A remote sensing image classification method based on evidence theory and neural networks
Authors:
- 1 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1, 240-244 Vol.1
Abstract
Neural networks have been widely used in remote sensing image classification. In this paper, we exploited the spatial information of the image to decide the classification result and proposed a remote sensing image classification method based on D-S evidence theory and neural networks. First, the original image to be classified is smoothed with the smoothed image obtained. Next, a B-P neural network is used to train and classify the original image and its smoothed image separately. Next, the two classification results (decisions) of the B-P neural network are fused with evidence theory. Finally, the fused result is as the final classification result of the original image. Experiment results show that the new method is efficient and improves the classification accuracy largely.Keywords
This publication has 6 references indexed in Scilit:
- Neural-fuzzy classification for segmentation of remotely sensed imagesIEEE Transactions on Signal Processing, 1997
- A review and analysis of backpropagation neural networks for classification of remotely-sensed multi-spectral imageryInternational Journal of Remote Sensing, 1995
- Classification of multisensor remote-sensing images by structured neural networksIEEE Transactions on Geoscience and Remote Sensing, 1995
- Multispectral classification of Landsat-images using neural networksIEEE Transactions on Geoscience and Remote Sensing, 1992
- Neural Network Approaches Versus Statistical Methods In Classification Of Multisource Remote Sensing DataIEEE Transactions on Geoscience and Remote Sensing, 1990
- Upper and Lower Probabilities Induced by a Multivalued MappingThe Annals of Mathematical Statistics, 1967