OCR (optical character recognition) has also been widely used in recent years in one particular area, that of digital preservation of old texts. Very often books published before a certain time, and all the data and information contained in them, are not digitally available and therefore cannot be consulted, for example, on the Internet. The transformation of these archives into digital data is especially important for those books that are no longer accessible over time.
The lack of accessibility also proves to be a problem for character recognition software, so much so that very often these books, especially the oldest and most delicate ones, cannot even be opened or the inside pages damaged. In this regard, a group of researchers at MIT has developed a special technology that uses microwave and infrared radiation to read the inside pages without having to open the cover of the pages themselves.
The technique, described in the last issue of Nature Communications, is currently able to correctly identify letters and words of only a few sheets (nine to be precise) but the researchers promise to improve the same technology to go deeper. The technology could be of particular interest to those museums and institutions that want to examine ancient books whose paper has become too weak to be leafed through or even touched.
The system uses various algorithms that capture images from single sheets in stacks of paper, algorithms developed by researchers at the Massachusetts Institute of Technology and Georgia Tech. The system uses a terahertz rate that can penetrate surfaces and distinguish between ink and white paper, something that not even X-rays can do. This radiation is emitted in short bursts and the system can measure the difference between the time of emission and the time when the radiation is reflected back to the sensor.
The same system takes advantage of the fact that between the very pages of a book there are tiny air pockets about 20 micrometers deep. The difference in the refractive index between this same area and the paper allows the difference to be detected.
Latest posts by Clive Sullivan (see all)
- New technology can read pages of a closed book with radiation - February 2, 2020
- New family of moths discovered in Kazakhstan - January 19, 2020
- Glyphosate linked to non-alcoholic liver steatosis according to study - January 11, 2020