LEADTOOLS OCR Module

What's New in the LEADTOOLS Document Engine V17.5 Re-mastered
The LEADTOOLS Document Imaging engine has received major enhancements, including significant improvements to the speed and accuracy of its OCR, PDF and Barcode engines, as well as new SharePoint integration capabilities.

OCR Advantage
The latest version of LEADTOOLS 17.5 includes dramatically increased speed and accuracy of the OCR Advantage engine.

• Increased speed by up to 300%
• Faster and More accurate table detection
• Auto Orient up to 8x faster
• Improved algorithm for small zones - yielding greater accuracy for Forms Recognition and Processing
• Greatly enhance accuracy of color image recognition using new segment based auto-binarization algorithm

PDF

• Load times between 400% and 800% faster
• PDF Support for v1.7
• Added support for reading with 128 bit AES and 256 bit AESv3 encryption
• Native multithreaded support

SharePoint Integration

• Load and Save images directly to and from SharePoint
• SharePoint integration with other LEADTOOLS features such as OCR, Barcode, File Format Conversion and more

Microsoft Phone 7 Silverlight

• Improved speed for Windows Phone 7 Barcode libraries
• New Live Capture Barcode demo

What's new in LEADTOOLS Document Engine V17.5

• Detect, read and write Barcodes in Silverlight and Windows Phone
• New high-level Barcode design for .NETsimplifies adding support for all barcode types in your application without any special knowledge of barcode symbologies
• Dramatically increased barcode recognition speed by up to 9X
• Automated cleanup for barcodes
• PDF updates and improvements
  • PDF runtime now requires only a single DLL for distribution
  • Merge multiple PDF or Postscript files into a single PDF document
  • Split a PDF into multiple documents
  • Extract text with properties such as bounding rectangle, font information and color
  • Extract hyperlinks, bookmarks and internal links
  • Load and save PDF metadata such as Title, Author, Subject, Keywords and Producer
  • New PDF Document Editor with extended support for adding, replacing, and deleting pages in any PDF file
  • Enhanced PDF encryption support
  • Convert Postscript files to PDF with full control over distiller, encryption and compatability settings
  • Convert existing PDF files to linearized PDFs(optimized for web viewing)
  • Convert existing PDF files to PDF/A-1B
  • Save fully customized UNICODE Bookmarks in PDF files
  • Added support for multi-threaded PDF load
  • Added support for writing PDF 1.3 and 1.6
  • New PDF Info Extractor gets information about PDF version, image width, height, bit depth, compression and DPI
• OCR Advantage Engine updates and improvements
  • AutoZone has been improved to provide fewer, more accurate zones, improving OCR results
  • Dramatically accelerated AutoZone detection, improving overall performance for OCR
  • Faster Advanced Table Detection, providing better performance when dealing with documents containing tables
  • New image pre-processing features to improve OCR results including fixing broken letters, dithered text, inverted text and more.
  • Now supports the following languages: Bulgarian, Catalan, Chinese (Traditional and Simplified), Czech, Danish, Dutch, Finnish, Greek, Hungarian, Indonesian, Italian, Japanese, Korean, Latvian, Lithuanian, Norwegian, Polish, Portuguese, Romanian, Russian, Serbian, Slovak, Slovenian, Swedish, Turkish, Ukrainian and Vietnamese
  • Added support for built-in Spellcheckingincluding NHunspell and Microsoft's System.Windows.Controls.SpellCheck engines
• Enhancements to Existing File Formats
  • Support for loading Office 2007 formats: DOCX, XLSX and PPTX
  • Optimized load performance for DOC format
  • Added ability to load Office formats from memory
  • Added support for JBIG2 with non-sequential headers (i.e. random access)
  • Support for loading refinement segments in JBIG2 format
• Increased the speed of the Line Remove image processing function
• New ImageColorType function for detecting whether the image is mostly black and white, grayscale or color.
• Virtual Printer now supports network printing

Overview

With the LEADTOOLS OCR module a developer can add many zoned areas to the same page and each zone can have its own options such as the OCR engine, filter and much more. LEAD makes OCR development easier by presetting OCR properties that work for most images by default. And to improve recognition results, LEADTOOLS OCR supports custom dictionaries to recognize words that may only exist in the documents being recognized. Other OCR module features include support for over 100 languages, output document options like document margins, paragraph options and more, and new output formats.

