基于半监督稀疏流形嵌入的高光谱影像特征提取

罗甫林; 黄鸿; 刘嘉敏; 冯海亮

doi:10.11999/JEIT151340

基于半监督稀疏流形嵌入的高光谱影像特征提取

doi: 10.11999/JEIT151340

基金项目:

重庆市研究生科研创新项目(CYB15052)，国家自然科学基金(41371338)，重庆市基础与前沿研究计划(cstc2013jcyjA40005)，中央高校基本科研业务费项目(106112013CDJZR125501, 1061120131204)

计量
- 文章访问数: 1411
- HTML全文浏览量: 120
- PDF下载量: 639
- 被引次数: 0
出版历程
- 收稿日期: 2015-11-23
- 修回日期: 2016-03-18
- 刊出日期: 2016-09-19

Feature Extraction of Hyperspectral Image Using Semi-supervised Sparse Manifold Embedding

Funds:

The Chongqing Postgraduates Innovation Project (CYB15052), The National Natural Science Foundation of China (41371338), The Basic and Advanced Research Program of Chongqing (cstc2013jcyjA40005), The Fundamental Research Funds for the Central Universities (106112013CDJZR125501, 1061120131204)

摘要

摘要: 高光谱影像具有波段数多、冗余度高的特点，因此特征提取成为高光谱影像分类的研究热点。针对此问题，该文提出一种半监督稀疏流形嵌入(S3ME)算法，该方法充分利用标记样本和无标记样本，通过基于切空间的稀疏流形表示来自适应地揭示数据间的相似关系，并利用稀疏系数构建一个半监督相似图。在此基础上，增加了图中同类标记样本的权重，然后在低维空间中保持图的相似关系不变，并最小化加权距离和，获得投影矩阵实现特征提取。S3ME方法不仅能揭示数据间的稀疏流形结构，而且增强了同类数据的集聚性，能有效提取出鉴别特征，改善分类效果。该文提出的S3ME方法在PaviaU和Salinas高光谱数据集上的总体分类精度分别达到84.62%和88.07%，相比传统特征提取方法提升了地物分类性能。
- 高光谱数据 /
- 特征提取 /
- 半监督学习 /
- 稀疏流形嵌入
Abstract: Hyperspectral image contains the properties of much bands and high redundancy, and the research of hyperspectral image classification focuses on feature extraction. To overcome this problem, a Semi-Supervised Sparse Manifold Embedding (S3ME) algorithm is proposed in this paper. The S3ME method makes full use of labeled and unlabeled samples to adaptively reveal the similarity relationship between data with the sparse representation of tangent space. It constructs a semi-supervised similarity graph via the sparse coefficients and enhances the weight between labeled samples from the same class. In a low-dimensional embedding space, S3ME preserves the similarity of graph to minimize the sum of the weighted distance. Then, it obtains a projection matrix for feature extraction. S3ME not only reveals the sparse manifold structure of data but also enhances the compactness of the same class data, which can effectively extract the discriminating feature and improve the classification performance. The overall classification accuracies of the proposed S3ME method respectively reach 84.62% and 88.07% on the PaviaU and Salinas hyperspectral data sets, and the classification performance of land cover is improved compared with the traditional feature extraction methods.
- Hyperspectral data /
- Feature extraction /
- Semi-supervised learning /
- Sparse manifold embedding

HTML全文

参考文献(23)

CHANG Y L, LIU J N, HAN C C, et al. Hyperspectral image classification using nearest feature line embedding approach [J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(1): 278-287. doi: 10.1109/TGRS.2013.2238635.

XUE Z H, DU P J, LI J, et al. Simultaneous sparse graph embedding for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(11): 6114-6133. doi: 10.1109/TGRS.2015.2432059.

李志敏, 张杰, 黄鸿, 等. 面向高光谱图像分类的半监督Laplace鉴别嵌入[J]. 电子与信息学报, 2015, 37(4): 995-1001. doi: 10.11999/JEIT140600.

LI Zhimin, ZHANG Jie, HUANG Hong, et al. Semi- supervised Laplace discriminant embedding for hyperspectral image classification[J]. Journal of Electronics Information Technology, 2015, 37(4): 995-1001. doi: 10.11999/ JEIT140600.

