Texture Classification of Aerial Image Based on Bayesian Network Augmented Naive Bayes

Classification is an open, old and basic problem in many domains. Recently, a lot of new methods come forth, such as Bayesian Networks. Bayesian Networks, one of probabilistic networks, are a powerful data mining technique for handling uncertainty in complex domains. In this paper, we apply Bayesian Networks Augmented Naive Bayes (BAN) to texture classification of aerial image and propose a new method to construct the network topology structure in terms of training accuracy based on the training samples. In order to validate the feasibility and the effectivity, we compare BAN to Naive Bayes Classifiers (NBC) and PCA-NBC. Thus six pieces of 23cm×23cm aerial image about Australia and ten pieces of 23cm×23cm aerial image about Wuhan city in China are used in the experiments. Experimental results demonstrate BAN outperform than NBC and PCA-NBC in the overall classification accuracy. Although it is time consuming, it will be an attractive and effective method in the future.