Return to ING home page5  Handling pixel data

This page is part of the ING document INS-DAS-31 Design notes for UltraDAS

Background

Udas_camera handles millions of pixels in each run. The pixels arrive from the camera as a stream of 16-bit numbers, the order of which depends critically on the number of output amplifiers - or "readout channels" -  in the camera and the readout format. Although there are a few cases in which the pixels arrive in the correct order for storage on disc, in general, udas_camera must do considerable work to arrange and process the pixels before outputting them.

The general course of pixel processing is this:

  1. Determine which areas of the camera will be read out; assess windows and binning.
  2. Make a template for the composite object in which each readout will be stored. Include objects to receive pixels as they come from the camera and objects into which the received pixels can be assembled.
  3. From the template, determine how the camera should be programmed.
  4. From the template, determine the order in which the pixels will arrive and decide what should be done with each pixel: this is called "mapping the pixel stream".
  5. Programme the camera for this readout format.
  6. Repeat for each readout in the run:
  7. Combine associated readouts.
Steps 1 and 2 are done by the Formatter object. Steps 3..5 are done by the Pixel-stream object. Steps 6 involves a collaboration of objects lead by the Run object. Step 7 is done inside the Run object.
 

Topics of detail