CONCLUSIONS
Multi-temporal ERS-1 synthetic aperture radar (SAR) and
Landsat TM data were used to evaluate an artificial neural
network approach for crop classification. Six major crops,
i.e., winter wheat, com (good growth & poor growth),
soybeans (good growth & poor growth), barley/oats, alfalfa,
and pasture/cut-hay-alfalfa, were classified into eight classes.
The results show that both a single-date and multi-temporal
SAR data yielded poor classification accuracies using a
maximum likelihood classifier (MLC). With per-field
approach using a feed forward artificial neural network
(ANN), the overall classification accuracy of three-date SAR
data improved almost 20%, and the best classification of a
single-date (Aug. 5) SAR data improved the overall accuracy
by about 26%. These accuracies (<60%), however, were not
high enough for operational crop inventory and analysis.
Using the combination of TM3,4,5 and Aug. 5 SAR data, the
best per-field ANN classification of 96.8% was achieved. It
represents a 8.5% improvement over a single TM3,4,5
classification alone. It also represents a 5% increase over
the best per-pixel classification. This indicates that a
combination of mid-season SAR and VIR data was best suited
for crop classification. The results also show that the best
ANN classification had a 5% higher accuracy than a
minimum distance (MD) classification using the same
dataset.
REFERENCES
References from Journals:
Ban, Y., P.M. Treitz and P.J. Howarth, 1995. Improving the
accuracy of synthetic aperture radar for agricultural crop
classification. Canadian Journal of Remote Sensing, Special
Issue, SAR on Vegetation and Landuse Studies , 21( 2), pp.
158-164.
Benediktsson J.A, P.H. Swain and O.K. Ersoy, 1990a. Neural
networks approaches versus statistical methods in
classification of multi-source remote sensing data. IEEE
Transactions on Geoscience and Remote Sensing, GE-28(4),
pp. 540-552.
Brisco B. and R.J. Brown, 1995. Multidate SAR/TM
synergism for crop classification in western Canada.
Photogrammetric Engineering and Remote Sensing, 61(8),
pp. 1009-1014.
Brisco, B., R.J. Brown, and M.J. Manore, 1989. Early
season crop discrimination with combined SAR and TM data.
Canadian Journal of Remote Sensing, 15(1), pp. 44-54.
Foody, G., M.B. McCulloch, and W.B. Yates, 1994. Crop
classification from C-band polarmetric radar data.
International Journal of Remote Sensing, 15(14), pp. 2871-
2885.
Foody, G., M.B. McCulloch, and W.B. Yates, 1995.
Classification of remotely sensed data by an artificial neural
network: issues related to training data characteristics.
Photogrammetric Engineering and Remote Sensing, 16(9),
pp. 1707-1723.
Guindon, B., P.M. Teillet, D. G. Goodenough, J.J. Palimaka
and A. Sieber, 1984. Evaluation of the crop classification
performance of X, L and C-band SAR imagery. Canadian
Journal of Remote Sensing, 10(1), pp. 4-16.
52
Kanellopoulos L, A. Varfis, G.G. Wilkinson and J. Mégier,
1991. Neural network classification of multi-date satellite
imagery. IEEE Transactions on Geoscience and Remote
Sensing, pp. 2215-2218.
Li, R. and H. Si, 1992. Multi-spectral image classification
using improved backpropagation neural networks. IEEE
Transactions on Geoscience and Remote Sensing, pp. 1078-
1080.
Liu, Z.K. and J.Y. Xiao, 1991. Classification of remotely
sensed image data using artificial neural networks.
International Journal of Remote Sensing, Vol.12, pp.2433-
2438.
References from Books:
Schalkoff R., 1992. Pattern Recognition: Statistical,
Structural and Neural Approaches, John Wiley & Sons Inc.,,
New York, Toronto, Singapore, pp.204-258.
References from Other Literature:
Benediktsson J.A, P.H. Swain, O.K. Ersoy and D. Hong,
1990b. Classification of very high dimensional data using
neural networks. Proceedings, IGASS'90, Vol.2, pp.1269-
1272.
Dixon, R. and A. Mack, 1990. Crop discrimination in the
carberry area with combined SAR and TM data”, In
Proceeding, 14th Canadian Remote Sensing Symposium, pp.
118-123.
Fiumara, A. and Pierdicca, N., 1989. Evaluation of
classification results obtained with combined multi-temporal
optical and microwave data. IGARSS'89, Vol2, pp. 787-
7990.
Fog B., J. Poulsen, I. Sandholt and H. M. Skriver, 1993.
Monitoring land cover and crop types in Denmark using
ERS-1 SAR and optical satellite images. In Proceeding,
International Symposium from Optics to Radar, SPOT and
ERS Applications, Paris, France, CNES series no. 312, pp.
71-80.
Hirose T., J. Cihlar, and A. Mack, 1984. Discrimination of
agricultural crops in Canada with SAR plus VIR data: initial
results. Proceeding, 8th Canadian Remote Sensing
Symposium, pp. 387-403.
Kohl H., C. King, and H.D. Groof, 1993. Agricultural
Statistics: comparison of ERS-1 and SPOT for the crop
acreage estimation of the MARS-project. Proceedings of the
Second ERS-1 Symposium: Space at the Service of Our
Environment, Hamburg, Germany, pp87-92.
PCI, 1994. Using PCI Software.
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B7. Vienna 1996