| parthitv | stsdas.hst_calib.hsp | parthitv |
parthitv -- Remove particle events from HSP data files.
parthitv infile outfile outtable scheme dthresh athresh sigmas
The term "particle event" is a general term for any observed spike in the data. This task removes particle events by assigning a user-specified number to a data point's corresponding pixel in its output mask file. The user can choose from two methods to remove such events: (1) the statistical method will flag any observed measurement higher than a user-specified multiple of standard deviations above the mean value of the whole data set, and (2) the threshold method flags any observed measurement higher than a user-specified threshold level.
The statistical method is done in an iterative fashion. After the first round of particle event removal take place, remaining data points are rechecked for points above the statistical criterion of the new data set. The process goes on until no new points are removed.
This task also converts count to count rate (for digital data), averages the input data, and produces an average count rate (or analog reading) and its mean error. The average count rate (or analog reading) will be written to the output table as a new row. Output observation epoch is the modified Julian date of the first packet time.
Only single-group GEIS files can be used as input to this task.
The following keywords are needed from the science file header:
'ROOTNAME' ID of the observation set (char*10).
'DATA_TYP' Data type (char*7).
'SAMPTIME' Sample time of the observation (real).
The following group parameters are required to calculate the
observation epoch:
'FPKTTIME' Time of the first packet, i.e., the starting
time of the exposure (char*24).
'ROOTNAME' ID of the observation set (char*10).
'SCHEME' Method of removal, as passed from the 'scheme'
parameter (char*12).
'DOBJ' Mean digital count rate of the object after
the particle event removal (real).
'DSKY' Mean digital count rate of the sky after
the particle event removal (real).
'AOBJ' Mean analog digital number (DN) of the object
after the particle event removal (real).
'ASKY' Mean analog digital number (DN) of the sky
after the particle event removal (real).
'DOBJ_ERR' Mean error of 'DOBJ' (real).
'DSKY_ERR' Mean error of 'DSKY' (real).
'AOBJ_ERR' Mean error of 'AOBJ' (real).
'ASKY_ERR' Mean error of 'ASKY' (real).
'DTHRESH' If 'scheme = threshold', then this value is
equal to the digital threshold level, as
passed from the 'dthresh' parameter; if
'scheme = statistical', then this value is
equal to the statistical criterion, as
passed from the 'sigmas' parameter (real).
'ATHRESH' If 'scheme = threshold', then this value is
equal to the analog threshold level, as
passed from the 'athresh' parameter; if
'scheme = statistical', this will be the
statistical criterion, as passed from the
'sigmas' parameter (real).
'OBJ_SAMPTIME' Pixel sample time of the object, in seconds (real).
'SKY_SAMPTIME' Pixel sample time of the sky, in seconds (real).
'DOBJ_REMOVED' Number of points been removed by this task
for the digital object file (int).
'DSKY_REMOVED' Number of points been removed by this task
for the digital sky file (int).
'AOBJ_REMOVED' Number of points been removed by this task
for the analog object file (int).
'ASKY_REMOVED' Number of points been removed by this task
for the analog sky file (int).
'EPOCH' Epoch of the first packet time in modified Julian
date (double).
1. Remove particle events from all data files in the current directory. Write the result to the output table output. The output mask files will be in the same directory. Use the statistical method to remove points 3 sigmas higher than the mean value of all valid points. Pixels corresponding to particle events are assigned the value of 1 in the output mask file.
hs> parthitv "*.d*h" "." "output" "statistical" sigmas=3 \ >>> lowflag=no hitval=1