OCR

From MozillaWiki
Jump to: navigation, search

Right mouse button drag and release allows Psuedo Optical Character Recognition option for universal Cut feature of cut and paste.

Detail.

When browsing user shall be able to use mouse and right mouse button and a drag and release action to select any area of the screen from (X1, Y1) to (X2, Y2). (A simple right button push and release without the drag action would still bring up the right button menu.)

A visible box is drawn around the area selected.

If the user then uses the short cut [ctrl]-c or menu item Edit > Copy, the bit map image of the selected area is fed to Character recognition software which would do it's best to extract ASCII text and formatting details from the bit map to the scratch-pad from where it can be pasted into search fields as per normal.

The reason I would like to see this feature is that there have been a number of times I have seen a term that I wanted to paste into a search field or other text input field, but hey, guess what the word is on the screen as part of a picture or other unselectable text so I have to grab a pen write it down then type it into the field with the attendant risks of a typo or other error.

I'm not enough of an expert to say if this cannot be done because of the underlying Operating System. Perhaps someone can add that. Perhaps this should be something added LINUX.

Optionally, if the character recognition software returns a very low probability that any text has been captured, the area could remain a bit map capture in scratch pad that could be pasted into graphics processing software such as GIMP.

I must admit I don't know if there is any public domain character recognition software out there and I don't imagine it would be easy to create from scratch. Others may be able to lend details to this point.