WO2019194673A1

WO2019194673A1 - System and method of tracking financial transaction with facial recognition

Info

Publication number: WO2019194673A1
Application number: PCT/MY2019/000011
Authority: WO
Inventors: Siah Poi Chong JONATHAN
Original assignee: Three Logic Concepts Sdn. Bhd.
Priority date: 2018-04-06
Filing date: 2019-04-05
Publication date: 2019-10-10

Abstract

A system and method to track and evaluate anti-money laundering and counter financing terrorism risk may include capturing monetary transaction through financial institutions and attach to a person with the means of biometric facial recognition. The tracking may include transaction performed over the counter or at self-service terminal. Person tracked may be an affiliate with the establishment such as member, account holder or even stranger. Risk evaluation may be determined by single or cumulative transaction value by the person.

Description

SYSTEM AND METHOD OF TRACKING FINANCIAL TRANSACTION WITH FACIAL

RECOGNITION

FIELD OF INVENTION

The present invention generally relates to system and method for tracking customer financial transaction for purpose of identify, track and evaluate Anti Money Laundering (AML) and/or Counter Financing of Terrorism (CFT) risk. The invention utilises facial recognition method and system to track over the counter and self-service terminal financial transaction required by reporting institution.

BACKGROUND OF INVENTION

The present invention relates to financial systems, products of the like and other non-fmancial businesses designated under the AML and/or CFT as reporting institution and more particularly to a method and system to track and record potential illegal use of products or services offered for purpose of money laundering, terrorist financing or similar schemes.

With the use of financial systems or its products of the like and non-financial business by terrorists and other criminals to finance and support their illicit activities, governments around the globe have introduced guidelines and regulations to detect and monitor use of such systems. products or services.

Compliance with these governmental guidelines and regulations can impose significant burdens as there are vast methods to launder money including "structuring" and "smurfing", making it a tedious task if every transaction is required to be monitored manually for the possibility of use for illicit or illegal activity. Or worst yet, based on "best effort". An audit trail, report or completed due diligence may be required to show compliance with governmental laws and regulations or institutional directives and policies.

There may also be a need to document why certain actions were taken or not taken or justification for a certain level of scrutiny placed on various customers especially when it relies on“best effort" basis. There may also be issues with respect to only profiling particular customers so additional basis or criteria for monitoring certain customers or taking particular actions may be needed.

SUMMARY OF THE INVENTION

Accordingly, the present invention provides a system of tracking currency transaction using facial recognition, the system includes a computing device, a first processor and a second processor, wherein the computing device having an application or an unmanned device detects or receives face of a detected person, receives details of the detected person, assigns a unique identifier to an unknown detected face, receives a currency transaction request input, wherein the currency transaction request input includes the detected person's unique identifier, currency and transaction value, the first processor verifies currency transaction details and transmission of the transaction to a system server and the second processor evaluates if the detected person is subject to Suspicious Transaction Reporting (STR) and/or Currency Transaction Reporting (CTR) based on a predetermined threshold.

Further the present invention provides a method for tracking and evaluating anti-money laundering and counter financing of terrorism risk for currency transactions performed at a financial institution (FI), the method includes the steps of (a) detecting face using a face detection system when a person approaches a transaction counter, a kiosk machine or an officer of the FI,

(b) matching a detected face to a facial database utilising a facial recognition solution, (c) sending output of a matched person's details to an application installed on a computing device or via interface wherein if the details have no match, the application will generate a unique association identifier at a database and then outputs details to the application, (d) receiving a request input, the request input includes currency type and value of an intended transaction on the application or kiosk, (e) transmitting the transaction details to a system server by a processor, (f) processing the transaction by the by the system server and (g) evaluating if a person is subject to Currency Transaction Reporting (CTR) or Suspicious Transaction Reporting (STR) based on transaction history and limit configured for various reporting threshold by the by the system server. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a flow chart of an example of a method to track and evaluate AML and/or CFT risk or the like in accordance with an embodiment of the present invention;

FIG. 2 illustrates the system front-end application used to identify the person and capture transaction details in accordance with ah embodiment of the present invention;

FIG. 3 is a flow chart illustrating a method in accordance with an embodiment of the present invention; and

FIG. 4 is a high level overview of possible flow based on FIG. 1.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

In accordance with an embodiment of the present invention, a system and a method to track and trigger/evaluate anti-money laundering and terrorist financing risk which include the use of facial recognition and provisioned system for over the counter, or self-service terminal financial transaction are being provided.

