Links
Here are some sites possibly interesting to Gamera users. For additional projects realized with Gamera, see the sections "Addons" and "Publications".
Projects that use Gamera
- The DDMAL group at the McGill University (Canada) uses Gamera for neume recognition. Among other things, they have created an online search tool for the Liber Usualis.
- The ISMS group at the Goldsmiths College of the University of London (England) uses Gamera for the recognition of lute tablature in order to build an Electronic Corpus of Lute Music.
- The Mount Allison University (Canada) uses Gamera for the recognition of 19th century polytonal Greek texts.
- The Junior Enterprise Catania (Italy) uses Gamera for an automatic processing of multiple choice tests.
- The didjvu software by Jakub Wilk uses Gamera for foreground/background separation during converting images to the DjVu file format.
Material on document image analysis
Of the few textbooks specifically devoted to document image analysis (DIA), most are out of print. Moreover they are better characterized as loose collections of research papers than as systematic textbooks. The following material might be helpful for getting started in the field:
- The survey Document image analysis: A primer. by Kasturi, O'Gorman, and Govindaraju can serve as a cursory introduction with many references for looking up the details.
- A textbook-like introduction into document image analysis is given in the lecture on document image analysis by Christoph Dalitz. On its website (in German), you can find material and exercises utilizing Gamera on specific stages of the recognition process.
Software projects on document image analysis
Beside Gamera, there are also other free software projects related to specific document image analysis tasks:
- Tesseract OCR is perhaps one of the most accurate open source OCR engines available. It was originally developed at HP between 1985 and 1995 and open sourced in 2005.
- Audiveris is a complete optical music recognition system written in Java.
- OCRopus is a framework for text document recognition which allows for pluggable layout analysis and OCR engines. Within the context of this project, the hOCR exchange format is defined for OCR and page layout information.
- Leptonica is a library that includes page segmentation functionality.
- OpenCV is a computer vision C++ library that also includes a number of classification algorithms.
Other software
The core of Gamera uses the following software:
- VIGRA by Ulrich Köthe, a template based C++ library for image processing
- Evolving Objects, a C++ library of genetic algorithm components
- wxPython, a cross platform GUI library
- libpng and libtiff for reading and writing images
- The python modules docutils and pygments for generating the documentation
Other useful software for scientific projects can be
- igraph is a Python graph library that can be used as an alternative to the Gamera graph API.
- SciPy/NumPy is a collection of numerical tools for Python. Among other things, it includes a Python wrapper around FFTW.
- matplotlib for directly creating plots within Python
- For the generation of publication quality plots, gnuplot is very flexible, because it allows the export of the plots in the vector graphics FIG format, which can be edited with xfig.