Experimental Physics and
Industrial Control System

Hi Pavel,

Since you mentioned pyepics and numpy, I'll jump in to recommend a
slightly different waveform -> Image conversion using these:

   import Image
   import epics

   waveform_pv = epics.PV('xx:waveform.VAL')
   waveform = waveform_pv.get()  # will be a numpy array of unsigned ints
   # to get a truncated waveform, pass in the length you'd like:
   #     waveform = waveform_pv.get(count=400)

   width = ??     # that is, you'll have to assert this here.
   height = len(waveform)/width

   image = Image.frombuffer('L', (width, height), waveform, 'raw', 'L', 0, 1)
   image.show()

where you'll have the know the image height and width somehow to
unravel the 1-d waveform.   The Image.frombuffer() call is slightly
cryptic, but documented ('L' means 8-bit greyscale -- if you're
waveform held RGB data from a color camera, you'd change that here to
get a color image), and is faster than converting to a string and then
using Image.open().

--Matt