| the 
poor 
| the 
facts 
ited. 
ame 
ould 
> the 
und 
ains 
EM 
ality 
ique 
nain 
the 
the 
nels 
the 
be 
ntly 
and 
'eter 
PU 
tical 
rom 
sual 
and 
data 
rom 
rban 
soil 
| the 
cted 
O Influence of terrain on SAR backscatter reduces the 
suitability of the data for recognition of the features, 
such as roads, cities, rivers and others. 
C The fusion of SAR with PAN using this type of band 
combination is not recommended in tropical areas 
with mountainous terrain. 
0 This type of technique does not solve the cloud cover 
problem because the range of digital numbers 
corresponding to clouds is preserved and even 
enhanced if not excluded from the calculation. 
Brovey Transform 
The images obtained from this special ratioing and 
multiplication technique (ERDAS, 1995) showed the 
following important elements: 
C) It preserves the spectral content of the VIR data 
while introducing the texture from SAR. 
0 The resulting image is not quite as sharp as the one 
produced from multiplication only. 
0 The water/land boundaries are well defined in the 
fused images; it allows one to assign colour to the 
water currents (e.g. tidal inlets). 
PCA 
PCA processing to fuse optical and microwave image 
data resulted in the following conclusions: 
C Radiometric pre-processing plays an important role 
in relation to the spectral content of the fused image. 
D Appearance of SAR significantly influences fused 
VIR/SAR image in terms of feature visibility. 
O As a consequence, features that are detectable on 
SAR data can be introduced to the VIR data by 
image fusion to complement the data (e.g. soil 
moisture, urban area, oceanographic objects). 
0 XS/ERS fused imagery shows more details than 
TM/ERS based on the higher spatial resolution of 
XS compared to TM. 
For SAR combinations using PCA the following remarks 
are valid: 
O Principal component SAR images show a potential 
for topographic mapping. 
O This is valid in particular for the 3D impression of 
topography and change detection. 
O Possibilities have not been fully explored, e.g. 
combination of principal components with optical 
data. 
IHS 
The IHS combinations are characterised by: 
O The capability of allocating data from the SAR to 
cloud covered areas without having to identify the 
clouds in an earlier stage. 
O The speckle is preserved from the SAR data in the 
fused image. 
O Similarities with the Brovey transformation in terms 
of spectral content of the imagery. 
659 
C Reduced spatial detail compared to original optical 
data. 
Mosaic 
The mosaic has an important position amongst the image 
fusion techniques as far as cloud removal from VIR and 
the replacement of radiometrically distorted SAR data is 
concerned: 
C) The result depends very much on the quality of the 
cloud/shadow mask designed for the mosaic. This is 
a critical point for the optical imagery. The 
identification of foreshortening, layover and shadow 
areas in the SAR is based on DEM calculations and 
pure geometry. These products are often delivered 
with the SAR image itself (e.g. GIM' from ESA). 
O Itis essential to match the histograms of the various 
input data to each other. 
C) It can be used in combination with any other image 
fusion technique. 
Combinations of techniques and images 
The best combinations of techniques and images were 
achieved with ‘triple sensor image fusion’. This refers to 
the fusion of SPOT XS and PAN with ERS-1 SAR. The 
images were developed during the process of improving 
published techniques and combinations. Their 
characteristics are: 
CJ Fused data contains the multispectral information 
from XS, the spatial resolution from PAN and the 
texture from SAR. 
C) The road network is clearly visible, the same is valid 
for housing structures in the urban area. 
CO Field boundaries are enhanced, variations inside the 
fields as well as differences to the field 
arrangements compared to the XS are visualised. 
O Speckle does not dominate the image as seen in 
other fused examples. 
6. CONCLUSIONS 
The results of the presented research prove the 
usefulness of fusing microwave remote sensing data with 
complementary optical imagery as far as topographic 
map updating is concerned. From the findings a first 
image map prototype was designed and printed at 
BAKOSURTANAL in Indonesia. The image map 
contains fused, multisensor, multitemporal satellite 
images from SPOT XS, ERS-1 SAR and JERS-1 SAR 
covering the research site in Indonesia in which all the 
clouds have been removed. 
The study contributed to operationalize VIR/SAR image 
fusion in presenting a processing line considering 
current operational satellite systems delivering images 
  
' Geocoded Incidence angle Mask 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996