Readiris Corporate 12 ($399 direct) converts documents from scanned images into standard file formats with the least fuss (and the fewest options) of any major OCR program. In corporate and other high-volume settings where you don’t want to spend time correcting text inside your OCR software, but simply want to get a printed page or an image file converted into a Word document or PDF file, Readiris’ streamlined approach may be suitable. But if you want to do much fine-tuning of output, Readiris isn’t the app for you.
And that’s not the only drawback to this app. In my informal tests, Readiris lags slightly behind both ABBYY Finereader 10 Professional Edition and Omnipage Professional 17 in recognition accuracy. Readiris also can’t match Finereader in disentangling complex page layouts—Readiris sometimes didn’t even recognize that a scanned-in table was a table, and not a set of random text-boxes.
Readiris comes in Pro and Corporate flavors; both have the same basic document-reading capabilities, but the Corporate version (which I tested) adds the ability to perform OCR automatically on files dropped into a “watched folder,” includes a feature that lets you create an XML-formatted document index, create encrypted PDFs, and streamlines processing of multiple documents. Like its corporate-level rival products, it supports bar code processing, and can automatically route documents to specified folders if the document has a bar code that includes a specified number. Readiris can optionally create an XML-based index of these documents, but, unlike its rivals, it won’t let you fine-tune the kind of processing it performs when it finds specific numbers in bar codes.
Readiris in Action
When you start Readiris, it optionally launches an OCR Wizard that walks you through the steps of getting a document from your scanner into whatever format you specify. Output formats include multiple varieties of PDF, Microsoft Word, and OpenOffice.org formats. I was puzzled for a moment by options to save to Netscape and Mozilla formats, but those options turned out to create standard HTML files. If you don’t use the OCR Wizard, you can choose from a list of standard tasks listed in the background of the main editing window. These tasks include scanning to Word, OpenOffice, or Excel, and saving to PDF and Microsoft’s XPS document-storage format.
When I chose from the list of standard tasks, Readiris ran automatically, and the only things I needed to do were select an image file to read (or click a few buttons in a dialog to tell the program to read from my scanner) and give a name to the output file. When I decided to perform a manual OCR operation—because the input material had a complicated layout that I knew was likely to cause problems in any automated operation, or because I wanted only to apply OCR to one part of a page—the whole process was equally straightforward. I like the way I could choose input and output options from a toolbar on the left of the screen. I also liked the way a smaller toolbar at the right of the screen let me second-guess the app’s interpretation of my documents’ page layout. I could combine multiple text regions into one, fix the app’s occasional mistakes in identifying text and image regions, and click a button to de-skew a distorted image.
I was less happy about the total absence of any text-editing features inside Readiris. With Finereader and Omnipage, I could open a text-editor window that displayed the scanned image in one pane and the text that the app extracted from the image in another pane, and I could type in manual corrections wherever the app misread some text or punctuation. In Readiris, I couldn’t even see the text that the app read from my document until I saved the text to another format. If I then wanted to make corrections, I had to open the saved text in Word (or whichever other editor I was using), and then find the corresponding part of the scanned image in Readiris. As you can guess, this isn’t an easy process, and I doubt I would want to use Readiris to recognize text in documents that would need extensive corrections.
One of my test documents for OCR software is a PDF file of a booklet, with the PDF formatted so that each PDF “page” is a spread of two facing pages from the printed booklet. Readiris insisted on reading each spread of facing pages as if it were a single page with two columns, and I couldn’t force the program to the split the two-page image into two separate pages. In contrast, Finereader correctly split each two-page spread into two separate pages, and didn’t need any help from me to do so. I was surprised by Readiris’ inability to handle a situation as simple and common as two facing pages on a single image.
Readiris also surprised me with an annoying interface glitch. The File/Open dialog, for example, can’t be resized, so you get only a tiny window near the top of the dialog for viewing the file list—and the limited space in that window is even more annoying when you have it set to display thumbnails of your image files.
Readiris gets basic OCR jobs done at a significantly lower cost than the competition, but it falls short in complex tasks, and completely lacks the text-editing features built into the competition. If you want low-priced, no-frills, automated OCR, Readiris isn’t a bad choice.