In accordance with another embodiment of the present invention, a system and a method to track and evaluate anti-money laundering and terrorist financing risk which include to input cash equivalent amount for the transaction either manually, systematically or automatically are being provided. A face may be selected and associated with the person performing the transaction and risk rating may be determined based on current and past transaction as configured in the system.

In accordance with another embodiment of the present invention, a system and a method to evaluate anti-money laundering and terrorist financing which include a server are being provided. A transaction history or storage may be operable on the server to store and evaluate anti-money laundering and terrorist financing risk. In one embodiment of the present invention, a risk evaluation tool may be adapted to determine a risk rating based on transactions history. In accordance with another embodiment of the present invention, a system and a method to track and evaluate anti-money laundering and terrorist financing risk may include providing server for facial recognition and transaction storage on the server are being provided.

Broadly, the present invention provides a system of tracking currency transaction using facial recognition, the system includes a computing device, a first processor and a second processor, wherein the computing device having an application or an unmanned device detects or receives face of a detected person, receives details of the detected person, assigns a unique identifier to an unknown detected face, receives a currency transaction request input, wherein the currency transaction request input includes the detected person's unique identifier, currency and transaction value, the first processor verifies currency transaction details and transmission of the transaction to a system server and the second processor evaluates if the detected person is subject to Suspicious T ransaction Reporting (STR) and /or Currency Transaction Reporting (CTR) based on a predetermined threshold, The currency transaction request may include a picture of the detected person. The computing device may include an inbuilt camera to capture picture of the person performing the transaction. The computing device may utilize an external camera to capture picture of the person performing the transaction. The face detection may be conducted either via a software application or a hardware if a detected face does not match any existing faces, a newly generated unique identification may be provided. The currency transaction request input may include the detected person's name. The unmanned device may be an unmanned kiosk

Or terminal. The computing device may include a database for storing current request input of an intended transaction. The system server may include a database for storing details of supposedly or completed transactions including unique association identification, picture, currency, value, date time, location, printer connection and authentication information. The database may store employee identification.

Also broad, the present invention provides a method for tracking and evaluating anti-money laundering and counter financing of terrorism risk for currency transactions performed at a financial institution (FI), the method includes the steps of (a) detecting face using a face detection System when a person approaches a transaction counter, a kiosk machine or an officer of the FI, (b) matching a detected face to a facial database utilising a facial recognition solution, (c) sending output of a matched person's details to an application installed on a computing device or via interface wherein if the details have no match, the application will generate a unique association identifier at a database and then outputs details to the application, (d) receiving a request input, the request input includes currency type and value of ah intended transaction on the application or kiosk, (e) transmitting the transaction details to a system server by a processor, (f) processing the transaction by the by the system server and (g) evaluating if a person is subject to Currency

Transaction Reporting (CTR) or Suspicious Transaction Reporting (STR) based on transaction history and limit configured for various reporting threshold by the by the system server. The face detection system may be either a hardware like camera or a software application. The intended or submitted transaction may include a picture of the person intending to perform a currency transaction. The face may be detected by an external camera or built-in camera on the computing device. The detected face may be matched to a facial database. Matching of the detected face to a facial database may be done using a facial recognition solution. If the detected face does not match the facial database, the system server may generate a new unique association identifier to the detected face. The new unique association identifier may be stored in a facial database by system server for future matching. The facial database may be updated using a third party system. The transaction and person's details may be stored in a system database. The detected person’s details may be created or received from a local system server located at a premises or remotely. The transaction details may be stored in the system database. A transaction performed by a matched face may be accrued in system database. Step [g) may be performed by Web Services server and it may further include a step of retrieving past transactions if any from system database. The method may further include a step of requesting for a person's details and history from the system server with the detected face of a person who intends or has submitted transaction with the person's details received.

