Tesseract-ocr
Description
Tesseract is probably the most accurate open source OCR engine available. Combined with the Leptonica Image Processing Library it can read a wide variety of image formats and convert them to text in over 60 languages. It was one of the top 3 engines in the 1995 UNLV Accuracy test. Between 1995 and 2006 it had little work done on it, but since then it has been improved extensively by Google.
Provider
Development of Tesseract is sponsored by Google. Its chief developer is Ray Smith.
Licensing and cost
Tesseract is an Open Source OCR engine, available under the Apache 2.0 license. It can be used directly, or (for programmers) using an API.
History
It was initially developed at HP during a 10 year period from 1984 to 1994. After a decade of minimal development it was released in 2005 for open source. Google acquired Tesseract in 2006 and currently maintains its development.
Platform and interoperability
- The latest downloads for Linux and Windows are found on GoogleDrive. Older versions of Tesseract and its language packs are found on the discontinued Google Code download page.
- The easiest way to install Tesseract on Mac OSX is with MacPorts. Once it is installed, you can install Tesseract by running the command sudo port install tesseract, and any language with sudo port install tesseract-<langcode>. A list of available langcodes can be found on the MacPorts Tesseract page.
- Dependencies for running Tesseract include Autotools and Leptonica . The Windows version requires installation of Visual Studio. More information about required Ubantu libraries and links to specific requirements are on the Tesseract Wiki.
- Other programs such as Scan Tailor, unpaper, ImageJ, Gimp or ImageMagick may be needed to properly prepare images for use in Tesseract.
Functional notes
Input supported
Any image readable by Leptonica is supported in Tesseract including BMP, PNM, PNG, JFIF, JPEG, and TIFF. GIF is not supported http://www.leptonica.com/library-overview.html.
Output generated
plain text, hOCR (HTML), PDF, invisible-text-only PDF, TSV and ALTO (XML)
Documentation and support
- Smith, Ray (2007). An Overview of the Tesseract OCR Engine http://static.googleusercontent.com/media/research.google.com/en/us/pubs/archive/33418.pdf
- Installation information is found on the ReadMe page of the project site.
- Support is offered and issues are addressed on the Issues page of the project site.
Included in
- Integration with the free Xena-Digital Preservation Softwarehttp://sourceforge.net/projects/xena/?source=navbar
- Integration with Free Online OCR http://www.free-ocr.com/faq.html
Usability
- Tesseract was primarily developed for English OCR capability, but 47 language packs have been developed for use with other languages https://tesseract-ocr.googlecode.com/svn/trunk/doc/tesseract.1.html. Tesseract 2.0x and 3.0x are trainable for other languages.
- There is no built-in GUI, but there are several available from the 3rdParty page.
User Experiences
- Lazorchak, Butch. (2014). Making Scanned Content Accessible Using Full-text Search and OCR http://blogs.loc.gov/digitalpreservation/2014/08/making-scanned-content-accessible-using-full-text-search-and-ocr/
- Texas A&M University. (2012-). Early Modern OCR Project Workflow http://emop.tamu.edu/about http://emop.tamu.edu/about http://emop.tamu.edu/about
- Adams, Chris. (2014). Content Search on a Budget-using Tesseract on large TIFF files http://chris.improbable.org/2014/3/17/content-search-on-a-budget/
- PSNC Digital Libraries Team. (2011). Tesseract 3.0 installation on Ubuntu 10.10 server http://dl.psnc.pl/2011/01/24/tesseract-3-0-installation-on-ubuntu-10-10-server/
- Lacy, David. (2014). Digital Library upgrade provides enhanced discovery http://blog.library.villanova.edu/digitallibrary/2014/02/18/digital-library-upgrade-provides-enhanced-discovery/#sthash.OwCtLlEc.dpuf
- Applied in an AQuA Mashup that resulted in the Solution page: Compare OCR results of the same source material in different formats (TIFF, JP2)
Development Activity
Source Commits : https://github.com/tesseract-ocr/tesseract/commits
Issues : https://github.com/tesseract-ocr/tesseract/issues