Hello there, I'm trying to implement OCR for license plate recognition. So far, I'm able to cut out each character individually using OpenCV and put it through Tesseract. I uploaded an image with my result after detection. About recognition, well, this is my first time using Tesseract so I'm simply selecting -psm 10 so each image is treated like a single character.
Most chars are being recognized correctly, but the "*W*", in *2.jpg*, is returned as "*H*" and also the zero "*0*", in *6.jpg* returned as "*l*" (low-case L). I suppose that the problem with the zero is more easy-going since I can assume that the last fours images contain numbers and Tesseract might get it right under this assumption, is that right? However, I'm really lost on the W problem. Maybe I could design some neural network to help me out, but I believe Tesseract can do this, I just don't know how yet. Can anybody help? :) Thanks in advance! Arthur -- 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 post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/c41b8b82-55dd-4c8b-ac59-47dbf8d28a53%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
