(Use this mode for recognizing full-page documents.)
(Best for single elements like tables or formulas.)
(Detects and locates specific elements in the image.)
No visualization yet.
