. Istanbul 2004 
05 
uth shift 
0.5 
uth shift 
> shifts computed 
€) & (d) Orgeval 
order of the test 
s (wheat, corn, 
"site where the 
ed land is rare 
The villages on 
ilar shape. The 
aphic Mapping, 
re than 70% of 
ustang test site, 
nterval between 
2 (d)) computed 
of the samples 
ergetic relief of 
ration using the 
siven in TABLE 
g. 1 (a) and (b) 
ge and azimuth. 
range direction, 
d we decide to 
lobale shifts. 
GER SHIFTS 
-PIXELIC ONES 
Mustang 
-4 
3 
-0.9 
-0.9 
ion algorithm is 
d after the sub- 
new sub-pixelic 
    
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004 
  
  
(a) 
a, 
) 
4 
r 
(c) 
  
  
(d) 
Fig. 2. SLC ERS interferometric pair tandem of (a) Orgeval and (b) Mustang and their corresponding coherence image (c) and ((d). The coherence maps 
were computed with the ISAR software. 
registration approach increase the number of coherent pixels and 
then the number of exploitable pixels for interferogram unwrapping. 
Elevation error estimation could be a good test for the validation of 
the developed algorithm. As we don't have some reference points on 
the test sites we couldn't achieve this test. 
V. CONCLUSION 
In this paper we developed a new approach, based on two steps, for 
the registration of an InSAR pair images with a sub-pixelic precision 
performed in the Fourier domain, using the Fast Fourier Transform 
(FFT). 
First, using the ISAR software, we perform an integer co- 
registration (coarse) of the interferometric pair. Than, we look for 
the sub-pixelic translation factors that maximises the cross-correlation 
between both images on a window dimensions of 2x2 pixels using an 
iterative process. However, in order to reduce the complexity given 
by the convolution in the spatial domain, we consider the transform 
model in the Fourier domain. 
This will be possible in the Fourier domain, since a shift in the 
spatial domain is achieved by a multiplication with an exponential 
factor (which is the real shift less than one pixel) in the frequency 
domain. The new algorithm was tested on two different kinds of 
interferometric pairs with different coherence map. In the case of 
coherente pair the algorithm gives satisfactory results. However, the 
local computed offsets are not reliable when the InSAR images 
contain decorrelated zones, shadow zones or layover. 
REFERENCES 
[1] L. C. Graham, Synthetic interferometer radar for topographic mapping, 
Proc. IEEE, vol. 62, pp. 763-768, 1974. 
[2] A. Gabriel and R. M. Goldstein, Crossed orbits interferometry : Theory 
and experimental results from SIR-B, Int. J. Remote Sensing, vol. 9, pp. 
857-872, 1988. 
[3] BAMLER, R., and PHILIPP, H., 1998, Synthetic aperture radar interfer- 
ometry, Inverse Problems, 14, , pp. R1-R54. 
[4] ROSEN, P. A., HENSLEY, S., JOUGHIN, I. R., LI, F. K, MADSEN, 
S. N., RODRGUEZ, E.. AND GOLDSTEIN, R. M., 2000, Synthetic 
Aperture Radar Interferometry, [EEE Transaction on Geoscience and 
Remote Sensing, 88, pp. 333-382. 
[5] LI, F. K., and GOLDSTEIN, R. M., 1990, Studies of multibaseline 
spaceborne interferometric synthetic aperture radar, /EEE Transaction 
on Geoscience and Remote Sensing, 28, pp. 88-97. 
ABDELFATTAH, R., NICOLAS, J. M., and BOUSSEMA, M. R., 1998, 
Topography optimization for accurate interferometric phase generation, 
[6 
—