More than 14 million scanned book pages appear online

Millions of scanned pages of books appeared on the Flickr resource, which the American organization Internet Archive posted for free use. At the moment, about 3 million images have already been published and this is only the beginning, the organizers of the project say that in the near future more than 14 million pages of scanned text will be in free use.

The implementation of such a large-scale project was made possible only thanks to the recently created algorithm, which automatically extracts images from more than 2 million books, which are currently stored in the public library of the Internet Archive organization. This program, developed by Kalev Litaru, a researcher at Georgetown University, is configured to automatically recognize text and extract 500 words from each scanned image, which allows text search through the collection.

It is noteworthy that all images of texts posted on behalf of Internet Archive will not be protected by copyright, which will allow their free distribution throughout the Internet. The creators of the project believe that over time they will be able to create a visual library in which information will be searched by reading images.

To date, the organization Internet Archive Book Images is included in the general project The Commons, whose goal is to create open libraries to preserve the heritage of the Internet. The service already has special subcategories for audio, video and graphic materials, outdated versions of sites, console games of the 1970s and 80s, as well as all kinds of software. The information stored in the Internet Archive is already 19 Pb and includes more than 600 million scanned texts of books of various genres and directions written over the past 500 years.

Svetlana Korableva © Gallerix.ru