A (en) Rental method, rental device, rental platform, and storage medium
US20220245651A1 (en) Systems and methods for enhanced resource protection and automated response
US20220398330A1 (en) System for image/video authenticity verification
US11314710B2 (en) System and method for database sharding using dynamic IDs
US20220292514A1 (en) Data processing method, apparatus, and system
CN104143141A (en) Financial payment method based on credit card scanning technology
Wang Overview of Digital Finance Anti-fraud
US9047637B1 (en) Method and system for converting printed checks into pre-paid debit card funds
CN111553702A (en) Payment risk identification method and device
US20160063620A1 (en) System and method of facilitating payday loans

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20141112