Method and system for payor-to-depositor cheque verification

Abstract

A method and system for verifying cheques in a financial transaction network that facilitates payor participation through real-time notifications and confirmation, to 5 reduce fraud risks and processing delays. The method includes receiving a preregistration submission from a payor containing cheque images and key details, receiving a deposited cheque submission from a depositing bank, and extracting cheque details using a vision and language model. The extracted details are validated against pre-registered data to determine duplicate cashing attempts, 10 fraud indicators, and unusual activity, generating a suspicion score. The system transmits validation results and alerts to the cheque issuing bank, enabling realtime payor confirmation. The system includes modules for cheque pre-registration, cheque data processing, fraud validation, and secure communication.

Description

- 1 - METHOD AND SYSTEM FOR PAYOR-TO-DEPOSITOR CHEQUE VERIFICATION FIELD OF THE INVENTION 5 The present invention relates generally to a cheque-clearing process. More particularly, the present invention relates to a method and system for verifying cheques in a cheque clearing process involving a network that connects payors, cheque issuing banks, and depositing banks. BACKGROUND OF THE INVENTION 10 Cheque-based transactions remain a significant component of financial networks, despite the increasing adoption of electronic payment methods. However, cheque fraud and unauthorized cashing present persistent challenges for banks and financial institutions. Traditional cheque verification methods rely on post-deposit analysis, often resulting in delayed fraud detection and financial losses. 15 Current cheque processing systems predominantly rely on sophisticated fraud detection techniques at the depositing bank level, employing state-of-the-art artificial intelligence and machine learning algorithms. However, these systems do not involve the payor in the verification process. Despite their advanced fraud detection capabilities, they exclude payor participation, limiting their overall 20 effectiveness. Fraud detection algorithms operating solely at the depositor bank level may overlook discrepancies that only the payor can confirm. This limitation reduces the system’s overall reliability, as potential errors or fraudulent alterations may go undetected until later stages. Additionally, without payor confirmation, banks must 25 rely on costly, reactive measures to address returned or disputed cheques. The expenses associated with resolving these issues, including manual intervention and prolonged investigations, are significantly higher than if real-time confirmation- 2 - were integrated into the process. Furthermore, the lack of immediate payor involvement often leads to substantial delays in cheque processing, particularly when resolving disputes related to returned cheques. The inability to swiftly verify the legitimacy of a cheque 5 increases the turnaround time for resolving these disputes, hindering timely customer service and increasing the operational burden on financial institutions. Existing cheque verification systems primarily focus on signature authentication, historical transaction patterns, and manual intervention, which introduce inefficiencies and processing delays. Real-time payor verification is generally 10 absent, leading to fraudulent transactions that are only detected after funds have been transferred. Accordingly, there remains a need for improvements in the field of cheque processing. SUMMARY OF THE INVENTION 15 What is desired is a cheque-clearing process that involves the payor in the verification process. Accordingly, a method and system for effecting the cheque-clearing process are provided to facilitate payor participation, including real-time notifications and the ability for the payor to confirm or deny the validity of an issued cheque. 20 The method and system actively involve the payor in the verification process. In contrast to conventional systems that rely solely on fraud detection algorithms at the depositing bank level, the present invention introduces real-time payor participation, ensuring that the payor is notified upon cheque processing and can confirm or deny the transaction. Advantageously, such real-time engagement will 25 help minimize processing delays, reduce operational costs, and enhance fraud detection effectiveness.- 3 - According to one aspect of the present invention, there is disclosed a method for verifying cheques in a financial transaction network, the method comprising the steps of: receiving, via a first API, a pre-registration submission from a payor 5 including high-resolution images of a cheque and cheque details including payor account information, depositor name, cheque amount, and cheque date; receiving, via a second API, a cheque image and depositor account information from a depositing bank; processing the received cheque image using a vision and language model 10 to extract details from the received cheque and compare the received cheque details to the pre-registered cheque details; validating the received cheque by determining whether the received cheque has been previously cashed, matches any records of fraudulent or stolen cheques, or exhibits unusual activity associated with the payor or depositor, and 15 generating a suspicion score based on the validating step; transmitting the suspicion score and a validation status to the cheque issuing bank; and facilitating real-time notification and confirmation procedures between the cheque issuing bank and the payor. 