The following detailed description of the preferred embodiments refers to the accompanying drawings which illustrate specific embodiments of the invention. Other embodiments having different structures and operations do not deviate from the scope of the present invention.

FIG. 1 is a detailed flow chart of an example of a method to identify patron, record and evaluate

AML and/or CFT risk or risk of similar activity in accordance with an embodiment of the present invention. Money laundering or activities of similar kind may involve any conduct or acts designed in whole or in part to conceal or disguise the nature, location, source, ownership or control of money, money equivalents, financial instruments or the like to avoid any transaction reporting requirements under state or federal law of to disguise the fact that the money was acquired by illegal means or may be used for illegal means. Anti-money laundering may include any activity designed to detect, monitor or thwart money laundering activity or related activity such as that described above. AML risk or risk rating is a measure that may assist in determining if additional steps or actions may be appropriate to detect, monitor or thwart money laundering or similar or related activity.

FIG. 1 may form at least a portion of a method and system to identify and track financial transactions for the purpose of evaluating AML and/or CFT or similar risk, A person approaches counter, kiosk or terminal, the provisioned camera detects a face 100 that is captured then followed by recognition process 102. If the detected face matches 104 an existing association id in the system, the association id may be displayed 108 prior to advancing. If the detected face resulted in empty/no match 104, the system will automatically generate an ad-hoc association id for the unmatched face 106 then followed by possibly displaying association id 108. An association id may be existing unique assigned id such as but not limited to membership, banking account, national id or system generated 106.

Once association id has been identified and possibly displayed 108, an amount of the intended transaction value is required to be entered 110; if over the counter, input may be performed by counter staff attending to the transaction. And if self-service terminal, the person performing the transaction will provide the input. Input of the amount may also be electronically if interface to third party system such as but not limited to banking system, automated teller machine (ATM), cash deposit machine (CDM), currency counter or casino cage system including RFID casino currency gaming chip system. The person may opt to not proceed with the transaction by selecting Cancel 112, Upon confirmation of the input amount 110 for the requested transaction, the system will perform various checks 114 including but not limited to system configured threshold limit also transaction history if any to determine if this person or transaction should be flagged 116 for further action required. The threshold limit may be a configured total value for all transactions by number of hours or days as mandated by local law. Non-flagged 118 transaction may proceed without further action required, and flagged transaction will trigger system process to check available/additional information 120 of the person requesting the transaction. The available/additional information may be part of Know Your Customer (KYC) which may include contact details, occupation, profession, employer details which may be required when submitting Currency Transaction Reporting (CTR) 126 upon completing the transaction which will be saved and notification as configured in the system. The available information may be obtained from existing system such as banking system for account holder or casino system for existing members, If available information for the person is none or insufficient, a process to perform Customer Due

Diligence (CDD) is required 124; completing the CDD form may include personal and contact details, occupation, profession, employer details, source and purpose of funds. If person comply With completing the CDD 128, a CTR may he filed 126; but if the person does not wish to proceed with CDD, the transaction is then cancelled 130 and a Suspicious Transaction Reporting (STR) 132 may be filed. The CDD may be performed electronically using this same system which performs the identity and track of transactions

FIG. 2 illustrates the system front-end application used to process transaction details installed and operational on compatible computing device 202 such as but not limited to desktop or portable computer, or mobile device such as phone or tablet. With reference to FIG. 2, the processing input Application 204 is shown schematically upon successful User logon, as displayed on the display of the computing device by the App. Detected person's face by either external camera or computing device's built-in camera will be displayed in this area 206; and details of the person if known will be displayed accordingly such as but not limited to person's name 208 along with association unique identifier 210 such as membership, account or anonymous assigned id number. And association type 212 to easily distinguish if the person is a member (M), or anonymous (A).

