Researchers from MIT in Massachusetts can decipher, with the aid of terahertz waves and an intelligent algorithm for character recognition, the first nine pages of a stack of paper. Radiation in the terahertz range has a number of advantages over X-rays, for example to ability to distinguish between ink and paper, which cannot be done with X-rays.
 
The system uses minuscule air pockets of about 20 µm deep that form between the sheets of paper. The difference in refraction between air and paper indicates a separation between these two and a fraction of the terahertz radiation is reflected back to the sensor. In an experimental setup a 'standard' terahertz camera transmits ultra-short pulses of radiation. The reflected radiation is picked up by the built-in sensor. The time between transmission and reception is a measure of the distance between the individual pages.

The biggest problem at the moment is the noise that is caused by that part of the radiation that is not immediately reflected back, but first bounces back and forth between the pages, before moving towards the sensor. In addition the sensor itself also has its own 'background noise'.


 
 

More information: http://news.mit.edu/2016/computational-imaging-method-reads-closed-books-0909.