Evaluating neural networks and evidence pooling for land cover mapping

TY - JOUR

T1 - Evaluating neural networks and evidence pooling for land cover mapping

AU - Aitkenhead,M. J.

AU - Flaherty,S.

AU - Cutler,M. E. J.

PY - 2008/8

Y1 - 2008/8

N2 - The diversity of data sources, analysis methodologies, and classification systems has led to a number of new techniques for monitoring land-cover change. However, this wide choice means that it is difficult to know which solution to choose. A system capable of integrating the results of different analyses and applying them to land-cover mapping would therefore be extremely useful. This study investigates the use of evidence pooling and neural networks in land-cover mapping. Neural networks were used to classify land-cover using evidence from spectral (Landsat-7 ETM1), textural, and topographic information. Mapping was performed using combinations of evidence source and evidence pooling techniques. The best performance was achieved using all available information with a method that summed evidence directly instead of categorizing it. While the methodology failed to reach the level of accuracy recommended elsewhere, a comparison of the number of classes used with other methods showed that the system performed better than these approaches. © 2008 American Society for Photogrammetry and Remote Sensing.

AB - The diversity of data sources, analysis methodologies, and classification systems has led to a number of new techniques for monitoring land-cover change. However, this wide choice means that it is difficult to know which solution to choose. A system capable of integrating the results of different analyses and applying them to land-cover mapping would therefore be extremely useful. This study investigates the use of evidence pooling and neural networks in land-cover mapping. Neural networks were used to classify land-cover using evidence from spectral (Landsat-7 ETM1), textural, and topographic information. Mapping was performed using combinations of evidence source and evidence pooling techniques. The best performance was achieved using all available information with a method that summed evidence directly instead of categorizing it. While the methodology failed to reach the level of accuracy recommended elsewhere, a comparison of the number of classes used with other methods showed that the system performed better than these approaches. © 2008 American Society for Photogrammetry and Remote Sensing.

KW - Artificial intelligence

KW - Computer networks

KW - Conformal mapping

KW - Accuracy assessment

KW - Neural networks

KW - Classification

KW - Land cover

KW - Landsat

KW - Topographic mapping

M3 - Article

VL - 74

SP - 1019

EP - 1032

JO - Photogrammetric Engineering and Remote Sensing

T2 - Photogrammetric Engineering and Remote Sensing

JF - Photogrammetric Engineering and Remote Sensing

SN - 0099-1112

IS - 8

ER -