摘要
提出了一种在期望最大化(EM)算法框架下同时实现混合概率主成分分析(PPCA)降维和聚类的高光谱图像非监督分类方法。它根据不同类别应各有自己代表性的特征集,将通常意义下的特征抽取和模式分类合并在一步内完成,尽可能地保留了可分性;同时该方法具有概率模型的优点,更适合高维数据处理。采用仿真数据和真实数据进行的比较实验表明,该算法较一般不加区分地对所有原始数据进行PCA降维再分类的方法能得到更好的分类结果。
An unsupervised hyperspectral image classification method simultaneously realizing the mixture of probabilistic PCA and clustering under the frame of EM algorithm is proposed. It is based on the fa ct that different class should have its own representative feature set, and it r eali zes feature extraction and classification in one step while preserving as much s eparability. It also possesses the advantages of PPCA model, which is more effec tive to high dimensional data processing. Applying the method to simulated data and real data shows that it can achieve better results compared with the method that applies PCA to all data without differentiation among classes.
出处
《国防科技大学学报》
EI
CAS
CSCD
北大核心
2005年第2期61-64,共4页
Journal of National University of Defense Technology
关键词
非监督分类
降维
混合概率主成分分析
期望最大化算法
unsupervised classification
dimensionality reduction
mixture of Probabilistic Principal Component Analysis (PPCA)
EM (Expectation Maximization) algorithm
