The way I would do this is use a rectangle-by-color extraction phase that produces all the cropped out colour rectangles with numbers and then perform ocr on each one which should be good success for the quality of text
Sent from my iPhone > On 21 Aug 2015, at 08:45, Rutger Rozendal <[email protected]> wrote: > > Dear People, > > We are using Tesseract to recognise room numbers on a floorpan of a ship deck. > Attached to this email two examples. > > We are trying different methods and have a mixture of results, let's say > recognising between 20% till 70% of the room numbers. > > Because the image come with color we are now wondering is results are better > when we take out colours upfront? > So making them black and white or grayscale. > Can we use Tesseract to do this color conversion with a certain profiling or > do we need to use an external program for that? > > Also we could work maybe with a search for specific patters, as these rooms > most of the time consist out of 4 digits. > > Any tips on direction for a best configuration is helpfull. > > Thanks in advance, > > Rutger > -- > 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 http://groups.google.com/group/tesseract-ocr. > To view this discussion on the web visit > https://groups.google.com/d/msgid/tesseract-ocr/bf5d8e06-7e42-464d-ab50-61951a89447e%40googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > <rci_hm_DECK09.jpg> > <cel_rf_DECK08.jpg> -- 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 http://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/51085070-72F3-4ED5-81C1-4BE34146C5BE%40gmail.com. For more options, visit https://groups.google.com/d/optout.
