OCR Software

I admired the Rennaissance people very much. Smart and learnt people like Leonardo da Vinci or Michelangelo were experts in so many fields: for example Leonardo was a painter and a sculptor, a scientist and inventor, a fine observer of the human body and many, many other things. Nowadays our field of activity has narrowed a lot, but not because we are more limited or because we are less intelligent. Quite the opposite: the average IQ has increased tremendously in the past decades. I think it is all about science and technology development. There are so many new things in every field of activity that you couldn't possibly keep up to date with, let's say 10% of all this information. Whereas in the Middle Ages all their science and literature could be kept in one large college library. But these technological innovations and discoveries never cease to amaze me, as my constant question I ask myself is: What more could they possibly invent next? And usually the answer to my question comes every month in another field of activity.

I was amazed when I say the first scanner. But I got even more amazed when I saw that all the images that I got when I scanned some documents could actually be "read" by the computer and transformed in Word documents that I could edit afterwards. And I immediately thought of all the ancient books and documents transmitted from generation to generation. It took so long to type those work of arts and it took even longer to edit them in electronic format,as somebody had to type all those bushy volumes the size of a tombstone. Well, now they can be scanned and then edited and saved in electronic and virtual libraries and anyone can have access to the original text.



That is in short what OCR software does. As the name suggests it, OCR means optical character reading, that is a computer software that "reads" the scanned document and recognizes the text in the scanned picture and does not treat it like a photo. It is quite impressive, but the document still needs editing after that, because the software performs automatic reading and some characters might be mistaken for other symbols sometimes, due to the quality of the copy or of the software.

The first person who invented and patented OCR was a German, Gustav Tauschek in 1929. However, a similar invention was patented by handel in 1933 in USA. Anyway, the basic principle of Tauschek's device was a mechanical machine that used templated. A photodetector lit the exact spot where a character appeared. If the template and the character, having the same size, aligned there was no light passing by them. It meant that the identification of the character was made.



The Postal service in the USA was among the first institutions to use OCR technique, in this case for sorting letters.

Nowadays there are customized OCR software packages, offering different features for each language, including Arabic and Chinese. I told you science and technology are amazing!

Some of the best OCR software seem to be : ABBYY Fine Reader Professional 9.0, OmniPage Professional16, Type Reader 2008, Adobe Acrobat 9 Pro Extended.If you are interested and want to know some other OCR software names visit the following web pages: www.cnet.com, www.pcmag.com,  www.zdnet.com.

All OCR software have improved in time and now they are all very good. However, there are some features that differentiate them and that is why some cost more and some cost less. The main differences are connected to the following features (their presence, absence or degree): the degree of accuracy in character recognition, the degree of accuracy in page reconstruction, the support for as many languages as possible (of course, the ones with more languages has a higher price), user interface, speed and support for searchable PDF output.



According to these criteria ABBYY FineReader was considered the best average OCR software, while Read IRIS was considered the best solution for the least money.

So ABBYY FineReader is considered to be the best OCR at the moment. Since all OCR doftware are highly accurate , it means that other additional feature brought the title. Let's find out what are these additional features that distinguish ABBYY from the other OCR software. It has many fine tools that improve its recognition of characters on all kinds of original texts, especially digital photos or photo copies of books and even PDF files. When using Abbyy FineReader the screen is divided into three main parts: the first on the left shows thumbnails of the pages that are scanned, the page in work is in the middle of the screen and the right part belongs to the text that was recognized and you are about to edit. You have the option of magnifying the text and edit it there, in a special panel or transfer it to Word and edit it later. There are predefined procedures to choose from when you want the software to perform another operation and this saves you a lot of time.



Another special feature of Abbyy checks the PDF files to see if they have embedded text. According to the degree of accuracy of the text, the software will take a shorter or a longer time to descyphre it, but eventually it will. Another interesting and particular feature is the possibility to operate a few changes on the image of a scanned book. You can split the image in two and make the curves of the original page return back to the rectangular shape of a book page. This important feature allows Abbyy to improve its degree of accuracy quite a lot.

OK, we have talked only about scanned images so far. Now we will talk about photos made with the digital camera. You know that if you take a photo of a book or newspaper with your digital camera, the result will not be the best and you can hardly read the text in the photo. Well, apparently Abbyy FineReader is indeed such a "fine reader" because it can isolate the text directly on these photos. I guess this ability makes it such a remarkable piece of work in OCR software.