Distinguishing between "1" and "/"

[Paul Grebenc]

I'm working with Tesseract 3.02, trying to perform an OCR on an original source image (not scanned, so there is no noise or other artifacts).  The image contains the text "6582044/1", but it is detecting "6582044I1".

I've tried setting tessedit_char_whitelist to "/0123456789" (because in this case I know my input text will only contains digits and slashes), but the result is then "658204411".

Is there anything I should try, to improve the accuracy in distinguishing forward slashes?

Thanks,

Paul