State estimation in a document image and its applciation in text block identification and text line extraction

> block.py contains some text-structuring stuff. I intended to use this as a replacement for the current text-line detection system, but I haven't been able to get it to work.

What is the issue(s) you encounter?

Also I noticed that https://github.com/phulin/rebook/blob/master/rebook/block.py#L174 is probably an error. To invert the image, rather use `im = 255 - im_inv`.