PDF recognition often misses several pages.

#33
by xldistance - opened

nanonets_ocr doesn't have this problem

            images = list(convert_from_path(
                pdf_path,
                poppler_path=r"H:\\Langchain-Chatchat0.3\\poppler-24.08.0\\Library\\bin",
            ))

The image has already been converted to the RGB mode.

Sign up or log in to comment