Fonts... leptonica, jbig2enc, tesseract and you

README

1. brew install ghostscript

2. Download leptonica: http://code.google.com/p/leptonica/downloads/detail?name=leptonica-1.69.tar.bz2

tar xvzf leptonica-1.69.tar.bz2
cd leptonica-1.69
./configure
make
make install

Get a pdf: 
curl -O http://www.rand.org/content/dam/rand/pubs/papers/2008/P4874.pdf

Run the pdf2tiff program (in prog):
mkdir 4874
leptonica/leptonica-1.69/prog/pdf2tiff P4874.pdf 4874

Remove pages you don't need
rm 4874/4874002.tif 
rm 4874/4874005.tif 
rm 4874/4874020.tif 

You can deskew:
leptonica/leptonica-1.69/prog/skewtest 4874001.tif 4874001-s.tif

3. git clone https://github.com/agl/jbig2enc
cd jbig2enc
./autogen.sh
# Turn on DUMP_ALL_SYMBOLS and DUMP_SYMBOL_GRAPH in jbig2enc.c
make 
make install

4. Run: jbig2 -s -t 0.5 -a 4874/4874*.tif