06639ca8   
   XPost: alt.usage.english, sci.physics   
   From: siegman@stanford.edu   
      
   In article   
   ,   
    Jonathan de Boyne Pollard    
    wrote:   
      
   > > It would be interesting to see what a spectrum analyser would say about   
   > > the colour purple. I suspect that it would find a mixture of two or more   
   > > colours, where for the purposes of this paragraph I define "colour" to   
   > > mean a band-limited range of frequencies.   
      
   Side note on this:  Take a picture, using a digital camera, of a spot   
   intensely illuminated by 1.06 micron infrared from a laser.  [Or,   
   spread the laser beam out and (carefully!) take a picture with the   
   beam coming right into the camera.]  The spot will probably appear top   
   be a strong purple color.   
      
   Reason:  The RGB sensors/pixels on the camera chip have filters in   
   front of them that are supposed to pass only R, G and B.  But all   
   three of these filters have transmission profiles that open up beyond   
   0.9 microns or so, and transmit a fair amount of infrared into the   
   pixels (the R filter is probably not blocked at all on the IR side),   
   and the sensor pixels behind the filter have at least some residual   
   sensitivity to 1.06 microns.  So, the camera thinks it's seeing purple.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)