Brainvire Automates The End-To-End Digital-KYC Process For Philippine Finance Leader
About
The client, a leading Philippines financial services provider, believes in prioritizing customer service using technology.
Project Highlights
The client was keen to completely automate its KYC process of digitizing and extracting text from customers for perfect data accuracy and efficiency in document processing. Their objective was to automate data extraction from over 65,000 accounts by scanning and mapping customer data with a selfie. Brainvire combined OCR technology with Open Source solution for data accuracy and improved document processing efficiency.
The Challenges
- ID and Facial Scanning:The client wanted to scan large volumes of at least 65,000 accounts documents
- Map Data With Customer Selfie:The client wanted to ensure that selfie data is as accurate as the customer’s photo ID
- Minimum accuracy for data extraction:There was a need to have at least 95% accuracy rate for printed and 85% for handwritten text.
- Ensure image match:Compare image received in base-64 encoded format from the client’s end without storing in the database
Tech Stack
Google Vision
Python
Result
Scanned documents for 65K customer accounts
Seamless data migration with facial scan that mapped scanned data to automate its KYC documentation process.Faster Document Processing With Accuracy
This helped in accurately processing KYC details with verified mapping of selfie with the actual customer’s ID photo.Automated extracting text from official documents
Brainvire helped reach a near 100% for accurately scanning both printed and handwritten text from documentsImage comparison from client side without storing in database
We first stored the base-64 encoded format in a temporary variable and decoded it using the NumPy module for accurate image comparison
