Could you provide sample images from the training and testing set? I haven't tried training Tesseract with single characters at a time but you might want to try training on whole expressions like x+y=0.
On Wed, Dec 18, 2019, 11:39 PM Haris Sheikh <[email protected]> wrote: > hi i'm using Linux (ubuntu), > i tried tesseract training by following this > https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract-4.00 > and i used data set like: > '=' folder -> 26,000 .jpg image files in which = is written in different > forms > '+' folder -> 30,000 .jpg image files in which + is written in different > forms > so on > > i take all the images from each folder and paste it into ground-truth > folder and converted those images into .tif format and also created their > labels in .gt.txt format > then execute the command: "make training" > it worked fine and it took 5-6 hours to train the dataset, after that i > used the data/foo.traineddata file and paste into > /usr/local/share/tessdata/ directory and > run command: "tesseract --list-langs" it showed me that there is my file > and then > > *Issue is this:* > > when i use a sample image having "x+y=0" written, and run tesseract as my > language it gives me output as "xxxx" *why?* > > *please tell me where i get wrong!* > > -- > 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 view this discussion on the web visit > https://groups.google.com/d/msgid/tesseract-ocr/9c24849b-69a5-4f6d-928f-da17420adfa3%40googlegroups.com > <https://groups.google.com/d/msgid/tesseract-ocr/9c24849b-69a5-4f6d-928f-da17420adfa3%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- 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 view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/CABtjQ9LOmTxbj6Jft7YbY2Yr6vg%3DQVnPqyNckM-7BmyTGLR5OA%40mail.gmail.com.
