Pixel!OCR

Full-featured OCR with easy to use learning and forms processing capabilities

Mentalix, Inc. provides Pixel!OCR, a document imaging application that converts scanned documents to editable text, typically with greater than 99% accuracy. Virtually any document can be scanned, converted to text, and imported into any word processing or page layout file with ease. This application offers comprehensive OCR capabilities and has the ability to learn new fonts on-the-fly through its unique Omni-Learning mode. Pixel!OCR also performs automatic zone decomposition, automatic barcode zone decomposition, zone typing and forms processing.

This member of the Pixel!FX Deluxe family of integrated imaging products has application in a wide variety of markets: Desktop Publishing; Document Imaging; GIS; Scientific & Technical Applications; Office Automation; Web Page Authoring; Manufacturing; Medical; Educational; and Federal, State & Local Government. It has been designed to operate efficiently on all popular UNIX platforms. Displays supported include most monochrome, grayscale, 8-bit or 24-bit color display controllers.

It offers an icon-based user interface that includes complete point-and-click functionality. The use of tear-off menus and icon toolbars allows you to customize your display. Extensive drag-and-drop support gives you intuitive control of powerful image manipulation operations.

It also provides unprecedented speed and performance in the most demanding image processing environments. The result is increased user productivity with less time spent waiting for display refreshes and file saves.

NOTE: If you are a systems integrator who needs batch OCR functionality, see information on our Pixel!ENGINE OCR application.

Feature Summary

Multiple engines maximize recognition effectiveness based on image quality
Omnifont recognition engine typically recognizes with 99.5% accuracy
One-step scanning/OCR when used with Pixel!SCAN
Perform OCR on line art or grayscale images
Arbitrary region selection
Automatic zone decomposition and zone ordering
Automatic barcode zone decomposition
Zone settings can be used to improve accuracy by constraining recognition within zone to:
- ANSI Uppercase (A..Z)
- Lowercase (a..z)
- Fixed Length Expression (e.g., phone #, area code)
- Digit
- Alphabet
- Alphadigit (a..z, A..Z, 0..9)
- Numeric (0..9 $ + - % / = . ,)
- Word List (i.e., a list of acceptable words)
- Images
- Barcodes
ASCII or WYSIWYG file output
Recognized text output formats:
- ANSI
- Spreadsheet (tab-delimited)
- RTF
- HTML
Text accumulated over multiple images
Unreadable characters are marked
Barcode recognition
Handprinted text recognition
Fax recognition
European language support
Dot matrix support
RTF support retains font sizing, attributes and embedded images for export to a word processor
Dictionary improves recognition
Learn new fonts with Omni-Learning mode support and save learned fonts
Drag image icon from Image Album to display image in OCR window
Drag image to OCR dialog from other Motif 1.2-compliant applications
Zone settings can be saved and retrieved to enable forms processing