HOME :: JOB LISTINGS :: ARCHIVES :: MEDIA KIT :: SUBSCRIBE


BusinessWire
March 29, 2006 08:00 AM US Pacific Timezone

Visionet Systems Launches Mortgage-Specific Optical Character Recognition (OCR) Toolkit and Service MBA's National Technology in Mortgage Banking Conference & Expo


CRANBURY, N.J.--(BUSINESS WIRE)--March 29, 2006--
Using a Combination of Template-Matching, Template Precedence Ordering and a Document Dictionary, VisiOCR(R) Simplifies Document Indexing for Lenders and Servicers

Visionet Systems has launched VisiOCR(R), a low-cost document indexing and data entry technology service for servicers and lenders. VisiOCR enables mortgage companies to reduce indexing and data entry costs and improve the accuracy of indexing and data through offshore data entry services and document-to-data processes. Properly indexed loan files help improve efficiency from underwriting to secondary marketing and servicing. Automation of data capture for loan review and legal description speeds up the mortgage supply chain. The news was announced at Mortgage Bankers Association's 2006 National Technology in Mortgage Banking Conference and Expo.

VisiOCR is an artificial intelligence-based engine running on parallel processors to deliver speed and accuracy. Manually indexing a 200-250 page loan file currently requires 10-15 minutes. VisiOCR minimizes human involvement while delivering a 50 percent cost reduction in indexing alone. As the documents change, the system detects this change and expands its library to improve automated indexing.

VisiOCR can be integrated into a lender's environment without disturbing its existing imaging infrastructure or outsourcing arrangements. It has an embedded workflow to channel indexing and data capture exceptions to separate queues so that the data in question and the document can be viewed and worked on in a side-by-side window arrangement. Once the exceptions are cleared, the data and documents are committed back to the system of record thus furthering the supply chain management paradigm in the mortgage industry.

"Document indexing has traditionally been a labor-intensive process for lenders," said Arshad Masood, president of Visionet Systems. "Now that the industry has embraced imaging and is ready to automate document-to-data entry to improve reporting, compliance and data integrity, many servicers are looking to take advantage of offshore centers to improve productivity by reducing errors associated with manual entry. VisiOCR helps servicers improve the accuracy of legal descriptions, loan applications and title policy."

Visionet Systems uses a combination of proprietary data extraction OCR tools and dictionaries to populate the forms with the necessary data, thus reducing cycle times and significantly lowering costs. Visionet Systems has also developed several tools that improve recognition rates including image and text identifiers used to recognize sections within documents and OCR-assisted page and document recognition and indexing.

"As with our other workflow-based products, VisiOCR furthers our commitment to providing servicers a low-cost option while also improving overall productivity," continued Masood. "By automating document validation, servicers are moving one step closer to being completely paperless. Based on our findings of human versus OCR metrics specific to the mortgage industry, we have concluded that OCR is a necessary component in a servicer's quest to become more efficient while also improving profitability."

About Visionet Systems

Cranbury, N.J.-based Visionet Systems is a full service mortgage consulting and business process outsourcing company that delivers products in the best-of-breed philosophy. Using a combination of consulting, metrics-based technology solutions and business process outsourcing, Visionet Systems helps lenders identify and remove inefficient processes to improve profitability. Founded in 1995, the company's customers include mortgage lenders and servicers. For more information, visit www.visionetsystems.com


 

All material on this site copyright © 2006 techfocus media, inc. All rights reserved.
Embedded Technology Journal
Privacy Statement