I have a 16 page TIFF document which is always rotated sizewise.

The recognition confidence is very low at 55% although the document is perfectly readable. That is api.Recognize(); does not work very well.

So what does Tesseract do here. Must we insist document are orientated vertically and can it handle multiple pages?


Can you help with some example code to both use OcrInvert, OcrRotate and ocr pageiterator ?

Edited by user Friday, November 24, 2023 6:39:34 AM(UTC)  | Reason: Not specified