A brief history of OCR: the technology inside your ScanMarker

A brief history of OCR: the technology inside your ScanMarker

Both the ScanMarker and the ScanMarker Air are types of OCR device. These multi-functional productivity tools are on the cutting edge of modern technology, providing a range of benefits to people all over the world. We have been contacted by students taking study notes, lawyers preparing legal briefs, doctors transcribing patient files, and a whole slew journalists, bloggers, and historians taking accurate quotes from original sources with their ScanMarker. More than that, we have received thanks from people with learning disabilities, people who are visually-impaired, people who are learning a foreign language, and parents of children with special educational needs for the assistance our device provides in their day-to-day lives.

But how did we get here? What is OCR technology, and where did it come from?

A brief explanation

OCR is short for optical character recognition. It describes the process of converting physical text – that is, a printed page in a book, newspaper, or magazine – into electronic data. That data can then be used in a variety of different ways. The ScanMarker, for instance, converts that data into audio so it can read text back to you. It can also translate it into different languages, copy it into documents, and much more besides. OCR is a remarkable blend of computer vision, pattern recognition, and artificial intelligence. The device is literally seeing the text through its scanner, recognizing the shapes of individual letters, and then understanding what those shapes mean as characters in their own right, as well as in conjunction with the other letters around them.

Where it all began

Surprisingly, OCR is a pre-War invention and by pre-War, we’re not talking about World War II – we’re talking about the Great War – World War I. For something that we imagine to be a modern innovation, OCR’s roots go back as far as 1914. In that year two men, Emanuel Goldberg and Edmund Fournier d’Albe developed, independently of one another, the first OCR devices. Goldberg invented a machine capable of reading characters and converting them into telegraph code. Meanwhile, d’Albe created a device known as the Optophone. This was a handheld scanner that could be moved across a page of printed text, producing distinct and separate tones, each of which corresponded to a certain character or letter. While this sounds more of a novelty than a serious OCR tool; it’s not impossible to imagine a sister device that could hear those tones being played in order and reconstitute the words in a new medium. Advances in technology aside, that is essentially how OCR works today.

Bigger and better

Goldberg continued to improve on this latent OCR technology, spending the years between the Wars developing a “Statistical Machine.” This device was, in effect, the world’s first search engine, and used OCR to search microfilm archives for particular patterns of characters. He was granted a patent in the USA for this invention, a patent that was letter acquired by none other than computer giant IBM.

Expanding the field

Early OCR devices had their limitations, of course. Specifically, they were only capable of recognizing perfect text – that is, text that was perfectly straight, perfectly clear, and printed in the single font that these devices were programmed to recognize. The machines were basically comparing the character that they scanned to a database of characters to see if they could find a match. With the limitations of computer memory being what they were, these databases were small by necessity and offered no wriggle room for users. Either the font matched exactly, or the reader failed to recognize it.

As computer processing power increased in the sixties and seventies, omni-font OCR readers became the norm. These scanners were not looking for an exact match. Instead, they were beginning to recognize general form and shape, despite differences in the font design.

Entering the modern age

OCR devices as we know it today undoubtedly began with Kurzweil Computer Products, Inc, founded in 1974. Another omni-font device, Kurzweil came up with an application for the hardware that was frankly revolutionary. His idea was to develop a reading machine for the blind, one that would allow the visually impaired to have a book read out to them by a computer. It was another invention ahead of its time and certainly beyond the budget of the paying public. However, not only was the machine itself some serious next-generation thinking, but it also produced a pair of technologies that have become an integral part of most OCR systems. These were the text-to-speech synthesiser and the flatbed scanner.

An ever-changing electronic world

In the 21st century, OCR really came into its own. By pairing it up with internet technology, the possible applications of OCR became a reality. Character recognition became broader, with most fonts accounted for, and the flatbed scanner gave way to the handheld device. As the recognition algorithms became more sophisticated and optical scanners started handling higher and higher resolutions, exciting new uses for the tech became available. Ever been stuck in a foreign country with signs in a language you don’t understand? Now you could just take a photo, process it into an OCR app and it will recognize the words, the language, and be able to translate back to you.

This kind of instant verification software has revolutionised the way we do business. They can recognize invoices and receipts, passports, car number plates, insurance documents, and so much more, and then know what to do with that information.

Sites like Project Gutenberg have been using the improved accuracy of OCR technology to scan old texts currently in the public domain, in order to produce accurate and complete electronic versions of the classics, which they give away for free. Google Books went a step further, allowing its users to search for words and phrases within the original scanned image, not just the electronic data.

With so many uses worldwide, OCR is one of those technologies that nearly everybody uses, but which nobody is talking about, and that’s a shame.

Where to next?

It’s been quite the journey for OCR, from the musical book-reading device of 1914 to the myriad of applications that use the technology today. Here at ScanMarker, we are always striving to improve the quality and functionality of our own OCR devices – the ScanMarker and the ScanMarker Air – consistently putting ourselves at the forefront of the field. Why not be a part of that journey yourself? Pick up a ScanMarker today and see how you can change your world.

Back to blog

Leave a comment

Please note, comments need to be approved before they are published.