Not sure if this is a good idea, but it would be easy and quick to try. What is the matrix of your image? Have you tried increasing the image size and then re-scanning?
On Wednesday, August 21, 2013 12:44:45 PM UTC-4, Morlock wrote: > > Hello, > > I'm using Tesseract v3.02.02. > > I'm unable to get it to consistantly recognizing a number that contains a > decimal point. Tesseract is recognizing the digits. Tesseract is > recognizing the leading minus sign when there. > But, it is always throwing out the decimal point. Has anyone else run > into this and found a solution? I'm doing this all from a C++ executable on > Ubuntu. > > Here are the things I tried and did not help: > > 1. I normally use png files. I've tried jpg and tiff but the results are > the same. > 2. I tried playing with contrast. I made the background black and the > color of the numbers bright yellow > 3. I tried setting the tessedit_char_whitelist to ".-0123456789" explicitly > 4. I've also use a TesseractRect to only OCR the numbers I'm interested in. > 5. I tried using GetUNLVText instead of GetUTF8Text. GetUNLVText was > worse and just returned garbage. > 6. The command-line tesseract command also gives me the same results; I > even tried using the 'digits' configuration. > > regards,m > -- 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/5e1a6049-0f50-4f5a-9d44-cbe08954e5a1%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
