IAPRS & SIS, Vol.34, Part 7, “Resource and Environmental Monitoring", Hyderabad, India,2002
IN-FLIGHT CHARACTERIZATION OF TES PAN SENSOR ON CHHARODI SITE
Shashikala Palsule, Yogdeep Desai, Anjali Garg, Bankim Shah
Data Quality Evaluation Division, SIIPA, Space Applications Center, ISRO, Ahmedabad-15, INDIA
shashi@ipdpg.gov.in, desai@ipdpg.gov.in, anjali@ipgpd.gov.in, bankim@ipdpg.gov.in
KEYWORDS: Square wave Response, Point Spread Function, Modulation Transfer Function, Ground Sample Distance
ABSTRACT:
Salient feature of IRS series of satellites is the availability of spatial resolution from several tens of meters to several meter data. The
improvement in spatial resolution gives challenge to calibration procedure such as selection of test sites, targets, and image sensor
characterizing parameters. IRS spatial resolution has improved over one and a half-decade from medium to high resolution. At initial
stages of IRS mission natural test sites like Kanij river sand, Sam desert as medium reflecting target, Kurukshetra tank and ICRISAT
bare soil as low reflecting targets were used. With improved spatial resolution, artificial targets like Black/White cloth, painted asbestos
sheets were experimented and deployed for characterizing various image sensor parameters.
The successful launch of Technology Experiment Satellite (TES) with step and stare technique has given high-resolution images in
panchromatic (PAN) band. In-flight performance of TES Pan camera was evaluated using special designed artificial targets on Chharodi
test site, located near Ahmedabad. Images can be described in terms of certain fundamental properties regardless of wavelength at which
image is recorded. The basic image sensor characterizing parameters are Point Spread Function, Radiance, Scale, Resolution, Contrast,
Square Wave Response and Modulation Transfer Function. The known target reflectance, shape & size and site dimensions aided in
analyzing multitemporal data.
1.0 INTRODUCTION
The term “Spatial resolution” of a sensor or its image is often
quoted as the ground instantaneous field of view (GIFOV),
which conveys a strong intuitive meaning as a measure to
Detect objects. It is seen in practice that even though objects
smaller than the GIFOV are Detectable due to high contrast
surrounding background, but they are not necessarily
Recognizable except by the general context of the image. The
continuous improvement of IRS sensor resolution from several
tens of meter to meters have made possible to understand
systematically the basic image sensor parameters.
The IRS optical sensor camera converts the
upwelling radiance (reaching at detector array) into an image
of the radiance spatial distribution. The radiance spatial
distribution at pixel level is an integrated effect of the signal
produced by the two components as target and background.
The reflected energy in simple words can be interpreted that
sensor detect a single unit for bright object against a dark
background even if it occupied only fractional percentage of
GIFOV of sensor. The same process can be viewed with
reference to object reflectance of 4.0% and the lighter bright
target reflectance of 8.0% will give a contrast ratio of 2:1. The
lighter bright target is still detectable if recorded digital
number differs from the pure background digital number by
threshold value. This threshold value is also a digital number
for detectability of object depends on radiometric quantization
of sensor coupled with the Target and Background reflectance
(contrast ratio) and the sensor GIFOV. An additional important
factor to be accounted for image resolution is the sample-
scene-phase i.e. the relative location of the image pixels and
the target. This relative scene-sample-phase is unpredictable
(and almost always unknown) for any given image and varies
from acquisition-to-acquisition with a uniform probability
distribution between +/- 2 pixel.
Therefore the above said terms like GIFOV, contrast ratio,
radiometric quantization, sample-scene-phase collectively
conspire to determine the "resolution" of the image. Remote
sensing system has "resolution" in the spectral, spatial
&temporal measurement domains. No instrument or sensor can
measure a physical signal with infinite precision. The
mathematical description applied to a wide range of
instruments is the output signal equals the input signal
convolved with the instrument response function.
This paper highlights the results of In-flight TES PAN
SENSOR imaging system parameters on Chharodi Test Site
using parameter specific specially constructed artificial targets.
2.0 TEST SITE DESCRIPTION
Chharodi Calibration Site at Sanand is a North Cot Cattle farm
situated about 30Km West of Ahmedabad. The farm contains
field, which are used for production of cattle fodder. This farm
is owned by Gujarat Agriculture University (GAU). Under
MoU between Space Application Center (SAC) & GAU, 300m
x 300m, bare soil field has been developed for IRS Sensor
Calibration-Validation Activity. This site is presently fenced
and with special efforts flat field is maintained after removing
grass & bushes. The basic requirement of site such as flat
uniform area, stability of surface reflectances, identifiable area
in satellite image, easily accessible for conducting experiment
is satisfying. Yearly maintenance contract has been given to
local party for the maintenance of the site with round the clock
security.
The measuring instrument is portable ASD field Spectro-
Radiometer, operating in the spectral range of 350 nm to
1050nm with spectral resolution of 1.4nm over complete
spectral range. This radiometer collects Sun Irradiance data in
2.5°FOV and target radiance/reflectance in 18°FOV.
3.0 TARGET DESIGN AND DEPLOYMENT ON SITE
Artificial Target are made of asbestos sheets of 4ft x 8ft size.
These are arranged specially with uniform / variable size and
painted with Black/Gray/White "paints to get variable
reflectivity. These targets are designed as per image measuring
paramet
deployrr
Image i:
paramet:
advantas
Image S
due to
targets
between
TARGET
TES IMAC
ARTI
Technol
panchroi
with a sj
560 Km.
