Contributing to Gamera
Gamera is an open source project and joining its development is always welcome. If you have written a Gamera based application or toolkit, please let us know so that we can link to it in the "Addons" section.
You are also welcome to directly contribute to the Gamera code. Here is a TODO list of possible projects. Some of these projects might nicely fit as part of a bachelor's or master's thesis. Just contact us before starting, to avoid any duplicate efforts. We are glad to provide all necessary additional information.
- More page layout analysis resp. page segmentation methods. E.g. text/graphics separation and more text line finding algorithms. Maybe some parts of Leptonica could be useful?
- Add multilevel thresholding.
- Support FFT on Gamera images by wrapping FFT from VIGRA.
- Feature selection via genetic algorithms (GA) as an alternative to the currently implemented GA based feature weighting. Investigate why the GA based feature weighting in Gamera produces so poor results. (Tobias Bolten is currently working on this)
- A new menu function in the interactive classifier dialog for highlighting outliers, i.e. possible manual misclassifications. Can be based on MNN editing.
- Add automatic clustering algorithms for training data.
- Adding more classifiers like Gaussian mixture models, support vector machines, or hidden Markov models.
- Add support for the hOCR exchange format in the form of a general writing and parsing library.
- Write wrappers for external OCR engines like Tesseract OCR as optional plugins.
- Completely overhaul (or rewrite?) the vector toolkit. Maybe lib2geom can be utilized? (does anyone still need vectorization?)
- Add more image formats: JPEG (reading/writing), PDF (writing)...