In: Wagner W„ Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
393 
ellipse of Figure 6, even the results shown in Figures 6 and 7 
would not be affected either. 
6. SPATIAL REGULARIZATION 
In order to improve the quality of the obtained output, we test 
here the usefulness of applying a spatial regularization 
technique as well. Again, in order to test the principle itself, we 
use a simple spatial regularization method: majority voting 
within a sliding window. Applying this method to the image 
shown in Figure 11, we obtain the result presented in Figure 12. 
Figure 12. Result of spatial regularization applied to Figure 11 
Although spatial régularisation does not affect the method itself 
either, it improves the quality of the final result by removing 
isolated points thus clearing the image and emphasizing the 
detected areas, as can be seen by comparing Figures 6 and 13, 
i.e., Figures 7 and 14. 
Figure 13. Same as Figure 6, but obtained after applying spatial 
regularization to the starting images. Region within the blue 
ellipse is zoomed in Figure 14 
Figure 14. Zoom of the region within the blue ellipse in Fig. 13 
7. VALIDATION 
Results presented here indicate that there is a region where 
human activities might have been performed, with changes in 
coherence values corresponding to works in building a hard 
surface on a sand terrain, and that these activities appear mainly 
in two parallel stripes as well as around them. 
A newspaper article from February 2008 (Greenberg, 2008) 
makes it possible to validate the obtained results. This article 
speaks about works at this same airfield, mentioning that a new 
runway is being built, together with other operational buildings 
around the runway, and that the works are about to be finished. 
The article also shows aerial photos which correspond to the 
region extracted by our analysis (Fig. 14). This means that our 
results are in a complete concordance with the reality, showing 
that the approach presented in this paper is promising for 
detecting human activities. 
8. CONCLUSION AND FUTURE WORK 
A simple method to detect potential human activities using 
CCD technique in repeat-pass SAR imagery is presented here. 
The results are validated, showing that the method is promising 
in extracting zones where a potential temporal change in the 
human activities occur, such as works in building-up areas of 
hard surface. As a good study case for a first research step, a 
desert type of soil is analyzed. Terrains which represent more 
difficult situations from the point of view of CCD technique 
will be studied in a next step. 
The presented method does not need any specific knowledge 
sources. Nevertheless, we show in this paper that if knowledge 
sources exist, it is possible to include them in order to improve 
final results. These knowledge sources can be related to the 
sensors, such as their operational principles, or to the situation 
at hand, referring to the context - terrain type, land-use, 
historical background etc. Three types of knowledge sources 
(information on the area of interest, pedologically extracted 
robustness and slope extracted from the digital elevation model) 
are included in the obtained results as an illustration. In a next 
step, possibilities of including some other knowledge sources, 
such as geological maps or aerial photos will be studied. 
In addition, in order to improve the aspect of the final result, the 
application of a spatial regularization method is tested, showing 
that it might be useful to further analyze ways to remove 
artefacts and preserve only the useful information. 
Finally, in future work, fusion of the obtained results with data 
coming from other sensors in order to further improve the 
quality and the robustness of the results will be investigated. 
REFERENCES 
Fanelli, A. et al., 2000. Understanding ERS Coherence over 
Urban Areas. In: ESA-SP-461 (Ed.), ERS-Envisat Symposium, 
Gothenburg, Sweden. 
Greenberg, H., 2008. “Longest landing field in Middle East to 
be constructed in Nevatim”. 
http://www.ynetnews.com/articles/0,7340,L-3508174,00.html 
(accessed 30 May 2010) 
Lu, D. et al., 2004. Change detection techniques. International 
Journal of Remote Sensing, 25(12), pp. 2365-2407.