Microsoft Azure OCR Xoriant SmartCapture For A Fortune 100 Bank
Our client is a leading Fortune 100 bank with a global client base of over a million customers. To maintain its credibility as a trusted international bank, the client had to ensure compliance with countrywide regulations and standards. However, the banking compliance process included manual data scanning across all Annexure documents. This was time-consuming and prone to human errors. Besides, there was a high risk of non-adherence to different regulations like Office of Foreign Assets Control (OFAC) compliance, Know Your Customer (KYC) that would have severe repercussions.
The key objectives of the client included:
- Automate the compliance data processing.
- Make use of a solution to accelerate the process, overcome challenges in complying with OFAC regulations.
- Ensure thorough scrutiny and review of all supporting documents involved in transactions to point out suspicious and bad entity names.
- Reduce the operational costs associated with business processes.
Xoriant Solution | Key Contributions
Xoriant used its very own state-of-the-art intuitive document digitization solution SmartCapture developed using Microsoft Azure OCR. The solution uses Microsoft Azure OCR services and Xoriant Machine Learning algorithms to extract printed text from images and multi-page PDF documents. Xoriant SmartCapture solution helped the client to scan all the transaction documents accurately to reduce manual efforts and errors.
The key contributions included:
- Designed the Xoriant SmartCapture solution to accept input files in various formats - plain files/OCR/HTML files seamlessly integrating with the client‚Äôs third-party OCR tools.
- Developed classification, decision-making, and machine learning algorithms for Address detections, categorized text/data into different entities, and re-used them for future checks.
- Implemented UI-based training framework to apply incremental feedback for named entities.
- Predefined text template to support context-based training using machine learning algorithms and improved scanning accuracy.
- Ensured the ability to locate and classify elements into pre-defined categories that helped in quick analysis of the transactions or documents and improved efficiency.
- Achieved 95% accuracy in compliance checks by implementing Xoriant Smart Capture solution.
- Reduced manual tasks and operational costs by 30% using Xoriant SmartCapture.
- Improved productivity in document scanning and processing with an automated workflow process.
Python | Computer Vision (OpenCV) | Object Detection (inception model) Image Processing (OpenCV) | Tesseract OCR | Azure OCR | Virtual Machines on Azure