Hello, I am running into real difficulty getting Tesseract to work with a faux segmented display.
At the moment I am processing a video frame by frame, removing the background and replacing the character's color to leave images like the attached. I have spent a long time training both using a set of the actual images produced by the above, as well as using tifs of LCD fonts that match, setting the expected format, whitelisting the characters expected and still am seeing no improvement in the recognition success - at best it is around 30% success, at worst 0%. I have also tried using SSOCR (https://www.unix-ag.uni-kl.de/~auerswal/ssocr/) without any success (0 digits recognized from the above), and exploring all the different settings of tesseract and SSOCR. Is there any advice for getting these characters recognized consistently? My target is to have the characters recognized with at least a 75% success rate. -- 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 tesseract-ocr+unsubscr...@googlegroups.com. To post to this group, send email to tesseract-ocr@googlegroups.com. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/73f12a10-45d4-4879-9d62-456dd5dd3abf%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
