Edward A. Guinness
Department of Earth and Planetary Sciences
McDonnell Center for the Space Sciences
Washington University
St. Louis, Missouri 63130
This document contains a description of the radiometric
properties and calibration of the Viking Lander cameras. Each
Viking Lander camera had a photosensor array (PSA) consisting of
the following twelve photodiodes: four high resolution broadband
(0.4 to 1.1 micrometers) diodes; one low resolution broadband
diode; six diodes with color and infrared filters; and one diode
with a red filter and no amplification for viewing the Sun. A
complete description of the camera system is found in the
VOLINFO.TXT
file located in the DOCUMENT directory.
This CALINFO.TXT file describes: A) radiometric calibration
files; B) conversion from digital number (DN) to reflectance; C)
internal calibration and scan verification images acquired during
the mission; D) reference test charts mounted on the Landers; and
E) the vignetting function. Data in the CALIB directory are
taken from Huck et al. [1975] and are derived from a series of
pre-flight component level and end-to-end calibration tests.
Descriptions of the radiometric calibration tests are found in
Wolf et al. [1977]. Because of the severe sterilization
requirements of the Landers and neutron radiation damage from the
radioisotope thermoelectric generators on the Landers, changes to
the pre-flight calibrations were expected. Calibration and
performance of the cameras during the Primary Mission are
documented in Patterson et al. [1977] based on internal
calibration data from a lamp. These internal calibrations were
performed frequently throughout the mission and provided a
database for monitoring the radiometric characteristics of the
cameras.
Data files in this directory are listed below. For each
data file, there is also a detached PDS label file with the same
name but an extension of .LBL.
1. Introduction
| Data File | Label File | Description |
|---|---|---|
GAINOFF.TAB
| GAINOFF.LBL
| Constants used for converting DN to photodiode output voltage. |
OPTICS.TAB
| OPTICS.LBL
| Spectral transmittance and reflectance of camera optical components. |
FILTERxx.TAB
| FILTERxx.LBL
| Spectral responsivities for photodiodes, where xx indicates the lander and camera number, respectively. |
PHOTOSEN.TAB
| PHOTOSEN.LBL
| Physical and electronic data for the photosensors. |
v = [(DN/4) * (2^GN) / Kg] + (K1 * OFN) - K2
| (1) |
where v is the photodiode output voltage in units of
volts, GN is the gain number listed in the PDS image
label, DN is the digital number from the image,
Kg is the gain constant, K1 is the offset
constant 1, OFN is the offset number listed in the PDS
image label, and K2 is the offset constant 2. Values for
Kg, K1, and K2 vary as a function of camera
and are found in the file GAINOFF.TAB. The digital number is divided
by 4 to convert it to the original 6-bit value. Gain and offset
constants vary by a small amount as a function of gain number, offset
number, and PSA temperature. These small variations are discussed in
Wolf et al. [1977]. The average PSA temperature is listed in the PDS
image label.
v = K * INT [L(w) * T(w) * R(w) * dw]
| (2) |
The symbol INT[ ] stands for a definite integral over
wavelength (w) with limits of zero to infinity, but with
practical limits of 0.4 to 1.1 micrometers based on the sensitivity of
the photosensor. In equation 2, v is photodiode output
voltage in volts, K is a constant described below,
L(w) is scene spectral radiance with units of power per
projected area per solid angle, T(w) is the transmittance
of camera optics (described below), and R(w) is the
spectral responsivity of a given photodiode (values tabulated in
FILTERxx.TAB files). No Sun diode data are listed in the FILTERxx.TAB
or PHOTOSEN.TAB because none are given in Huck et al. [1975]. The Sun
diode consisted of a red filter with no amplification [Patterson et
al., 1977].
In equation 2, the constant K, is:
K = kc * Rf * G * ((PI/4) * B * Dl)^2
| (3) |
where kc is a calibration factor to account for
uncertainty in the camera model (equation 2), calibration hardware,
and test procedures; Rf is the preamplifier feedback
resistance for a given photodiode; G is the channel
(photodiode) gain; PI is the constant 3.14159...;
B is the photodiode instantaneous field of view in
radians (Note values of B in PHOTOSEN.TAB are given in
degrees); and Dl is the diameter of the camera lens.
