摘要
提出了一种半监督稀疏多流形嵌入方法,并应用于高光谱影像分类.该方法充分利用少量标记和大量无标记样本,采用稀疏表示方法得到样本的稀疏系数,并选取来自同一流形的点作为近邻点,然后构建相似图来表征多流形结构,得到样本在每个流形上低维鉴别特征,增加来自同一流形的数据点聚集性,进而提升分类性能.本文方法在PaviaU和Salinas两个高光谱数据集上的总体分类准确度分别达到84.91%和89.74%,相较于其他方法明显提高了地物分类性能.
In this paper,a semi-supervised learning method called semi-supervised sparse multi-manifold embedding（S3 MME）was proposed for the classification of hyperspectral image.S3 MME exploits both labeled and unlabeled samples to adaptively find neighbors of each sample from the same manifold by using an optimization program based on sparse representation,which constructs an appropriate graph to characterize the manifold structure.Then it tries to extract discriminative features on each manifold in low dimensional space such that the data points in the same manifold become closer.The overall classification accuracies of the proposed method can reach 84.91% and 89.74% on PaviaU and Salinas hyperspectral data sets respectively, which significantly improves the classification of land cover compared with the conventional methods.
出处
《光子学报》
EI
CAS
CSCD
北大核心
2016年第3期126-132,共7页
Acta Photonica Sinica
基金
国家自然科学基金(Nos.41371338
61101168)
重庆市基础与前沿研究计划(No.cstc2013jcyjA40005)
中央高校基本科研业务费项目(Nos.106112013CDJZR125501
1061120131204)
重庆市研究生科研创新项目(No.CYB15052)资助~~
关键词
高光谱影像分类
维数约简
多流形
稀疏表示
半监督学习
Hyperspectral image classification
Dimensionality reduction
Multiple manifolds
Sparse representation
Semi-supervised learning
