摘要
基于BP算法的神经网络方法目前已广泛运用于遥感影像分类,提出一种主成分分析(PCA)与BP神经网络相结合的遥感影像分类方法——PCA-BPNN,实验证明该方法是可行并且有效的,在减少计算量和加快收敛的同时,提高了分类的精度。
A neural network based on back propagation has been widely used in the classification of remote sensing images. In this paper, PCA-BPNN is proposed for classification of multispectral remote sesing images, which combines principal component analysis with BP neural networks. The experiment results demonstrate its feasibility and it can cut down training time and improve the accuracy.
出处
《地理空间信息》
2006年第1期15-17,共3页
Geospatial Information
关键词
BP神经网络
影像分类
主成分分析
BP neural network
image classification
principal component analysis