TY - JOUR
T1 - On the use of consensus algorithms to address variability in the results of neural network classifications
T2 - Preliminary tests involving two northern study areas
AU - Leverington, David W.
AU - Moon, Wooil M.
N1 - Funding Information:
The helpful comments of two anonymous reviewers are appreciated. This work was supported in part by the Geological Society of America, the Northern Scientific Training Program (Department of Indian Affairs and Northern Development Canada), Falconbridge Ltd., and the Smithsonian Institution (D.W.L.), and by a Natural Sciences and Engineering Research Council of Canada Discovery Grant (A-7400) (W.M.M.).
PY - 2005/8
Y1 - 2005/8
N2 - In past work it has been recognized that variations in parameters such as learning rate, momentum, and network architecture can influence the results in neural network classifications of satellite images. New tests suggest that variation in the results of neural network classifications, caused solely by differences in weight initializations, can also be substantial. This issue has the potential to limit the applicability of neural networks in remote sensing classifications. The negative effects of variation in neural network results can potentially be reduced or eliminated through application of consensus algorithms in which the outputs of multiple neural network classifications are combined. Research results presented here were based on training and test data with low sample sizes for many classes and, accordingly, the results must be interpreted with caution. Early results using majority-vote and evidential-reasoning consensus algorithms, however, suggest that near-optimum neural network classification accuracies can be achieved through application of these algorithms.
AB - In past work it has been recognized that variations in parameters such as learning rate, momentum, and network architecture can influence the results in neural network classifications of satellite images. New tests suggest that variation in the results of neural network classifications, caused solely by differences in weight initializations, can also be substantial. This issue has the potential to limit the applicability of neural networks in remote sensing classifications. The negative effects of variation in neural network results can potentially be reduced or eliminated through application of consensus algorithms in which the outputs of multiple neural network classifications are combined. Research results presented here were based on training and test data with low sample sizes for many classes and, accordingly, the results must be interpreted with caution. Early results using majority-vote and evidential-reasoning consensus algorithms, however, suggest that near-optimum neural network classification accuracies can be achieved through application of these algorithms.
UR - http://www.scopus.com/inward/record.url?scp=30644474645&partnerID=8YFLogxK
U2 - 10.5589/m05-017
DO - 10.5589/m05-017
M3 - Article
AN - SCOPUS:30644474645
VL - 31
SP - 269
EP - 273
JO - Canadian Journal of Remote Sensing
JF - Canadian Journal of Remote Sensing
SN - 0703-8992
IS - 4
ER -