A comparison of neural network, rough sets and support vector machine on remote sensing image classification

作者：Xiao, Hang; Zhang, Xiubin^*

来源：Journal of Computational Information Systems, 2008, 4(6): 2555-2564.

• Incomplete data
• Kernel function
• Remote sensing image classification
• Variable precision rough sets model

This paper first reviewed the relevant theories of neural network, rough sets and support vector machine (SVM). All of them have great advantages on dealing with various imprecise and incomplete data. However, there exists essential difference among them. Except for neural network, rough sets and support vector machine are seldom used in the field of remote sensing image classification. How to combine the theories with the application of remote sensing is an important tendency in the later research. In the paper, neural network, rough sets and support vector machine are applied to the area of remote sensing image classification. Different networks, thresholds and kernel functions are used in three methods respectively for the purpose of comparing the experimental results. The paper provides us a new viewpoint on remote sensing image classification in the future work.

• 出版日期2008
• 单位上海交通大学