If available, history of the person's transaction may be viewed from the history option button 214.

To complete submission of the transaction details, a currency 216 may be selected if multi- currency is supported followed by the intended transaction value 218 which may be manually entered with provisioned on screen numeric pad 220, or quick denomination shortcut 222; alternatively the value may also be retrieved from third party device such as but not limited to counting and/or verification solutions for currency notes, casino gaming chip or cashable vouchers.

User may proceed to Submit 224 the transaction upon completing all required details outlined above.

Current date and time 226 along with logged on User's details 228 also displayed on the Application.

Cancel option 230 is available should the User or person decides not to continue with the transaction; which may trigger a Suspicious Transaction Reporting (STR) submission with information of the cancelled transaction included such as photo of the person, association ID, amount/value and date time. FIG. 3 illustrates the system process flow for each transaction; the computing device 302 may be but not limited to desktop or portable computer, mobile and tablet has an application 304 installed and store in its memory. The Application 304 may be downloaded and installed on the computing device 302 either over the air or sideload. The computing device 302 may also be a kiosk such as Automated Teller Machine (ATM), Cash Deposit Machine (CDM) or Currency

Exchange Machine including cryptocurrency which in this case, the kiosk will interface to system server 316 instead of Application loaded. When a person's face is detected either via built-in camera 306 on the computing device or external camera 308 either wired or wireless, the application sent instructions to the Central Processing Unit (CPU) 310 of the computing device, the CPU 310 executes the program instructions of the application. This causes the processor to execute steps 104 to 108 of FIG. 1 by invoking Web Services 318 located at system server 316.

The face detection may be done at front-end application, backend server or camera hardware level.

The application 304 may access local device database 312, also web services 314 to access backend systems 316 over intranet or extranet.

The database 312 on the computing device 302 stores information such as current input and previous transaction history for ease of referencing. Person's details downloaded form in the local system server 316 maybe cached locally on the computing device database 312 for performance reasons. Ih this case, a check may be made when the Application 304 is launched.

The built-in camera 306 and external camera 308 may be controlled and accessed by the

Application 304 for face detection and possibly photograph purpose.

If the system server 316 is hosted locally, then it will be accessed via Local Area Network (LAN) or Wi-Fi. However, if the system server 316 is hosted remotely, then it will via accessed via Wide Area Network (WAN) or Extranet. There may be a number of components within the platform, and these are described below.

The system server 316 implements a Web Services 318 server for access by Application 304 on the computing device 302 using web service module. When one or more sequences of face detected instruction received by the Web Services 318 from Application 304, the Web Services

318 proceeds to execute the steps outlined in 102 to 108 of FIG. 1. Web services interface may use JSON or REST technologies and use HTTP over IP as the underlying communication method. The System Database 320 which may be located on the Web Services 318 server or hosted separately stores transaction details such as person's association ID, transaction currency and value, along with time and date, and possibly a picture of the person when performing transaction. Also stored on the System Database 320 would be completed STR, CTR details and form in electronic format. The computing device 302 and external camera 308 connection, configuration and authentication information are also stored.

FIG. 4 is a high level flow diagram of FIG. 1, an example of a method detecting face, recognize detected face, attach transaction details to detected face, store transaction, process and prompt if required to evaluate AML and/or CFT risk and prompt next action accordingly.

