Cheque Image and Signature Mismatch Flagging

• Head of Clearing Operations
• Inward Clearing Verification Officer
• Fraud Risk Manager
• Head of Branch Operations

In cheque truncation the drawee bank sees only images, and must confirm signature, amount, date and mandate before the return deadline. As CTS moves to continuous clearing with settlement on realisation, that window gets shorter, while officers still compare signatures by eye against scanned specimen cards.

We build a verification assistant for inward clearing. For each cheque image it reads the MICR band, date, payee and the amount in figures and words, and locates the signature area. A signature comparison model scores similarity against the specimens in your signature repository, and a rules layer checks the operating mandate, account status, stop-payment instructions, cheque series and any Positive Pay details the customer submitted. Cheques are ranked by combined risk and value, and the verification officer sees the cheque, the specimens and every flag side by side. The officer decides to pass or return every item; the assistant never returns a cheque on its own.

• Reads MICR, date, payee and amounts in figures and words from CTS images, including handwriting
• Signature similarity scoring against one or more specimens per signatory
• Mandate checks for joint, either-or-survivor and corporate signatory rules
• Positive Pay, stop-payment, cheque series and account status checks
• Detection of alterations, overwriting and amount-in-words mismatches
• Risk-ranked queue combining value, flags and account context

1. 01Receive inward imagesInward clearing images and data arrive from the CTS interface into the bank's inward processing system.
2. 02Read the chequeOCR and handwriting recognition read the date, payee, amounts and MICR fields, and the signature region is located.
3. 03Compare and checkSignatures are scored against specimens, and mandate, Positive Pay, stop-payment and account status checks run.
4. 04Rank by riskEach cheque gets a risk band with reasons, combining flags, value and account context.
5. 05Officer verificationOfficers review high-risk items first with cheque and specimens side by side, and record pass or return with a reason code.
6. 06Return and learnDecisions flow back to the inward system, and confirmed mismatches feed periodic threshold reviews.

1. Intake
   • Inward clearing image and data feed
   • Signature repository and account master lookup
2. Image understanding
   • Image quality check
   • Field OCR and handwriting recognition
   • Signature region detection
3. Verification models
   • Signature similarity model
   • Alteration and overwriting detector
   • Amount words-figures consistency check
4. Rules and ranking
   • Mandate and account status rules
   • Positive Pay and stop-payment matcher
   • Risk ranking with reason codes
5. Officer workspace
   • Side-by-side verification screen
   • Pass or return with reason codes
   • Audit log

Verification officers make every pass or return decision, with high-risk items shown first. Signature similarity is a decision aid; the bank's signature verification policy and the officer's judgement prevail.

• Inward clearing or CTS processing system
• Core banking for account status, mandates and cheque series
• Signature verification repository
• Positive Pay data from the bank's channels
• Fraud case management system

• No automatic returns; officers decide every cheque
• Return reason codes always chosen by the officer
• Thresholds tuned per customer segment and reviewed with fraud risk
• Images and account data stay inside the bank's environment
• Score drift monitored when specimen quality or scanner sources change

• Works within the bank's CTS procedures and RBI's Positive Pay arrangements; the drawee bank's verification responsibility stays with its officers.
• Supports RBI IT governance expectations by processing images on bank-controlled infrastructure.

• Banking

  Account Opening Document Verification

  Check OVDs, PAN, masked Aadhaar, photographs and entity documents in an account opening file against each other and against CKYC, with mismatches routed to an officer.

• Banking

  Customer Complaint Triage and Routing

  Classify complaints from email, branches, social media and RBI CMS, spot urgent cases such as unauthorised transactions, route them to resolver teams and draft acknowledgements.

• Banking

  Branch Queue and Footfall Analytics

  Turn existing branch CCTV into anonymous counts of footfall, queue length and counter wait times, so branch and regional managers can plan staffing and act on long queues.

• Fintech & Payments

  Selfie-to-ID Face Match and Document Tamper Checks

  Compare a live selfie with the photo on an ID, check liveness and spot edited or replayed ID images, sending borderline cases to a reviewer with highlighted evidence.