Scannable Docs - Can you get the text out?

[calbloke 2]

I understand that you can search for words in the scanned doc so clearly OCR is involved. 

My question is whether you get the text out of a scanned image?

[gazumped 12,074]

Hi.  No.  Scanned images include some fuzzy logic so 'house' might also be listed as 'horse' or 'hose' - there's no narrative of continuous text,  just a tree index of possible interpretations.  It is (mostly) possible to extract the text from images through third party apps (an internet search menioning your OS will suggest a few) and paste that with an image to create a better index.

[PinkElephant 8,834]

EN will not modify the document uploaded by you.

The text information created by EN OCR is in a second pdf. This is invisibility stored in the note, not accessible for the user.

[laurence.glazier 147]

If you are running Windows, PowerToys can do this - a shortcut key can copy text from an image.

[PinkElephant 8,834]

On most Macs to select and copy text from pictures is a native function of MacOS, without additional tools. 

To create searchable text in PDFs OwlOCR is a nice tool. It can even nest itself into the right click mouse menu.

[bmcl26 578]

57 minutes ago, gazumped said:

It is (mostly) possible to extract the text from images through third party apps (an internet search menioning your OS will suggest a few) and paste that with an image to create a better index.

I use Snagit for this, the latest version has a widget hidden under the top of the PC Screen which you can hover the mouse over to make it visible, and it can be programmed for up to four functions.  The function I use very often is "Grab Text" It only takes a few seconds to process the text grabbed from images and can paste it into where it is required.

[calbloke 2]

Thanks to everyone who replied.  I use a Mac and was unaware of owlocr and will try out the evaluation.  

For the record, I found Google Keep does a nice job of extracting the text via OCR.