Spatial-Spectral Semi-Supervised Local Discriminant Analysis for Hyperspectral Image Classification

被引：0

作者：

Hou B. ^{[1
]}

Yao M. ^{[1
]}

Wang R. ^{[1
]}

Zhang F. ^{[1
]}

Dai D. ^{[1
]}

机构：

[1] Department of Information Engineering, Rocket Force Engineering University, Xi'an, 710025, Shaanxi

来源：

Guangxue Xuebao/Acta Optica Sinica | 2017年 / 37卷 / 07期

关键词：

Hyperspectral image classification; Remote sensing; Semi-supervised local discriminant analysis; Spatial neighbor; Spatial-spectral distance;

D O I：

10.3788/AOS201737.0728002

中图分类号：

学科分类号：

摘要：

In traditional hyperspectral image classification algorithm based on feature extraction, spectral information is usually considered while spatial information is ignored. To address this problem, a hyperspectral image classification algorithm based on semi-supervised spatial-spectral local discriminant analysis (S3ELD) and spatial-spectral nearest neighbor (SSNN) classifier is proposed in this paper. Combining the spatial consistency of hyperspectral images and on the basis that the discriminant information of the labeled samples is used to maintain the separability of the data set, we define the spatial local pixel scatter matrix to preserve the spatial-domain neighborhood structures of pixel. A similarity measure method based on the spatial-spectral distance is then proposed to discover the local manifold structure and to construct SSNN. S3ELD algorithm not only reveals the local geometric relations of the data set but also enforces the compactness of the spectral-domain same class pixels and the spatial-domain local neighbor pixels in the low-dimension embedding space. Combining SSNN to classify, the classification accuracy is further enhanced. The experiments on the PaviaU and Salinas data sets show that the overall classification accuracy of S3ELD algorithm reaches 92.51% and 96.29%, respectively. Compared with several existing algorithms, the proposed algorithm can efficiently extract the information of discriminant characteristics and obtain higher classification accuracy. © 2017, Chinese Lasers Press. All right reserved.

引用

共 20 条

[1] Fauvel M., Tarabalka Y., Benediktsson A., Et al., Advances in spectral-spatial classification of hyperspectral images, Proceedings of the IEEE, 101, 3, pp. 652-675, (2013)
[2] Liu D., Han L., Han X., High spatial resolution remote sensing image classification based on deep learning, Acta Optica Sinica, 36, 4, (2016)
[3] Ye Z., Bai L., Nian Y., Hyperspectral image classification algorithm based on Gabor feature and locality-perserving dimension reduction, Acta Optica Sinica, 36, 10, (2016)
[4] Li Z., Zhang J., Huang H., Et al., Semi-supervised Laplace discriminant embedding for hyperspectral image classification, Journal of Electronics & Information Technology, 37, 4, pp. 995-1001, (2015)
[5] Bandos T., Bruzzone L., Camps-Valls G., Classification of hyperspectral images with regularized linear discriminant analysis, IEEE Transactions on Geoscience and Remote Sensing, 47, 3, pp. 862-873, (2009)
[6] Roweis S., Saul L., Nonlinear dimensionality reduction by locally linear embedding, Science, 290, 5500, pp. 2323-2326, (2000)
[7] Belkin M., Niyogi P., Laplacian eigenmaps and spectral techniques for embedding and clustering, Proceedings of the 14th International Conference on Neural Information Processing Systems: Natural and Synthetic, 14, 6, pp. 585-591, (2001)
[8] He X.F., Niyogi P., Locality preserving projections, Proceedings of Advance in Neural Information Processing Systems, 16, 1, pp. 153-160, (2005)
[9] He X.F., Cai D., Yan S.C., Et al., Neighborhood preserving embedding, Proceedings of the 10th IEEE International Conference on Computer Vision, 2, 23, pp. 1208-1213, (2005)
[10] Cai D., He X.F., Han J., Semi-supervised discriminant analysis, Proceedings of the 11th IEEE International Conference on Computer Vision, 5, pp. 1-7, (2008)

← 1 2 →