20 According to another aspect of the present invention, there is disclosed a system comprising: a registration module configured to receive and store pre-registration data including cheque images and associated details; an API module for receiving cheque images and depositor information from 25 depositing banks; an image processing module utilizing a vision and language model to extract cheque details; a validation module configured to receive and compare the extracted cheque details with the pre-registered data and compute a suspicion score; and 30 a communication module configured to transmit cheque validation results- 4 - and suspicion scores to the cheque issuing bank for real-time notification and confirmation. BRIEF DESCRIPTION OF THE DRAWINGS Reference will now be made to the preferred embodiments of the present invention 5 with reference, by way of example only, to the following drawings in which: FIG. 1 shows schematically interactions between a cheque payor, a cheque issuing bank, a cheque depository bank, and a cheque verification network, within a financial transaction network, for cheque verification according to an embodiment of the present invention; 10 FIG. 2 shows schematically optional steps of stage 1 involving interactions between the cheque payor, the check issuing bank, and the check verification network of FIG. 1, for cheque pre-registration; FIG. 3 shows schematically a portion of FIG. 1 detailing steps of stage 2 involving in a cheque depositor bank submitting a received cheque into the cheque 15 verification network; FIG. 4 schematically illustrates a portion of FIG. 3 detailing stage 3 verification, in which the cheque payor’s pre-registered cheque data is utilized, if available, and a notification is sent to the cheque payor for early-stage anomaly detection; and 20 FIG. 5 schematically illustrates a portion of FIG. 1 detailing stage 4 verification and cheque processing, in which the cheque payor can confirm or dispute the cheque's validity, before the cheque is honoured or returned. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS 25 The present invention is described in more detail with reference to exemplary embodiments thereof as shown in the appended drawings. While the present invention is described below including preferred embodiments, it should be understood that the present invention is not limited thereto. In the figures, like- 5 - elements are given like reference numbers. For the purposes of clarity, not every component is labelled in every figure, nor is every component of each embodiment of the invention shown where illustration is not necessary to allow persons skilled in the art to understand the invention. 5 FIG. 1 schematically shows a method for verifying cheques in a financial transaction network, according to an embodiment of the present invention. Preferably, the method may proceed in four stages involving interactions between a cheque payor 10, a cheque issuing bank 12, a cheque depository bank 14, and a cheque verification network 16, within the financial transaction network. Stage 1 10 is an optional cheque pre-registration stage, in which the cheque payor 10 submits details of a cheque to the cheque verification network 16 through a series of steps, as indicated by arrow 1. In stage 2, the cheque depository bank 14 submits details of a cheque received to the cheque verification network 16 through a series of steps, as indicated by arrow 2. In stage 3, the cheque verification network 16 15 verifies the received cheque and assigns a suspicion score through a series of steps, as indicated by arrow 3. In stage 4, the cheque verification network 16 facilitates the honouring (i.e. acceptance) or return (i.e. rejection) of the received cheque through a series of steps, as indicated by arrow 4. The optional cheque pre-registration stage 1 allows a cheque payor 10 to pre- 20 register a cheque before sending it to the cheque depositor 18 for deposit. The cheque payor 10 may use a bank mobile application, an online banking interface, or integrated cheque-issuing software. The optional cheque pre-registration stage 1 includes capturing high-resolution images of the front and back of the cheque, along with cheque details such as the cheque depositor’s 18 name, the cheque 25 amount, and date. The cheque details are submitted to the cheque verification network 16 through an application programming interface (API) provided by the cheque verification network 16. The cheque verification network 16 uses vision and language models to extract the cheque details from the high-resolution images. The cheque payor 10 may be provided with incentives for pre-registering- 6 - a cheque such as reduced processing fees, and fraud loss mitigation. The optional cheque pre-registration stage 1 improves cheque verification capabilities in later stages, as discussed in more detail below. Stage 2 employs an API, provided by the cheque verification network 16, that 5 enables the cheque depository bank 14 to capture and submit high-resolution images of the front and back of the received cheque, along with the cheque depositor’s account information, to the cheque verification network 16. The cheque verification network 16 uses vision and language models to extract the received cheque details from the high-resolution images and compare them to the 10 pre-registration cheque details, if available. In stage 3, the cheque verification network 16 validates the received cheque by comparing the received cheque details against pre-registration cheque details stored in its internal pre-registered cheque database 24. It determines whether the received cheque has been previously cashed, compares it against records of 15 stolen or dark web-sourced cheque data, and evaluates any unusual activity associated with the cheque payor 10 or the cheque depositor 18. The cheque verification network 16 may preferably generate a suspicion score using the validation results and send it to the cheque issuing bank 12 via an API, facilitating real-time notifications and enabling additional verification measures if necessary. 