When a face is detected 402, an image of the detected face is captured then sent for face recognition 404 for profile retrieval from database. If recognition process does not return any result, a new profile is created in the database with unique association identifier and assigned to the detected face. The profile database may be a collection of association or membership preloaded or continuously updated with new registration or unknown face. Once profile is available, transaction details 406 such as but not limited to currency and value is required for submission and store 408 in database locally and centrally. Upon submission or cancellation, the transaction will be processed 410 to verify if the action is subjected to prompt

412 STR or CTR accordingly.

Claims

1. A system of tracking currency transaction using facial recognition, the system includes a computing device, a first processor and a second processor, wherein the computing device haying an application or an unmanned device detects or receives face of a detected person, receives details of the detected person, assigns a unique identifier to an unknown detected face, receives a currency transaction request input, wherein the currency transaction request input includes the detected person's unique identifier, currency and transaction value, the first processor verifies currency transaction details and transmission of the transaction to a system server and the second processor evaluates if the detected person is subject to Suspicious Transaction Reporting (STR) and/or

Currency Transaction Reporting (CTR) based on a predetermined threshold.

2. The system of claim 1, wherein the currency transaction request includes a picture of the detected person.

3. The system of claim 1, wherein the computing device includes an inbuilt camera to capture picture of the person performing the transaction.

4. The system of claim 1, wherein the computing device utilizes an external camera to capture picture of the person performing the transaction.

5. The system of claim 1, wherein the face detection is performed either via a software application or a hardware.

6. The system of claim 1, wherein if a detected face does not match any existing faces, a newly generated unique identification is provided.

7. The system of claim 1, wherein the currency transaction request input includes the detected person's name.

8. The system of claim 1, wherein the unmanned device is an unmanned kiosk or terminal.

9. The system of claim 1, wherein the computing device includes a database for storing current request input of an intended transaction.

10. The system of claim 1, wherein the system server includes a database for storing details of supposedly or completed transactions including unique association identification, picture, currency, value, date time, location, printer connection and authentication information.

11. The system of claim 26, Wherein the database stores employee identification.

12. A method for tracking and evaluating anti-money laundering and counter financing of terrorism risk for currency transactions performed at a financial institution (FI), the method includes the steps of:

(a) detecting face using a face detection system when a person approaches a transaction counter, a kiosk machine or an officer of the FI;

(b) matching a detected face to a facial database utilising a facial recognition solution;

(c) sending output of a matched person's details to an application installed on a computing device or via interface wherein if the details have no match, the application will generate a unique association identifier at a database and then outputs details to the application;

(d) receiving a request input, the request input includes currency type and value of an intended transaction on the application or kiosk; (e) transmitting the transaction details to a system server by a processor;

(f) processing the transaction by the by the system server; and

(g) evaluating if a person is subject to Currency Transaction Reporting (CTR) or

Suspicious Transaction Reporting (STR) based on transaction history and limit configured for various reporting threshold by the by the system server.

13. The method of claim 12, wherein the face detection system is either a hardware like camera or a software application.

14. The method of claim 12, wherein the intended or submitted transaction includes a picture of the person intending to perform a currency transaction.

15. The method of claim 12, wherein the face is detected by an external camera or built-in camera on the computing device.

16. The method of claim 12, wherein the detected face is matched to a facial database.

17. The method of claim 16, wherein matching of the detected face to a facial database is done using a facial recognition solution.

18. The method of claim 17, wherein if the detected face does not match the facial database, the system server generates a new unique association identifier to the detected face,

19. The method of claim 18, wherein the new unique association identifier is stored in a facial database by system server for future matching,

20. The method of claim 12, wherein the facial database is updated using a third party system.

21. The method of claim 12, wherein the transaction and person's details are stored in a system database.

22. The method of claim 12, wherein the detected person's details created or received from a local system server located at a premises or remotely.

23. The method of claim 12, wherein the transaction details will be stored in the system database.

24. The method of claim 12, wherein transaction performed by a matched face is accrued in system database.

25, The method of claim 12, wherein step (g) is performed by Web Services server.

26. The method of claim 25, wherein step (g) performed by Web Services server includes a step of retrieving past transactions if any from system database.

27. The method as claimed in any one of claims 12 to 25, wherein the method further includes a step of requesting for a person's details and history from the system server with the detected face of a person who intends or has submitted transaction with the person’s details received.