ROJECT 
‚stem 
sitive 
ected 
e use 
ment, 
f the 
h the 
ration 
aphic 
ıch a 
oped, 
ion in 
. The 
high 
ata is 
nuous 
SAR 
> data 
factor 
zically 
  
In this study, multitemporal SAR images were 
acquired during the 1996 growing season. 
These were examined as a complete 
multitemporal data set, and also were examined 
in combination with various band combinations 
from a Landsat TM image. These data sets 
were classified and the accuracy of crop 
identification was evaluated. In general, 
ploughed and stubble fields were confused with 
newly planted corn, some cotton and 
uncultivated grass. New citrus plantations were 
confused with matured sunflower fields due to 
very similar radar signatures. These problems 
may be overcome by using crop rotation 
information, historical data, crop calendar 
differences for example. Owing to the inherent 
speckle, and hence high local variability, 
information about small land parcels is less 
reliable. 
Accuracies from SAR based multitemporal 
classifications and combined SAR and TM based 
classifications were investigated. It was found 
that accuracies could be improved by 
complementing the SAR data with optical data. 
The reason for these increases are of interest and 
need to be investigated. 
Future work will involve : 
1. The use of segmentation 
2. The use of a per field classifier approach 
3. Work to achieve a better understanding of 
the relationship between non-agricultural 
classes and radar multitemporal profiles. 
4. Work to extend the use of radar into the 
more difficult terrain. 
AKNOWLEDGMENTS 
The authors thank to ESA for their support of 
the IMP project. They are also grateful for the 
Support and consultation provided by the 
Manager of Dalaman State Production Station, 
and from Mr. Sefaattin Zeybek, Mr. Miimtaz 
Yurdagül and Mr. Orhan Ilhan, agriculturists. 
  
REFERENCES 
Brisco B. and R.J. Brown, 1995. Multidate SAR / TM 
Synergism for Crop Classification in Western Canada, 
Photogrammetric Engineering & Remote Sensing, Vol.61, 
No.8, August pp. 1009-1014. 
Fitzgerald, R.W., and B.G. Lees, 1994. Assessing the 
Classification Accuracy of Multisource Remote Sensing Data, 
Remote Sensing of Environment, 47:362-368. 
Keil, M., Scales, D., Winter, R., Kux H. and J. Roberto 
dos Santos, 1995. Tropical Rainforest Investigation in Brazil 
using Multitemporal ERS-1 SAR Data, Proc. of the Second 
ERS Applications Workshop, London, 6-8 December, UK, 
pp.77-82. 
Maktav, D., Sunar, F., Kapdash, S., and N. Musaoëlu, 
1996, Use of Remote Sensing and Monitoring the Lake 
Kóycefiz and Its Environment in Turkey, NATO/AGARD 
Meeting, France, 22-25 April. 
Manore, M., D’lorio, M. and J. Harris, 1996. SAR Data 
Fusion, Proc. of the First Latino-Americano Seminar on Radar 
Remote Sensing - Image Processing Techniques, Buenos Aires, 
Argentina, 2-4 December, pp.91-96. 
Mather, P.M., 1987. Computer Processing of Remotely- 
Sensed Images, An Introduction, John Wiley & Sons. 
Nieuwenhuis, G.J.A. and H. Kramer, 1995. Monitoring of 
Agricultural Crops with ERS-1 and JERS-1 Multi-temporal 
SAR Data and Crop Growth Modelling, Proc. of the Second 
ERS Applications Workshop, London, 6-8 December, UK, 
pp.25-29. 
Scotten, C.G.J., Van Rooy, W.W.L., and L.L.F. Janssen, 
1995. Assessment of the Capabilities of Multi-Temporal ERS- 
1 SAR Data to Discriminate Between Agricultural Crops, Int. 
J. Remote Sensing, Vol.16, No.14, 2619-2637. 
Williams, V.L., Philipson, W.R., and W.D. Philpot, 1987. 
Identifying Vegetable Crops with Landsat Thematic Mapper 
Data, Photogrametric Engineering and Remote Sensing, 
Vol.53, No.2, February, pp.187-191. 
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 565