Am Donnerstag, 17. November 2016 17:39:28 UTC+1 schrieb Art Rhyno: > Your example has such good contrast that you might consider using the > colors to identify single characters. I have attached a quick sample of > what I mean. I used opencv and defer greatly to the blog post I reference > at the top of the script, but the idea would be to try to catch single > characters using opencv’s “inrange” function. I would use tesseract on the > image first and weed out blobs for further processing based on the > coordinates of what tesseract has already detected. I would then use single > character mode on what’s left. Feel free to ping me if you are interested > in this approach. >
Thanks for your suggestion! Looks like a neat way to circumvent the problem. However, I'd prefer to find the reason why tesseract rejects those blobs first. (See my other post.) Maybe this can be fixed in tesseract, once I know some background... :-) Cheers, --leo -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/d16bb097-f4a7-4deb-a5bd-fa1545e25c33%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
