An OCR SDK is a software development kit for adding optical character recognition capabilities to forms processing applications, document imaging management systems, e-discovery systems and records management solutions.
In order to avoid the difficulties of incorporating OCR technology, some OCR SDKs contain a high number of APIs, support multiple operating systems and programming languages.
Here is a non-exhaustive comparison of optical character recognition software:
Name | Founded year | Latest stable version | Release year | License | Online | Windows | Mac OS X | Linux | BSD | Programming language | SDK? | Languages | Fonts | Notes |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
ABBYY FineReader | 1989 | 11 | 2011 | Proprietary | Yes | Yes | Yes | Yes | Yes | C/C++ | Yes | 186[1] | ? | ABBYY also supplies SDKs for embedded and mobile devices. Professional, Corporate and Site License Editions for Windows, Express Edition for Mac.[2] |
AnyDoc Software | 1989 | ? | ? | Proprietary | No | Yes | No | No | No | VBScript | ? | ? | ? | Works with structured, semi-structured, and unstructured documents. |
CuneiForm/OpenOCR | ? | 12 | 2007 | BSD variant | No | Yes | Yes | Yes | Yes | C/C++ | Yes | 28 | Any printed font | Enterprise-class system, can save text formatting and recognizes complicated tables of any structure |
ExperVision TypeReader & RTK | 1987 | 7.1.170.1125 | 2010 | Proprietary | Yes | Yes | Yes | Yes | Yes | C/C++ | Yes | 17 | 2618 | Won the highest marks in the independent testing performed by UNLV for X consecutive years (in 1994).[3]
The speed of ExperVision’s OpenRTK is four to eight times faster than competition. — PC Magazine[4] but also "Not as accurate as rival products, clumsy interface, limited options for proofreading, couldn't open some files in standard PDF or image formats." [5]PC Magazine |
OCRFORMS[6] | 2009 | 11.10 | 2011 | Proprietary | No | Yes | No | Yes | No | C/Python | No | Any language based on latin alphabet | Printed and written latin fonts | Features a complete GUI and has a command-line tool for batch processing. Propietary algorithms for OCR/ICR/OMR and advanced string correction technology |
GOCR | ? | 0.47 | 2009 | GPL | No | Yes | Yes | Yes | Yes | C | ? | ? | ? | |
LEADTOOLS[7] | 1990[8] | 17 | 2010 | Proprietary | No | Yes | No | No | No | various | Yes | 56[9] | Any printed font | Supports Latin, Asian, Arabic, and MICR character sets.[7] For full page, zonal, and form image processing. Includes OCR, barcode, OMR and forms recognition.[10] ICR (handwritten text recognition) is supported.[11] |
Java OCR | ? | Java OCR | 2010 | {{{1}}} | No | Yes | No | No | No | ? | ? | ? | ? | Uses Java |
Microsoft Office Document Imaging | ? | Office 2007 | 2007 | Proprietary | No | Yes | No | No | No | ? | ? | ? | ? | Uses OmniPage |
Microsoft Office OneNote 2007 | 2007 | ? | 2007 | Proprietary | No | Yes | No | No | No | ? | ? | ? | ? | |
Ocrad | ? | 0.20 | 2010 | GPL | Yes | Yes | Yes | Yes | Yes | C++ | Yes | Latin alphabet | ? | Command line |
OCRopus | ? | 0.3.1 | 2008 | Apache | No | No | No | Yes | No | C++ and Lua | ? | ? | ? | Pluggable framework which can use Tesseract |
OCRFeeder | ? | 0.7.7 | 2009 | GPL | No | No | No | Yes | No | Python | ? | ? | ? | Features a full user interface and has a command-line tool for automatic operations. Has its own segmentation algorithm but uses system-wide OCR engines like Tesseract or Ocrad |
OmniPage | 2005 | 18 | 2011 | Proprietary | No | Yes | Yes | No | No | C/C++/C#[12] | Yes | ? | ? | Product of Nuance Communications |
PSI:Capture | 1995 | 4.1 | 2011 | Proprietary | No | Yes | No | No | No | C# | No | 99 | Any printed font | Scan, capture and extract data from business documents such as invoices, forms and correspondance and export images/data to over 50 different backend systems including Microsoft SharePoint. |
Puma.NET | ? | ? | ? | BSD | No | Yes | No | No | No | C# | Yes | 28 | Any printed font | .NET OCR SDK based on Cognitive Technologies' CuneiForm recognition engine. Wraps Puma COM server and provides simplified API for .NET applications |
Readiris | ? | 12 Pro | 2009 | Proprietary | No | Yes | Yes | No | No | C++ | Yes | ? | ? | Product of I.R.I.S. Group of Belgium. Asian and Middle Eastern editions. |
ReadSoft | ? | ? | ? | Proprietary | No | Yes | No | No | No | ? | ? | ? | ? | Scan, capture and classify business documents such as invoices, forms and purchase orders integrated with business processes. |
RelayFax | ? | ? | ? | Proprietary | No | Yes | No | No | No | ? | ? | Many | ? | Converts faxed pages into editable document formats (doc, PDF, etc...). |
Scantron | ? Cognition | ? | ? | Proprietary | No | Yes | No | No | No | ? | ? | ? | ? | For working with localized interfaces, corresponding language support is required. |
SimpleOCR | 2002 | 3.5 | 2008 | Proprietary | No | Yes | No | No | No | ? | ? | ? | ? | |
SmartScore | ? | ? | ? | Proprietary | No | Yes | Yes | No | No | ? | ? | ? | ? | For musical scores |
Tesseract | ? | 3.01 | 2010 | Apache | No | Yes[13] | Yes | Yes | No | C++, C | ? | 35+[14] | ? | Created by Hewlett-Packard; under further development by Google |
Transym OCR | ? | 3.0 | 2008 | Proprietary | No | Yes | No | No | No | C#, C/C++, VB, VB.NET | Yes | 11 | ? | |
Zonal OCR | ? | ? | ? | Proprietary | No | Yes | No | No | No | ? | ? | ? | ? | |
Name | Founded year | Latest stable version | Release year | License | Online | Windows | Mac OS X | Linux | BSD | Programming language | SDK? | Languages | Fonts | Notes |