Notes2Self.net

Stephen McGibbon's Web Journal

IRIS announces innovative document conversion solution.

Have you recently scanned a document and run Optical Character Recognition on it? The results these days are very impressive.

irisA few months ago I visited the office's in Belgium of one of the leaders in this field, I.R.I.S. (Image Recognition Integrated Systems), and met with Pierre De Muelenaere, Group President and CEO. Pierre's been an innovator in OCR technology from the 80's when he completed his PhD thesis and set up a company to put his ideas into silicon. The picture shows Texiris 1.0, the first commercial product of iris-1988-1the company, composed of a dedicated image processor that can be inserted in an IBM PC and a downloadable software.

Watching the way I.R.I.S.' software worked it struck me that they also had effectively developed a novel approach to document translation. Instead of trying to read what the file says about how the document should look this approach essentially creates an image of the page and analyses the image to figure out how the document should look when translated to a different format.

Pierre agreed, and today I.R.I.S. announced that we're cooperating together to provide document conversion solutions from a wide variety of electronic document formats to OpenXML.

Specifically, a future version of I.R.I.S. OCR and document compression server will be capable of reading more than 75 different file formats and convert to over 25 file format including: ODF, OpenXML, WordPerfect, WordML,SpreadsheetML, PDF, XPS, HTML, Jpeg, jpeg 2000, HD‐Photo,…

My friend and colleague, Bruno Schröder, (National Technology Officer, Microsoft Belux) is quoted saying:

“ I.R.I.S. technology brings an extremely promising new solution to the translation of documents between different file formats. It is very robust and deliver a very faithful representation of the document, independently of the complexity of the page and the complexity of the XML.”

It's hard to picture, so I will try to stop by I.R.I.S. next time I'm nearby and see if I can make a quick video with Pierre to give an overview of how this works. It's very cool.

Comments

Doug Mahugh said:

As we near the end of the standards process for DIS29500, the final positions of various countries are

# March 24, 2008 7:29 PM

Notes2Self.net said:

Today is " Document Freedom Day ". The EU's IDABC "Open Source Observatory" observed back in February

# March 26, 2008 1:49 AM