20 Based on the validation results and / or the suspicion score, the cheque issuing bank 12 may contact the cheque payor 10 to confirm the legitimacy of the received cheque, before deciding to honour or return the received cheque. In stage 4, the cheque verification network 16 facilitates obtaining payment confirmation from the cheque payor 10 within a predefined response window. If 25 the cheque payor 10 denies the legitimacy of the received cheque or the cheque verification network 16 detects a mismatch, the cheque verification network 16 will initiate steps to return the received cheque. To ensure maximum data privacy and security, all data may preferably be stored in a one-way hashed format.- 7 - Referring now to FIG. 2, there is shown a portion of the method for verifying cheques in a financial transaction network, specifically illustrating the optional preregistration stage 1 for clarity and ease of discussion. As can be seen, the cheque payor 10 initiates the cheque pre-registration stage at step 20, which includes 5 uploading a cheque image, a cheque amount, and cheque depositor 18 (i.e. payee) information, which is collected and sent to the cheque issuing bank 12. Then in step 22, the cheque issuing bank 12 sends the pre-registration cheque details to the cheque verification network 16. An API provided by the cheque verification network 16 is used to allow the cheque issuing bank 12 to receive the 10 pre-registration cheque details, and to record and store them, before relaying them to the cheque verification network 16. The cheque verification network stores the pre-registration cheque details in a pre-registered cheque database 24. Referring now to FIG. 3, there is shown a portion of the method for verifying cheques in a financial transaction network, specifically illustrating the depository 15 bank submission stage 2 for clarity and ease of discussion. As can be seen, the cheque depositor 18, after receiving the cheque from the cheque payer 10, interacts with an API in step 26 to send an image of the received cheque to the depository bank 14. In step 28, the depository bank 14 uses the API to send the image of the received cheque along with the account details of the cheque 20 depositor 18 to the cheque verification network 16. Referring now to FIG. 4, there is shown a portion of the method for verifying cheques in a financial transaction network, specifically illustrating the cheque validation and suspicion scoring stage 3 for clarity and ease of discussion. In step 30, the cheque verification network 16 uses a vision and language model to 25 analyze the cheque image to extract details of the received cheque, such as, for example, the cheque amount, date, signature, and the like. In step 32 the received cheque details are compared to any pre-registration cheque details that may be stored in the pre-registered cheque database 24. In step 34, the cheque verification network 16 may analyze the behaviour of the cheque payor 10 and- 8 - cheque depositor 18 to identify anomalies or suspicious activity indicative of potential fraud. For example, the cheque verification network 16 may aggregate data on the activities of the cheque payor 10 and the cheque depositor 18 and use it to generate a suspicion score. In step 36, the cheque verification network 16 5 may send the suspicion score to the cheque issuing bank 12 via an API, enabling real-time notifications and facilitating additional verification measures if necessary. In step 38, based on the validation results and / or the suspicion score, the cheque issuing bank 12 may contact the cheque payor 10 to request additional information or confirmation regarding the legitimacy of the received cheque. It is contemplated 10 that cheque issuing banks 12 can apply additional verification measures if the suspicion score exceeds a preset threshold. Furthermore, immediate alerts may be generated for discrepancies, potentially prompting the need for direct communication with the cheque payor 10 or cheque depositor 18. Referring now to FIG. 5, there is shown a portion of the method for verifying 15 cheques in a financial transaction, specifically illustrating stage 4 verification and cheque processing, in which the cheque payor can confirm or dispute the legitimacy of the received cheque, within a window of time, before the received cheque is either honoured or returned to the cheque depositor 18 by the cheque depository bank 14. In step 40, the cheque payor 10 provides a response 20 confirming or denying the legitimacy of the received cheque, and the cheque payor’s 10 response is sent to the cheque issuing bank 12. In step 42, the cheque issuing bank 12 relays the cheque payor’s 10 response to the cheque verification network 16. If the cheque payor 10 denies the legitimacy of the received cheque, or if inconsistencies persist, the cheque verification network 16 may initiate a rapid 25 cheque return process in step 44 to mitigate potential fraud and reduce processing delays. Preferably, all transaction and verification data is stored using a one-way hashing algorithm, ensuring data integrity and privacy. As will be appreciated, embodiments of the method for verifying cheques in a financial transaction are implemented using a system interconnected computers,- 9 - servers, and data storage devices, which communicate via wired and / or wireless connections. Portions or the entirety of the system may be cloud-based. Preferably, the cheque payor 10 and the cheque depositor 18 interact with the system via the Internet. 5 While reference has been made to various preferred embodiments of the invention, other variations, implementations, modifications, alterations and embodiments are comprehended by the broad scope of the appended claims. Some of these have been discussed in detail in this specification and others will be apparent to those skilled in the art. Those of ordinary skill in the art having access 10 to the teachings herein will recognize these additional variations, implementations, modifications, alterations and embodiments, all of which are within the scope of the present invention, which invention is limited only by the appended claims.

