Color OCR information lost?

[rs_nuke]

Hello all, I and I must admit that the the ocropus project to be a
great one.

I wonder how I can get 'color' information, or if that is even
possible? I am using ocropus to ocr "linked" (hyperlinked) documents,
and hyperlinks are usually underlined and in color blue. I wonder how
I cannot lose this information. Is there a way to know what words were
underlined? or in blue color?

[Ilya Mezhirov]

Hello,

Thanks for your interest.
Currently the color information is just lost and there are no plans
currently to get it through.

Ilya

[Thomas Breuel]

Well, there are plans to support color and other font properties at some point, just not in the upcoming releases.

If you really need it, it's not that hard to write a post-processor that takes the hOCR output, reads the corresponding image, determines the color information from the image for the bounding boxes, and adds it to the hOCR info.  It's probably a few hundred lines of Python code (using PIL, DOM, and building on some of the hOCR sample tools in Python).

Cheers,
Thomas.

[Nitin Shinde]

Is there any updates regarding support for color and other properties ??