Any Example for Document Segmentation with hOCR output?

[zmwangGH]

It seems a bit quite hard to get an example regarding the document layout analysis with tesseract-ocr python binding.

I want to analyse the pdf document and separate the regions into text and images. But I cannot find the relevant pages/threads regarding this.

Thanks a lot.

[sirfz]

Have you tried the Iterator over the classifier choices for a single symbol: example?

You can get both the text and image at the level you want (e.g. RIL.WORD for words), the PyResultIterator class offers access to many components at any level.

If you have examples for the C++ APIs, they should be very easily replicated using tesserocr (all examples in the README are from the C++ API examples ported to tesserocr).