Introduction

The 3-Heights™ OCR Enterprise Add-On compliments several 3-Heights™ products with a high performance optical character recognition (OCR) function. There are no page limits.

Even large archives can be quickly and reliably converted into PDF- or PDF/A-Files that can be searched in full text. Multiple languages are supported. Together with the corresponding basic product, the add-on ensures a reliable OCR functionality.

Brief Description

Performance Properties

• High recognition quality  
• Multiple control options  
• Comprehensive language support
• No page limits  
• Easy to use
• Universally applicable 

Functions

• Recognition of machine generated texts  
• Recognition of typewriter scripts and barcodes (1D)
• Image manipulation
• Image pre-processing

Areas of Application

Embedding of text contents while ...

• scanning incoming mail  
• unpacking scanned E- mail attachments  
• preparing for archiving  
• archive migrations

Branches

• Public sector  
• Automotive industry
• Telecommunication
• Bank and insurance business  
• Archives and libraries  
• Health field  
• Pharmaceutical industry

Functions

The 3-Heights™ OCR Enterprise Add-On is an OCR module, which is used as an option with several 3-Heights™ products.  Based on the ABBYY FineReader Engine it recognizes text contents and embeds these as Unicode Text in the PDF- and PDF/A-File. This makes the PDF files full-text searchable. Numerous options in image manipulation, image pre-processing and text recognition allow a recognition process ideally coordinated to your needs. Almost 200 languages are supported; almost 50 languages are supported by dictionaries and morphologic tools.

Functions

• Recognition of machine generated texts  
• Recognition of typewriter scripts and barcodes (1D)
• Image manipulation
• Image pre-processing 

Benefits

Properties and Use

The 3-Heights™ OCR Enterprise Add-On provides high quality text recognition. According to your needs you can determine an ideal rate of recognition or a
high speed in the recognition process. In the add-on itself there are no page limits at all so that even the largest incoming documents or document batches can be made full-text searchable. Future versions also support load balancing, for optimally scaled performance. Thanks to theadd-on required information is quickly retrieved, which of course leads to noticeable cost reductions. Complicated manual indexing of documents is no longer necessary because the recognized texts can also be used in the metadata.
 

Performance properties

• High recognition quality  
• Multiple control options  
• Comprehensive language support
• No page limits  
• Easy to use
• Universally applicable  

Areas of Use

Inbox

Recognition of texts while scanning incoming mail. Usage of texts in the metadata of incoming documents and in the downstream business processes, for example ERP- and Workflow-Systems. Direct archiving of incoming documents with text recognition. Text recognition in scanned E-Mail attachments for easier processing.

Areas of Application

Embedding of text contents while ...

• scanning incoming mail  
• unpacking scanned E- mail attachments  
• preparing for archiving  
• archive migrations

Archiving

Text recognition when converting archives from TIFF or PDF to standardised PDF/A-Format. Conversion of proprietary formats to PDF/A and embedding of texts. Recognition of information on index pages and transmittal to the metadata of the document or dossier.

Branches

• Public sector  
• Automotive industry
• Telecommunication
• Bank and insurance business  
• Archives and libraries  
• Health field  
• Pharmaceutical industry

Technical Details

Architecture and Application Variants

The 3-Heights™ OCR Enterprise Add-On can be used together with the 3-Heights™ Image to PDF Converter, the 3-Heights™ PDF to PDF/A Converter and the 3-Heights™ Document Converter Service. It can be used with the Windows operating system and Linux (in preparation). It is activated via API, command line or the Windows service of the basic product. The options necessary for a recognition process are combined to a profile and stored as a simple string or text file. This allows optimal alignment of texts and scenarios that are to be recognized.

Variants and Required Basic Products

Product Variants

• The product is offered in an Enterprise-Variant without page limits.

Options

• Recognition of CJK-Scripts/languages (Chinese, Japanese, Korean)
• Recognition of 2D-Barcodes

Required Basic Products

The 3-Heights™ OCR Enterprise Add-On can only be used with the following basic products:

• 3-Heights™ Image to PDF Converter
• 3-Heights™ PDF to PDF/A Converter
• 3-Heights™ Document Converter Service

Extended Properties

General Parameter

• Optimal rate of recognition OR high speed recognition 
• Specification of document languages

Image Manipulation and Image Pre-Processing

• De-Skewing: Automatic image alignment   
• Image clean-up: Unwanted artefacts are recognized and eliminated
• Filtering of non-relevant backgrounds  
• Recognition and correction of page orientation 

Recognition Mechanisms in OCR (Optical Character Recognition)

• Recognition of almost 200 languages with machine generated contents,
• Extended support of almost 50 languages with dictionaries and morphological tools
• Recognition of multilingual documents  
• Recognition of typewriter scripts   
• Recognition and decoding of barcodes (1D)
• Recognition of type of content (images vs. texts)

The required options can be combined to one profile. Multiple profiles are possible.

Platforms

Operating Systems

• Windows 2000, XP, 2003, Vista, 2008, Windows 7  – 32 bit
• Linux: (SuSE and Red Hat on Intel) (in planning)

Interfaces and Languages

Interfaces

According to the basic product:

• API: C, Java, .NET, COM
• Shell Tool: command line for batch processing
• Windows Service: Windows service with monitored directories

Programming Languages

According to the basic product

Further Product Details

The 3-Heights™ OCR Enterprise Add-On compliments several 3-Heights™ products with a high performance optical character recognition (OCR) function. There are no page limits. Even large archives can be quickly and reliably converted into PDF- or PDF/A-Files that can be searched in full text. Multiple languages are supported. Together with the corresponding basic product, the add-on ensures a reliable OCR functionality.

Function

The 3-Heights™ OCR Enterprise Add-On is an OCR module, which is used as an option with several 3-Heights™ products. Based on the ABBYY FineReader Engine it recognizes text contents and embeds these as Unicode Text in the PDF- and PDF/A-File. This makes the PDF files full-text searchable. Numerous options in image manipulation, image pre-processing and text recognition allow a recognition process ideally coordinated to your needs. Almost 200 languages are supported; almost 50 languages are supported by dictionaries and morphologic tools.

Architecture and Application Variants

The 3-Heights™ OCR Enterprise Add-On can be used together with the 3-Heights™ Image to PDF Converter, the 3-Heights™ PDF to PDF/A Converter and the 3-Heights™ Document Converter Service. It can be used with the Windows operating system and Linux (in preparation). It is used via API, command line or Windows Service, according to the basic product. The options necessary for a recognition process are combined to a profile and stored as a simple string or text file. This allows optimal alignment of texts and scenarios that are to be recognized.

Properties and Use

The 3-Heights™ OCR Enterprise Add-On provides high quality text recognition. According to your needs you can determine an ideal rate of recognition or a high speed in the recognition process. In the add-on itself there are no page limits at all so that even the largest incoming documents or document batches can be made full-text searchable. Future versions also support load balancing, for optimally scaled performance. Thanks to theadd-on required information is quickly retrieved, which of course leads to noticeable cost reductions. Complicated manual indexing of documents is no longer necessary because the recognized texts can also be used in the metadata.