Three specialized OCR recognition engines are supported:

MOR OCR Engine

• Supports 114 languages
• Supports up to 500 zones on one image
• Supports Omnifont, Draftdot24 and OCR-A filling methods
• Supports character training to achieve improved accuracy
• Provides 3 page-level accuracy and speed trade off settings including Accurate, Balanced and Fast
• Provides Checking Subsystem based correction

MTX (Mtext) OCR Engine

• The fastest of the selectable OCR engines
• Support for 12 languages
• Supports up to 64 zones on one image
• Supports Omnifont, Draftdot9 and Draftdot24 filling methods
• Provides 2 page-level accuracy and speed trade off settings including a combined Accurate & Balanced value and Fast
• Provides Checking Subsystem based correction

FireWorX OCR Engine

• Optimized for speed
• Support for 54 languages
• Supports up to 2,500 zones on one image
• Supports Omnifont filling methods
• Supports character training to achieve improved accuracy

Output formats supported:

• Adobe PDF edited
• Open eBook 1.0
• XML
• 2G Type 2
• 2G Type 3
• Over 40 output formats supported

Additional Features:

• Recognize and export text, choosing from a variety of text, word processing, database, or spreadsheet file formats
• Select the language of documents to be recognized. Choose from English, Danish, Dutch, Finnish, French, German, Italian, Norwegian, Portuguese, Russian, Spanish, or Swedish
• Documents with multiple languages can be recognized
• Set accuracy thresholds prior to recognition to control the accuracy of recognition
• Learn, save and load character recognition data for similar documents. The software learns as a result of normal recognition, and acquires additional information by using the OCR's text verification system
• Recognize text from 5 to 72 points in virtually any typeface
• Increase recognition accuracy with built-in lexical classes and user defined lexicons
• Verify or correct text during the recognition process based on confidence levels set prior to recognition. If a word or character falls within the set range, a dialog box can be brought up to allow the user to see the original image and the preliminary results of the recognition. From the dialog box, the user may make any necessary corrections to the recognized text
• Automatically detect fax, dot matrix, and other degraded documents and compensate accordingly
• Recognize all document pages at once while managing multiple pages of a document when pages are in different stages of processing (segmentation vs. recognition)
• Recognize a document page(s) and export text in raw format to over 40 different formats including MS Word, PDF*, MS Excel, Dbase, WordPerfect and ScanSoft's defined mark-up language format (XDoc/XDoc Lite)
• Automatically segment the page to correctly recognize text on pages with complex or irregular layouts, including tables, reverse video, and line art as well
• Process both text and graphics. The recognition software's ability to distinguish graphics from text provides the basis of creating a compound document processing system
• Process documents in two-page mode for open-faced books and magazines
• Process images from newspapers with special image processing designed for newsprint
• And more...

LEADTOOLS provides an interface to verify or correct text during the recognition. LEAD's OCR verification dialog ties the text being edited directly to the image, providing a visual reference to the original bitmap data. The OCR engine can perform Automatic area segmentation creating multi-layered zones, recognizing areas such as tables, rules, images and text. Or, you can manually designate up to 2500 such zones.

Fax, dot matrix and halftones can be preprocessed to improve recognition results. The OCR Engine supports major European and Scandinavian languages (Danish, Dutch, Finnish, French, German, Italian, Norwegian, Portuguese, Russian, Spanish, and Swedish) as well as English. Recognized text can be exported to more than 40 different formats, including MS Word, PDF*, MS Excel, Dbase, WordPerfect and ScanSoft's defined mark-up language format (XDoc/XDoc Lite). You get superior OCR processing speeds, for use in form recognition and processing applications.