You must have struggled many-a-times reading a friend's messy handwriting or copying someone's notes! Have you ever wondered if it is possible to read it precisely through any device? With so much of our lives computerized, it's important that machines and computers understand one another and pass information back and forth. Thats when OCR comes into picture.
OCR stands for Optical Character Recognition. It can convert images of typed, handwritten and printed text into machine encoded text, whether from a scanned document, a photo of document, text on billboards in a landscape photo or from a subtitle text superimposed on an image. In other words, its similar to producing something like a TXT or DOC file from a scanned JPG of a printed or a handwritten page. It avoids the need to retype already printed material for data entry. Also, once a printed page is in its machine readable text form, you can do all kinds of things you couldn't do before like searching through it by keyword, editing it with a word processor, compressing it into a ZIP file and much more.
Earlier such softwares were not much efficient. Back in 1970’s, one of the first major uses of OCR was in a photocopier like device, which could read printed books out loud to blind people.With technological advancements, OCR software has improved in the last few years. Today, it can recognize characters, words and sentences with hardly any mistakes, making the software quite accurate. Also, most OCR software has a high rate of recognition, but documents containing images along with texts may still not be converted completely. Still commercial OCR nowadays have 81% to 99% accuracy. Here 99% accuracy means 1 character in 100 characters is uncertain, which can be correct or incorrect. Thus if further improved OCR can really be a revolution in exactation.