Values for these parameters are tabulated in the file PHOTOSEN.TAB as
a function of camera and photodiode. Values of B in
PHOTOSEN.TAB were computed from formulas in Huck et al. [1975], and
are a function of the photodiode aperture radius, distance between a
photodiode and the camera lens, the lens focal length, and the
photodiode in-focus object distance. These values are also included
in PHOTOSEN.TAB.
Also in equation 2, the optical spectral transmittance T(w),
is:
T(w) = tcw(w) * tw(w) * rm(w) * tl(w)
| (4) |
where tcw, tw, and tl are the transmittances
of the contamination cover window, the camera window, and the camera
lens, respectively, and rm is the mirror reflectance.
Note that tcw is included only if the contamination cover
window was in place, which it was for all cameras at the start of the
mission. The purpose of the contamination cover was to protect the
camera from sand blasting or dust coatings. It was designed to move
aside if it became coated. The cover was deployed (i.e., moved out of
the optical path) for two cameras during the Extended Mission. It was
deployed for camera 1 on Lander 1 during camera event 11F252 on VL 1
Sol 470. The cover for camera 2 on Lander 2 was also deployed during
camera event 22G255 on VL 2 Sol 593. Analysis of images taken before
and after deployment indicates that there was no significant coating
on the covers. The contamination covers remained in place for the
other 2 cameras during the entire mission.
r = Vs / M
| (5) |
where Vs is the photodiode voltage generated from the
scene and M is the voltage from a normally illuminated
Lambertian surface with unit reflectance at the same heliocentric
distance. Reflectance computed with equation 5 is a radiance factor
defined as the ratio of the scene radiance to the radiance of a
Lambertian surface of unit reflectance illuminated normally at the
same heliocentric distance [Hapke, 1981].
In equation 5, the quantity M is a function of camera,
photodiode, time, and whether the contamination cover window was
in place:
M = [(D0/D)^2 * K / PI] * INT [F(w) * T(w) * R(w) *
dw]
| (6) |
where D0 is the mean Mars-Sun distance of 1.52 AU;
D is the Mars-Sun distance at the time of the image;
K is the constant from equation 2; F(w) is
the solar spectral irradiance at 1.52 AU; T(w) is the
optical transmittance from equation 4; R(w) is the
photodiode responsivity; and w is wavelength.
The reflectance computed with equation 5 does not account for atmosphere contributions of attenuation and diffuse illumination. Several authors have reported on ways of dealing with the atmosphere using the brightness of shadows in the scene and optical depth measurements made with the Sun diode [e.g., Arvidson et al., 1989; Guinness et al., 1987; and Guinness, 1981].
| line 1 | - | entire line with ambient illumination |
| line 2 | - | first half with lamp off to sample dark current, second half with lamp warming up |
| line 3 | - | first half lamp continues to come on; second half lamp is at full power |
| line 4 | - | lamp is off and data collected with ambient illumination |
The nominal sequence of photodiode measurement during the 16 cycles of an internal calibration was as follows: UND, UND, IR1, Red, BB3, BB4, IR3, Blue, Survey, UND, IR2, Green, UND, Sun, BB1, BB2. Note UND means that no photodiode was used during that cycle. Internal calibrations were always done immediately before or after another image. Internal calibration images showed that there was no significant degradation of the BB1, BB2, BB3, BB4, Survey, and color diodes, but that the IR diodes degraded [Patterson et al., 1977].
Each Viking Lander had three reference test charts, which
contained a set of patches for radiometric, color, and spatial
resolution calibration. Specifically, each chart had a series of
11 gray patches of varying reflectance, a blue, green, and red
patch for color balancing, three sets of tribar patterns with
different spatial frequencies, and two patches coated with paint
that darkened when exposed to ultraviolet radiation. Data on the
reflectance properties of the individual patches on the reference
test charts can be found in Wall et al. [1975].
The three reference test charts were mounted on the lander
deck. Two charts were oriented so that they were directly viewed
by one of the cameras (i.e., normal to the camera line of sight
and about 1 meter from the camera). The third reference test
chart was mounted in the center of the back of the lander deck so
that it could be seen by both cameras. This third chart also
contained a set of ring magnets that were part of the magnetic
properties experiment.
The surfaces of the charts acquired thin coatings of dust
during the mission. The magnets also acquired magnetic particles
during landing and remained about the same throughout the
mission. The patches of ultraviolet sensitive paint degraded
with time [Moore et al., 1987; Zent et al., 1980; Hargraves et
al., 1979].
6. Reference Test Charts