We blogged the other day about Google's new document imaging initiative, which promises to make "hard" PDF documents searchable on the World Wide Web. The technology being used to accomplish this feat, we noted, is called Optical Character Recognition, or OCR.
OCR is one of those technologies that has a definite cool factor. Through OCR software, a computer can "read" a hard copy document and render it into bits and bytes--in short, make a paper document digital.
At that point, sharing and searching that document becomes much easier, because the digital world is not dependent on things like planes, trains, and automobiles. You want something, it's at your fingertips.
But is OCR really ready for widespread corporate use? For example, say you run a law firm, and you are interested in implementing a document imaging solution, but your files are hard for a non-lawyer human being to comprehend, let alone a non-lawyer computer.
Can you rely on OCR technology to accurately transport your paperwork to your computer accurately and completely?
The answer is, yes but. Depending on your provider (so check around), OCR software has bugs and troubles just like any other software. Nevertheless, accuracy rates on typed words are 99%.
Cursive writing, however, is still being figured out.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment