I have been trying with some success to have tesseract recognize text from iPhone's about screen.
I have applied various contrast / brightness tweaks, removed blue from the images, converted to grey, and stretched to 4x actual size. Using --psm 6 with -l que gives good results, but not consistently. I've also had to add logic to vary the pre-image processing by model and ios version as there are subtle differences that throws off the ocr. Has any worked out a way to do this in a consistent way? -- 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 visit https://groups.google.com/d/msgid/tesseract-ocr/af1fd2dc-7bbf-4f1b-ae31-0966caba3d9an%40googlegroups.com.