Claims

- 10 - THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A method for verifying cheques in a financial transaction network, the 5 method comprising the steps of: a) receiving, via a first API, a pre-registration submission from a payor including high-resolution images of a cheque and cheque details including payor account information, depositor name, cheque amount, and cheque date; b) receiving, via a second API, a cheque image and depositor account 10 information from a depositing bank; c) processing the received cheque image using a vision and language model to extract details from the received cheque and compare the received cheque details to the pre-registered cheque details; d) validating the received cheque by determining whether the received 15 cheque has been previously cashed, matches any records of fraudulent or stolen cheques, or exhibits unusual activity associated with the payor or depositor, and e) generating a suspicion score based on the validating step; f) transmitting the suspicion score and a validation status to the cheque issuing bank; and 20 g) facilitating real-time notification and confirmation procedures between the cheque issuing bank and the payor.

2. The method of claim 1, wherein the pre-registration in step (a) is performed via a mobile banking application, an online banking interface, or integrated cheque issuing software. 25 3. The method of claim 1, wherein the vision and language model in step (c) extracts cheque details by analyzing high-resolution images to determine cheque amount, date, and other pertinent information.

4. The method of claim 1, further comprising the step of incentivizing the payor- 11 - to pre-register the cheque by offering reduced cheque processing fees and fraud loss mitigation benefits.

5. The method of claim 1, wherein the validation step (d) further comprises cross-referencing the cheque against a database of dark web-stolen cheques and 5 monitoring unusual depositor or payor activity patterns.

6. The method of claim 1, wherein the network facilitates a confirmation stage in which the payor is given a predetermined period to confirm or deny the cheque payment, and upon denial, initiates a cheque return process.

7. The method of claim 1, wherein all cheque and transaction data is stored 10 using a one-way hashing algorithm to ensure data integrity and privacy.

8. A system for verifying cheques in a financial transaction network, the system comprising: a registration module configured to receive and store pre-registration data including cheque images and associated details; 15 an API module for receiving cheque images and depositor information from depositing banks; an image processing module utilizing a vision and language model to extract cheque details; a validation module configured to receive and compare the extracted 20 cheque details with the pre-registered data and compute a suspicion score; and a communication module configured to transmit cheque validation results and suspicion scores to the cheque issuing bank for real-time notification and confirmation.