Image Recognition Procedures for Extracting Images

[Padanges]

Hi,
I have many PDF documents with scanned pages. Images mode can't extract images from such pages (where whole page is an image itself). I was wondering, is there a way to use image recognition procedures to extract images from scanned pages? I suppose such procedures are common for DSP packages, so it would be possible to use them for our purposes. But can we mess with image extraction using Custom Scripts?


Regards

[RTT]

Image segmentation is something the OCR library, that PDFE uses for the quick OCR assistant, can do, so it would be possible to achieve this functionality.
But I don't see many users needing such service, especially from a script interface, to worth the effort to expose the functions to the script interface.
Maybe a feature more suitable to the extract images tool, where the user can manually enable the image segmentation, when trying to extract images from such type of scanned documents.