|
|
 | |
 | |
 | |
| |
 | |
| |
 | |
| |
 | |
| |
 | |
| |
 | |
| |
 | |
| |
 | |
| |
 | |
| |
 | |
| |
 | |
 |
Production Imaging The entire Archimedes Palimpsest has been imaged using one particular technique, which was designed in response to the failed experiments of the year 2000. It was clear from these experiments that images had to be taken at a higher resolution, that post-processing needed to be simplified, and that the problem was not to make the prayer book text disappear, but rather to clearly differentiate it from the Archimedes text underneath, and to make the Archimedes text itself more visible. Moreover, since the whole manuscript needed to be imaged in this way, the technology had to be simple and effective, both in terms of cost and in terms of time. The imagers, Roger Easton, Keith Knox, and William Christens-Barry, devised an apparatus for imaging the Palimpsest on an XY stage.  The imagers decided to combine two different wavelengths of light. The first of these was in the visible spectrum. They noted that while the Archimedes text was extremely faint in RGB light, it disappeared almost completely when viewed through the Red channel of RGB light. This does not sound useful but it will in a minute.  The second wavelength they used was the blue channel from Ultra-violet florescent light. Again, this needs a little explaining. Ultra-violet is invisible to the eye, but if it is directed upon certain substances, these substances are energized, and emit light in the blue part of the visible spectrum. Specifically, when Ultra-violet light is directed against parchment, the parchment will give off blue light. However, the Ultraviolet light does not energize the ink on the surface of the parchment. The result is that if one looks at texts written on parchment under UV light, the letters stand out much better and appear to be “backlit”. Many faint texts are successfully read using UV light. UV light alone works well on the Palimpsest, but there are some problems. The first is that UV light is somewhat “soft”, or blurry at the edges. The second and more important difference is that the under-text appears exactly the same as the over text, making it very difficult to differentiate between the two when they overlap.  The solution that the imagers came up with was to combine an image taken in the Red channel of RGB light with an image of the blue channel of Ultra-violet fluorescent light. The three principle components that the imagers were trying to separate will now come out as different colors. The parchment will come out “white”, because it reflects both red light and blue light, the prayer book will appear black, because it absorbs both red light and blue light, and the under text will come out red, because it absorbs the blue light but reflects the red light.  The results were astonishingly successful, and this is the way that most of the Archimedes Palimpsest has been imaged, at about 700 dpi.  If you would like to see a report on the spectral imaging of the Archimedes Palimpsest, click here. If you would like to see a description of the software used for the spectral imaging of the Archimedes Palimpsest: |