emeventsproj projects a list of events onto an image, projecting multiple events onto several pixels (using the geometry contained in PATTERN). The output image is always 600x600 (even in Window mode) and covers the CCD only (not the under and overscans). The main use of emeventsproj is to prepare an image for embadpixfind.
emeventsproj works both on raw events files straight from the ODF, on merged PPS event lists, and on single CCD events file processed by emevents. The latter solution is the only one allowing full functionality. If the input file is a merged PPS event list, the mergedeventlist and ccdnr parameters must be set.
When rejectbadevents=Y, events flagged for rejection are not included in the projection. The flags OUT_OF_FOV and REJECTED_BY_GATTI are ignored, because it is useful to find bad pixels outside the field of view, and masking the truncated events would prevent from detecting single bright pixels with charge larger than the upper threshold. The rejectbadevents option is not supported if the input events file is straight from the ODF, because there is no FLAG there. If another selection on events is required, then it should be done via evselect prior to calling emeventsproj.
When projectenergy=Y, energy is projected instead of counts.
For each event, ENERGYE1 is added to the map at the events position
(instead of 1), and ENERGYE2 is shared among all secondary pixels
of the events (exact reconstruction of the original charge pattern
is impossible for events larger than 2 pixels).
ENERGYE3 and ENERGYE4 are not projected.
If a single frame were selected in the events file, the output image
would look like the original CCD map read-out on-board after thresholding,
minus all charge patterns which did not pass the EDU (and flagged events
if rejectbadevents=Y).
The same algorithm is used in emevents
when analysepatterns=Y.
The projectenergy option is not supported if the input events file
is a merged PPS event list, because there is no ENERGYE1 and
ENERGYE2 there.
