By Topic

RBF Neural Network Supported Classification of Remote Sensing Images Based on TM/ETM+ in Nanjing

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Cao Kai ; Dept. of Geogr. & Resource Manage., Chinese Univ. of Hong Kong, Hong Kong ; Huang Bo ; Lu Heng ; Liu Biao

The classification of remote sensing images is more and more important along with the development of society and economy. According to the defects general classification methods have, such as the accuracy, the efficiency etc, the design of `robust' classification system based on a Gaussian RBF neural Network is used in this article to classify the TM/ETM+ image in Nanjing. The choice of this neural network model is justified by some of its particular properties, i. e., local learning, fast training phase, ability to recognize when an input pattern has fallen into a region of the input space without training data, and capability to provide high classification accuracies on remote sensing images. For appraising the precision of the model in brief, over 1000 examples are chosen in this research, and the result shows that in the whole research area there is obvious improvement (86.6-89.7%) between MLC and this model. Besides, it is also better than the MLP NN model (87.9-89.7%). The result indicates that the model of RBF NN is a good approach for the classification of remote sensing in this area based on TM/ETM+. Of course, there are also many aspects need to be revised and improved in the future research such as the accuracy and for other data source.

Published in:

Geoscience and Remote Sensing Symposium, 2008. IGARSS 2008. IEEE International  (Volume:4 )

Date of Conference:

7-11 July 2008