GUIDE深度 · Deep dive
PDF Text Extraction vs OCR
Digitally generated PDFs have a text layer and extract directly; scanned/photographed PDFs are images and need OCR. OCR misreads numbers/names/tables, so proofread important documents. Check a few known values before trusting the whole extraction.
Selectable text means a text layer.
Only image selection means a scan.
Clean scans and good OCR do better.
Tables/handwriting/low resolution do worse.
FAQ
Are tables preserved?
Usually not; output follows text flow.
Can I batch?
One PDF for now; batch may come later.