abbyy recognition server 5.0 at #abbyysummit17
TRANSCRIPT
ABBYY TechnologySummit2017
© ABBYY Confidential
ABBYY NAHQ, 2017
FlexiCapture Technical Track
ABBYY Recognition Server 5 Alpha
What’s New
© ABBYY Confidential
Contents
• Smart PDF quality detection and processing
• Support of PDF/E standard
• Import of email messages in MSG format
• Advanced document editing
• Support of user languages and patterns
• Extracting index fields by template of fixed regions
• Other improvements
© ABBYY Confidential 3
Smart PDF Quality Detection and Processing
© ABBYY Confidential
The Goals
• Produce PDF files with highest quality of text layer
• Preserve original meta-data and properties
Initial Challenges
1. Input files can be created in various applications, scanned by MFP with some OCR, merged from several documents with and without text on pages
2. ReOCR leads to lose of meta-data, bookmarks and originally good text
3. Skipping the OCR leads to non-searchable inline pictures, no detection of rotation, non-unified export standard of output files
Smart PDF Quality Detection and Processing
© ABBYY Confidential 5
New Features
• Page-by-page detection of PDF quality (auto mode)
• Merging OCRed text from pictures and the original text layer
• Preserving bookmarks from source PDF
• Verification and indexing of PDF files with no re-OCR
Support of PDF/E Standard
© ABBYY Confidential 6
• ISO 24517-1:2008
Engineering document format using PDF—Part 1: Use of PDF 1.6 (PDF/E-1)
PDF/E
Import of MSG Files
• Processing of emails in MSG format from folders and document library
Advanced Document Editing at Indexing and Verification Stations
© ABBYY Confidential
The Goal
• Offer a tool for operator to improve the input document
New Features
• OCR editor and Image editor
• Rotating, reordering and deletion of pages
• Text redaction
Support of User Languages and Patterns
© ABBYY Confidential
The Goal
• Fine-tune the OCR for non-standard text and fonts
New Features
• Creation of user languages
• Pattern training support
Extracting Index Fields by Template of Fixed Regions
© ABBYY Confidential 9
The Goal
• Simplify indexing for fixed fields
New Features
• Creation of template at Indexing Station
• Automated recognition of fields from the given zones
Other Improvements
• Native 64-bit support
• Access from Admin Console to server on remote machine
• Communication via TCP/IP protocol by default
• OCR Technology 15 support
© ABBYY Confidential 10
Recognition Server Release Plan
© ABBYY Confidential 11
ABBYY Recognition Server 5 Landing Page
© ABBYY Confidential 12
• https://www.abbyy.com/en-us/lp/recognition-server/alpha/
ABBYY Recognition Server 5 Wiki Pages
© ABBYY Confidential 13
https://wiki.abbyy.com/display/DAT/Recognition+Server
Thank You!
© ABBYY Confidential 14