ISIS 3 Application Documentation
|Printer Friendly View | TOC | Home|
Radiometrically calibrates a MOC image
moccal performs radiometric corrections to images acquired by the Mars Global Surveyor (MGS) MOC camera. MOC has one Narrow Angle (NA) detector and 2 Wide Angle (WA) detectors. It also has two camera systems, A and B. The B camera system is a backup and currently has very little or no radiometric calibration information. It is supported to a very limited extent but is not in use at this time.
The MGS MOC NA detector has a total of 2048 pixels. The WA Red and WA Blue detectors have a total of 3456 pixels. Images from MOC may or may not include all pixels in the acquired image. There are special summing modes that are utilized on-board the spacecraft to average detector pixels to combine them into a single output pixel value. Both NA and WA detectors can utilize crosstrack (sample) and downtrack (line) summing modes. The value of these modes indicate the number of samples and lines, respectively, that were summed and averaged to result in the pixel values stored in the input file to moccal specified in the FROM parameter. These values are stored in the ISIS labels as CROSSTRACK_SUMMING and DOWNTRACK_SUMMING, respectively. Note that this will reduce the number of samples in the output image by a factor of at most the crosstrack summing mode value. The NA is restricted to modes of 1 through 8 for both crosstrack and downtrack summing. Furthermore, the crosstrack and downtrack modes must be the same. For both WA detectors, crosstrack and downtrack modes range from 1 to 127. Unlike NA, WA crosstrack and downtrack summing modes can differ. There are two additional special WA crosstrack summing modes. These modes are indicated by crosstrack summing mode values of 13 and 27. Both of these modes use specially configured tables to indicate the relative detector hardware pixels involved in computing the output pixel value. These modes are intended to maintain approximately equal spatial resolution from nadir to limb. Crosstrack summing mode 13 has a resolution of 7.5 km/pixel and maximum number of output pixels of 768; mode 27 has a resolution of 3.75 km/pixel with the maximum number of output pixels of 384. moccal recognizes both these modes and applies appropriate measures to radiometrically correct them.
The MOC camera has the ability to acquire images of differing sizes in both line and sample. The starting hardware detector pixel for the acquired image is specified by the ISIS label keyword, FIRST_LINE_SAMPLE. The first pixel in the detector is indicated by a value of 1. (Note that in the original PDS image products this is the EDIT_MODE_ID keyword and is 0-based. In ISIS, FIRST_LINE_SAMPLE is a PDS standard and is 1 plus this value, a 1-based numbering scheme.) Note that all corrections are applied relative to the actual hardware pixel.moccal has a number of parameters required to process MGS MOC image data. Most of these parameters are read from an ISIS parameter defaults file located in the the mgs directory specified in the IsisPreference file. This file has the form:
$mgs/calibration/moccal.kernel.xxx where "xxx" is a positive number.moccal will search for the parameter file with the highest number and load all its defaults needed to process MOC images. Note that the caller may explicity provide an alternative ISIS parameter defaults file name in the "CALKERNEL" input parameter if they have their own set of defaults that applies to MOC data. See an existing calibration file for details on how to configure this file.
The MOC response equation (without pixel-to-pixel variation terms) is as follows:
dn = a * (r * ex + dc * ex + g) + (z - off) where r is the average signal being generated at the focal plane (in DN/msec at minimum gain), z is the fixed zero offset, off is the commanded variable offset, dc is the dark current term (in DN/msec at minimum gain), g is the gain- dependent offset (in DN at minimum gain), a is the commanded system gain (where minimum gain is 1 and all other gains are >1) and ex is the exposure time in milliseconds.moccal computes "r" from "dn" at every pixel in the input MOC image. Hence, the actual equation moccal implements is as follows:
r = ((dn - z + off) / a - g) / ex - dc
Additional pixel-to-pixel variations are corrected by a detector coefficient file also specified in the parameter defaults file. This file has been provided after analysis of the calibration data taken prior to and after the launch of MGS. Certain detectors may not have these coefficients derived and thus will not have this correction applied. If these files do exist, they are typically taken from the mgs directory specified in the IsisPreference file. The coefficent file is expected to contain any number of comments indicated by a '#' as the first character in a line. The first non-blank or uncommented line must contain the number of values (i.e., one per line) of pixel coefficients for this detector. This must be the exact number of pixels in the given detector, 2048 for the MOC NA detector and 3456 for the MOC WA Red and Blue detectors. All non-blank, uncommented lines after the detector count expects two values per line. The first is a multiplicative value followed by an additive value. These values are applied to the value "r" computed in the above equation to correct for individual pixel variation. Note that some detectors, namely WA Blue, may not have a pixel variation correction derived. For these cases, moccal will apply the calibration equation without pixel-to-pixel variation applied.
