I am working on extracting scoreboard data from game images in real-time, with high accuracy and speed. The template and language (English) stay the same, but fonts may vary (ignore the one Chinese name).
I annotated the data in Roboflow, trained a YOLO model, and achieved 99% detection accuracy. For text extraction, I used Tesseract-OCR with some preprocessing, the results are good but it still misses some digits. What can i do to increase the accuracy of the extraction? -- 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/35317a99-4b96-4475-9eb9-33d65c4afcafn%40googlegroups.com.
