Abbyy Finereader 8

The paperless office is a great idea, but if you have large amounts of paper-based records needing to be converted into electronic format, things can start to get tricky. Abbyy Finereader 8 Professional can greatly reduce the time it takes to capture printed material and convert pdfs.

As you would expect, this new version improves on accuracy and speed, but there are also several additions. A fast recognition mode has been introduced along with support for capturing text from digital photographs. It is also far more accurate when capturing text with different font sizes on one line.

The package is organised into three main sections, each with their own large icon within the main interface. A wizard is also on hand for hassle-free and quick optical character recognition (OCR).

A column down the left-hand side of the screen shows all the pages that have been added to the project while icons on each page show whether it has been recognised and saved.

After the OCR process, words that need attention are highlighted in different colours for characters that Finereader is unsure of and words that are not in the dictionary. One extremely useful addition to Finereader 8 is the ability to use the custom dictionary from Word on the computer; something anyone with many job-specific words will appreciate.

The ability to recognise text directly from digital photographs will be of real interest to anyone doing research where photocopying is not an option, such as genealogists working in libraries where photocopying of records is not permitted.

Abbyy recommends a 5megapixel camera is used, with 2megapixels being the absolute minimum. In our tests, the latter was adequate for a single page at a time, as long as there was no particularly small text.

Try taking a photo of a page of text and you’ll probably find the lines of text appear distorted due to the shape of the lens and angle of the shot. Finereader includes image manipulation routines to straighten these lines, which in turn improves OCR accuracy.

The pdf tool is another new feature. Although it might sound odd to load a pdf into OCR software, it allows you to edit documents that have been locked and reduces the need to use Adobe Acrobat. Finereader attempts to use the best of both worlds by scanning a pdf as if it was an image and accessing the text of the document.

Other time-saving tools include recognising both web and internal pdf links within documents and saving them into the finished document. The fast scan takes less time at the cost of accuracy.

The idea is that the fuzzy search capabilities of tools such as Google Desktop remove the need for absolute accuracy. However, if the content is important it is probably better to recognise the text properly to begin with.

Jobs with Finereader can be automated in a similar way to creating Word macros. This comes into its own if several output formats are required. The disadvantage is that you are at the mercy of the automated text box layout; performing such a scan of the Abbyy press release resulted in some of the images being in text boxes with undesirable effects.

Finereader is not cheap but it is a very comprehensive OCR suite that gets the job done quickly and with the minimum of fuss. New tools such as the digital camera options are impressive and give it wide appeal.

System Requirements:

• 500MHz processor
• 128MB of Ram
• 350MB hard disk space
• Windows 2000/XP/Server 2004