loaded into SARscape in ENVI using the ASAR product
standard format reader. The processing of the multi-
temporal ASAR APS data involved several steps including:
l. Data import of the original ESA datasets.
2. Data multilooking (number of looks: 1 and 5
respectively in azimuth and range),
Data coregistration.
4. Data multitemporal Filtering (using a De Grandi
method).
5. Extraction of SRTM-3 version 4 DEMs for use
with geocoding.
6. Data geocoding with radiometric calibration and
normalization (using a grid size 25 m to match the
resolution of the ASAR data).
Ww
These steps roughly consist of image calibration or
conversion to the radar backscattering coefficient sigma
nought (co), image registration or geocoding, and image
spatial filtering (Nguyen, Armando, Thuy, Young, Trung, &
Bouvet, 2009). Image calibration consists of correcting SAR
images for incidence angle effect and for replica pulse power
variations to derive physical values (Nguyen, Armando,
Thuy, Young, Trung, & Bouvet, 2009).
The processed SAR datasets were then fused with Landsat
ETM+ scenes from 2011, and ENVI was used to classify the
images in order to quantify the area covered by the rice
Crops.
3. RESULTS
The processed SAR images for each of the dates during the
growing season are represented in Figure 1 below. The rice
crops are represented by the darker areas at the beginning of
the growing season (Figure 1a) as the backscatter coefficient
for water is lower. Once the plants emerge from the water,
the backscatter coefficient gets higher and these areas
become brighter in the images from August through October
(Figure 1: b), c) and d)). As the crops mature and are
harvested starting in November (Figure 1: e), the areas begin
to appear darker again.
k Bs
a) July 2011
September 2011
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B7, 2012
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia
“ e) November 2011
Figure 1. Processed SAR Multi-temporal Images; HH
Polarization
After the ENVISAT ASAR data was processed, SARscape in
ENVI was used to produce an RGB color composites using
HH and HV bands in order to improve the visualization of
the objects in the images. In the series of RGB color
composites shown in Figure 2 below, red = (input 1 - input 2)
/ (input 1 + input 2), green = input 2, and blue = input 1.
This type of color composite enhances the differences in the
backscatter results of the rice crops during different periods
of the growing season. The forests (green) are more easily
distinguished from the rice crop fields (red to purple
depending on water content).
) July 2011
c) September 2011 d) October 2011
- €) November 2011.
Figure 2. Processed SAR Multi-temporal Images; Color
Composites
In the ETM+ image from Figure 3 below, at the end of the
growing season in November, the light green areas represent
the forested areas, the pink/purple areas represent the fields
where the rice has been harvested, and the blue areas
represent water.
