text2image: Segmentation fault: 11

[Niko Partanen]

Sorry if I've missed something very basic. I've compiled Tesseract and the training tools, but when I try to run text2image I get as only output the line:

Segmentation fault: 11

I've tried different input text files, and they all give the same error. Same with different fonts. I'm using Mac OS X Yosemite. I'm just learning how this model training works and try to get it run for something, so my latest input file has just contained the English "Quick brown fox" example sentence I found from Tesseract wiki in GitHub. 

This is the command I've run in tesseract-3.04.00 folder:

training/text2image --text=../training_texts/model.txt --outputbase=test.arial.exp0 --font='Arial Medium' --fonts_dir=~/Library/Fonts/

I installed Tesseract 3.04 from the release in GitHub.

https://github.com/tesseract-ocr/tesseract/releases

Here is the version information:

tesseract 3.04.00

 leptonica-1.71

  libgif 4.2.3 : libjpeg 9a : libpng 1.6.18 : libtiff 4.0.4 : zlib 1.2.8 : libwebp 0.4.3 : libopenjp2 2.1.0

It is difficult to find information about this segmentation fault 11 with Google, apparently it isn't so common. The book I want to OCR is in Komi language with rather specific but not very complicated orthography. I have lots of text in the same variant, and thought to try to develop a language model for the purpose.

I would appreciate any help! Can it be that I'm missing some dependency?

Best wishes,

Niko

[Niko Partanen]

I got it work fine running Ubuntu in VirtualBox. Of course this is not really a surprise to anyone. I think the problem is that I don't have cairo-devel version on my Mac, though I have problems installing it through MacPorts as there are so many dependencies to cairo already which don't allow me to switch into cairo-devel version. I don't know if this is the issue, but maybe.