Hey all, just wondering, does Okular use Optical Character Recognition?

I've been fidling with it, I took an image with text on it, put it into a PDF...and Okular can read it. I'm quite suprised by it, actually. I was wondering if I was correct, or if there was some other magical system in effect here?

I've done a few more experiments, I put the image in at an angle, the words 'knowledge plane' came out like this on the clipboard;

K w no led lan eP g e


So it seems it DOES use some sort of OCR/pattern recognition.

Dante Ashton wrote:Hey all, just wondering, does Okular use Optical Character Recognition?

It does not.

pinotree wrote:

Dante Ashton wrote:Hey all, just wondering, does Okular use Optical Character Recognition?

It does not.

Would be nice though.

Any plans to implement it?

pinotree wrote:

Dante Ashton wrote:Hey all, just wondering, does Okular use Optical Character Recognition?

It does not.

I had that working as well. It's a bit mysterious which component does the OCR -- maybe some scanners do it when scanning to PDF?

scummos wrote:

pinotree wrote:

Dante Ashton wrote:Hey all, just wondering, does Okular use Optical Character Recognition?

It does not.

I had that working as well. It's a bit mysterious which component does the OCR -- maybe some scanners do it when scanning to PDF?

Some PDF do include transparent(?) OCR text on top of the scanned image. But that’s up to the PDF creator not the reader.

It would be great feature if Okular could OCR a selection – e.g. when the user draws a rectangle with the selection tool, it would run an OCR program over that area/image and provide an option to copy that text to clipboard instead of image.