The text displayed alongside the article image has been created through a process called Optical Character Recognition (OCR). This is when a computer software program converts words in an image file into text. The software OCR eye is less accurate than a human eye and reads character by character, so words that appear clear in the page image to a reader may sometimes be converted incorrectly and appear as odd sets of characters. Any user is able to correct these characters and words so they match what is in the image.
You may correct the OCR text of any line of this article by double-clicking the OCR'ed text needing correction.
The page image window will be positioned to show the line of text you are editing at the top left, with the first word of the line highlighted in green.
Press F12 to rapidly move to and highlight the next word in the text - this can make correction much faster!
To move to the next line of OCR'ed text, just press Enter.
To discard your corrections to a line, press the discard button at the start of each line of edited text.
When you are finished, press the "Save OCR changes" button which appears once you start correcting, or just navigate away from the page without saving
to discard all changes.
It is very important that you correct the text line by line, that is, that the original line boundaries are
preserved. The system needs to be able to correlate your corrections with the originally OCR'ed words, so that phrase
searching and text highlighting will continue to work on your corrected
text. Hence, make you corrections line by line, matching the lines of text in the page image.