MathFinder
MathFinder copied to clipboard
Uses Tesseract OCR outputs to locate and segment math regions from document images
This project requires Qt, libpng-dev, libjpeg-dev, and Leptonica (http://www.leptonica.com/) to be pre-installed and on the system path. Need to install the png and jpeg libraries before Leptonica for it to be able to read/write png and jpeg images. It optionally requires Java for some of the debugging features work.
This project can be compiled and installed by running the following commands from this directory:
autoreconf -vfi ./configure make sudo make install
The TESSDATA_PREFIX and SCROLLVIEW_PATH environment variables also need to be set in order to run Tesseract and view results respectively. This is because Tesseract only looks on those specific paths and won't look on the system path. This can be done by running the following commands:
export TESSDATA_PREFIX=/usr/local/share export SCROLLVIEW_PATH=/usr/local/share/scrollview
And then copying and pasting the above lines to the ~/.bashrc file to persist the changes.
The following commands will run the application:
MathFinder [path to directory or image on which to run]
To run the project for training, evaluation, or to print information, run with interactive menu as follows:
MathFinder -m