How a Payments Bank used Computer Vision for Fraud Detection

Head Quarters

New Delhi, India

Industry

BFSI

Company Size

10,000+

Challenge

At the time of application for FASTag issuance, vehicle registration number & vehicle Registration Copy (RC) is uploaded by the customer on the app. These details are manually verified by the firm before dispatching to their customers. The industry has been on a lookout for solutions that can help reduce their processing time and reduce costs associated with fraudulent transactions.

Outcome

AmyGB’s VisionERA Intelligent Document Processing Platform offers the ability to do away with the erroneous and time-consuming manual verification process, to delight customers with a single-click, Instant approval and activation of their fast tags. Technology such as Computer Vision, OCR/ICR, Machine Learning, Neural Linguistic Programming and API verification have been used to detect fraudulent transactions and flag them for manual intervention, while automating remaining transactions with instant approvals.

Approach

AmyGB’s Intelligent Document Processing Platform offers the ability to do away with the erroneous and time-consuming manual verification process, to delight customers with a single-click, Instant approval and activation of their fast tags. Technology such as Computer Vision, OCR/ICR, Machine Learning, Neural Linguistic Programming and API verification have been used to detect fraudulent transactions and flag them for manual intervention, while automating remaining transactions with instant approvals.

HOW A PAYMENTS BANK USED COMPUTER VISION FOR FRAUD DETECTION

Key Outcomes

  • Enhanced speed with 90% reduction in processing time
  • 70% reduction on manpower dependence
  • 80% higher accuracy than manual intervention
  • Fraud detection capabilities with intelligent detection of fake / duplicate RC copies
  • Improved scale of operations with ability to process more applications in lesser time, aiding revenue generation
  • Customer Delight with Instant Approvals and Instant Activation of Fast Tags
  • Improved distribution efficiency by allocating color coded Tags based on vehicle class automatically derived from the Document Processing Platforms and APIs