宋相法, 焦李成. 基于稀疏表示及光谱信息的高光谱遥感图像分类[J]. 电子与信息学报, 2012, 34(2): 268-272. doi: 10.3724/SP.J.1146.2011.00540.

SONG Xiangfa and JIAO Licheng. Classification of hyperspectral remote sensing image based on sparse representation and spectral information[J]. Journal of Electronics Information Technology, 2012, 34(2): 268-272. doi: 10.3724/SP.J.1146.2011.00540.

FENG Z X, YANG S Y, WANG S G, et al. Discriminative spectral-spatial margin-based semi-supervised dimensionality reduction of hyperspectral data[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(2): 224-228. doi: 10.1109/ LGRS.2014.2327224.

ROWEIS S T and SAUL L K. Nonlinear dimensionality reduction by locally linear embedding[J]. Science, 2000, 290(5500): 2323-2326. doi: 10.1126/science.290.5500.2323.

BELKIN M and NIYOGI P. Laplacian eigenmaps for dimensionality reduction and data representation[J]. Neural Computation, 2003, 15(6): 1373-1396. doi: 10.1162/ 089976603321780317.

HE X F, CAI D, YAN S C, et al. Neighborhood preserving embedding[C]. IEEE International Conference on Computer Vision, Beijing, 2005: 1208-1213. doi: 10.1109/ICCV.2005. 167.

HE X F and NIYOGI P. Locality preserving projections[C]. Advances in Neural Information Processing Systems, Whistler, B. C., Canada, 2003: 153-160.

YAN S C, XU D, ZHANG B Y, et al. Graph embedding and extensions: A general framework for dimensionality reduction [J]. IEEE Transactions on Pattern Analysis Machine Intelligence, 2007, 29(1): 40-51. doi: 10.1109/CVPR.2005. 170.

SHAO Z and ZHANG L. Sparse dimensionality reduction of hyperspectral image based on semi-supervised local Fisher discriminant analysis[J]. International Journal of Applied Earth Observation and Geoinformation, 2014, 31: 122-129. doi: 10.1016/j.jag.2014.03.015.CHEN X B, CAI Y F, CHEN

CHEN X B, CAI Y F, CHEN L, et al. Discriminant feature extraction for image recognition using complete robust maximum margin criterion[J]. Machine Vision and Applications, 2015, 26(7): 857-870. doi: 10.1007/s00138-015- 0709-7.

BACHMANN C M, AINSWORTH T L, and FUSINA R A. Exploiting manifold geometry in hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(3): 441-454. doi: 10.1109/TGRS.2004.842292.

QIAO L S, CHEN S C, and TAN X Y. Sparsity preserving projections with applications to face recognition[J]. Pattern Recognition, 2010, 43(1): 331-341. doi: 10.1016/j.patcog. 2009.05.005.

ELHAMIFAR E and VIDAL R. Sparse manifold clustering and embedding[C]. Advances in Neural Information Processing Systems, Granada, Spain, 2011: 55-63.

HUANG H and YANG M. Dimensionality reduction of hyperspectral images with sparse discriminant embedding[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(9): 5160-5169. doi: 10.1109/TGRS.2015.2418203.

LU G F, JIN Z, and ZOU J. Face recognition using discriminant sparsity neighborhood preserving embedding[J]. Knowledge-Based Systems, 2012, 31(7): 119-127. doi: 10.1016/j.knosys.2012.02.014.

ZANG F and ZHANG J S. Discriminative learning by sparse representation for classification[J]. Neurocomputing, 2011, 74: 2176-2183. doi: 10.1016/j.neucom.2011.02.012.

SONG Y Q, NIE F P, ZHANG C S, et al. A unified framework for semi-supervised dimensionality reduction[J]. Pattern Recognition, 2008, 41(9): 2789-2799. doi: 10.1016/j. patcog.2008.01.001.

SONG Y Q, NIE F P, and ZHANG C S. Semi-supervised sub-manifold discriminant analysis[J]. Pattern Recognition Letters, 2008, 29(13): 1806-1813. doi: 10.1016/j.patrec.2008. 05.024.

ZHAO M B, LI B, WU Z, et al. Image classification via least square semi-supervised discriminant analysis with flexible kernel regression for out-of-sample extension[J]. Neurocomputing, 2015(153): 96-107. doi: 10.1016/j.neucom.2014.11.048.

施引文献

资源附件(0)

访问统计