Prior to MGS MOC mapping phases, MOC NA downtrack summing modes greater than 1 was implemented differently in onboard camera software. A software patch was made at some point prior to normal mission mapping phase that would be used throughout the mapping and extended phases for the rest of the mission. moccal applies the following algorithm for all MOC NA images: *ALL* NA images have the exposure times multiplied by the downtrack summing mode. All NA images prior to mapping phases has the recoreded GAIN mode value divided by the downtrack summing mode and a new GAIN value is looked up in the table that is closest to this value. It is substituted in the calibration equation for the commanded gain value (a). The critical piece of information needed for this algorithm is the time (spacecraft clock count or SCLK) when this patch was uploaded to the spacecraft. This time has been determined to be just prior to the first image acquired in the normal mapping phase. This is image m0000001.imq and its SCLK is 607568463:128 which corresponds to April 3, 1999, 01:00:40.4405 UTC. None of the MOC WA images are affected by this algorithm.
References: "MOC2 Calibration Report, October 1997 [SCCS 10/17/97 version 1.3]" "Software Interface Specification, Narrow Angle and Wide Angle Standard Data Products (September 1999 (revised for MGS) (formatted April 7, 2000))"
A uncalibrated MOC image. If the image is from the narrow angle camera, it should be run through mocevenodd and mocnoise50 prior to moccal.
The resultant radiometrically calibrated cube
If you do not enter a file, mocal will search for a file with the form:
$mgs/calibration/moccal.kernel.xxx where "xxx" is a positive number.The kernel with the highest version (xxx) will be used. See an existing calibration file for details on how to configure this file.
|Internal Default||Internal Default|
The user has the option to request the output of moccal to be in either units of percent reflectance (IOF="YES") or in average signal level at the focal plane (IOF="NO") in DN/msec. moccal has a set of default I/F (percent reflectance) factors for each detector, NA and both WA. Should they not have this value, a W0 of 0.0 will be detected and I/F cannot be generated. moccal will detect occurances of W0=0.0 and report this to the user whilst functioning as if IOF="NO" irregardless of what may have been provided by the user for IOF. Conversion to I/F takes place after calibration and pixel-to-pixel variation correction.
The MOC WA camera has an additional capability to change the commanded gain state (a) and offset (off) at any time during image acquistion. The approximate time these changes are made is provided in a file called "wago.tab". This file can be found on all MOC CDROM products under the "index" directory. This table, called a "WAGO" (Wide Angle Gain Offset) table, contains one entry per line of four fields. All fields are double quoted. The first field is the WA detector for which the gain and offset change occurs, the second is the SCLK time of the change, the third is the hexidecimal code of the new gain mode identifier and the fourth is the offset mode identifier. The WA detector will be either "RED" or "BLUE". The SCLK is the spacecraft clock count of the time at which the command to change gain and offset was initiated. The gain mode identifier indicates a gain value that is stored in a table of gain values. The offset mode identifier is actually specified in factors of 5 DN. Hence, if the commanded offset mode is "5", then "off" in the above calibration equation is set to 25 (5 * 5 = 25). Any number of these WAGO changes can occur during WA image acquisition. WAGO changes are apparent as an abrupt change in brightness at a particular line. Beware, not all changes are WAGOs as this is also a characteristic of dropped data.
As the timing knowledge of the gain-offset changes is only accurate to 1/8 of a second, and the WA line rate is somewhat faster than this and not an even multiple of 1/8 second, the line on which the brightness discontinuity occurs may not match precisely the line computed from the image start time, line rate, and known time of state change. In addition, changes in highly-summed images (such as global map swaths) may result in summed output lines having been created in the instrument from unsummed input lines taken in different states. moccal attempts to detect lines that exhibit this signature and corrects them using a brightness matching algorithm. With the WA line rate at 75ms, for a downtrack summing mode of 1, the change should occur no more than two lines before or after the computed line of the change. moccal looks in this window of transition lines and applies an averaging techinque to correct them if NULLWAGO="NO". If NULLWAGO="YES", then these lines are assigned the ISIS NULL value and can be handled later using alternative processing techniques.