The main part of the SCRIBO project is based on text processing and semantic analysis. Text can easily be extracted from many kind of documents except images. That's why SCRIBO provides also image processing tools to detect, locate and extract text from images.

The image processing part of the SCRIBO project is based on Milena, a generic C++ image processing library developed by the LRDE. During the project, several toolchains have been developed: Text detection in natural images, text extraction from scanned document and page segmentation.

Some of them are now available online! Read more.

...

The Scribo project will take part in the Open World Forum, the leading global summit bringing together decision-makers from across the world to discuss the technological, financial and social impact of open technologies. On October 1,2 2009, the Open World Forum 2009 will bring together all the key players in the open source ecosystem to cross-fertilize initiatives for innovation and economic growth.

On this occasion, the Scribo team is taking part in a conference about the benefits of semantic technologies on the desktop. Professor Stefan Decker, director of the Digital Enterprise Research Institute will give a ...

In the Scribo interview, Stéphane explains really well what we are looking at with Scribo. Let's sum it up again.

We are handling too many informations (documents, mails, web pages, ...). The desktop is just providing fairly isolated tools to handle these informations. The user is doing the link between all these informations. And there are more and more informations but the 100 000 000 000 of neurons in my head (and 1 000 000 000 000 000 synapses) are decreasing every day (and yes, we use 100% of them). And the desktop does not help us. It does not make the link between the mails, the document and the urls. We are using an application oriented desktop.

What we are developing, is a task oriented desktop. We want to provide a way for a user to handle at once all informations related to a given "task" (or context whatever you prefer). You want to work on the project FutureWorld ...

The 2009 convention of the System@tic Competitiveness Cluster took place in Ecole Centrale Paris at the beginning of the summer. Stéphane Laurière from Mandriva was invited to present the objectives of the Scribo project. In this interview (in French), he emphasizes the disruptive change that the semantic technologies are bringing to the desktop. The natural language processing algorithms developed within Scribo will let you get a high-level contextual view of the information you deal with daily. Note that you can get a preview of the combined Scribo and Nepomuk technologies by installing the Nepomuk extras package on ...

Scribo is bringing natural language processing to the Nepomuk KDE semantic desktop. Sebastian Trüg, lead Nepomuk KDE architect, presents the current developments and integration status on the following blog entry: Scribo ? Getting Natural Language Into The Mix.

The SCRIBO Consortium took part in the 3d System@tic convention, June 18 2008, at Ecole Polytechnique, Palaiseau. The free software System@tic thematic group was represented by 4 projects:

• Couverture
• Helios
• Squale
• SCRIBO

A few slides and a demo were presented to a delegation from Paris Région, Direction Générale des Entreprises, OSEO: SCRIBO slides (in French).