Re: [tesseract-ocr] inaccuracy in plane text

2023-12-22 Thread Ger Hobbelt
Couple of things to check/test: - tesseract expects black text (lettering) on white background: that's what is has been trained on and that's what will work best. Hence: try to convert anything to look like that before feeding it to tesseract. - tesseract was trained on text, if I recall correctl

Re: [tesseract-ocr] Numbers detection

2023-12-22 Thread Ger Hobbelt
On Thu, 21 Dec 2023, 15:22 Art Rhyno, wrote: > If > Important extra note (as I see a new image that's white text on black background): Tesseract was trained on black text on white background, targeting books, publications and academic papers' OCRing. To improve your chances, ALWAYS make sure y