Classification of Hyperspectral Images Based on Supervised Sparse Embedded Preserving Projection
Abstract
1. Introduction
2. Relation Work
2.1. Graph Embedding
2.2. Sparsity Preserving Projection
3. Supervised Sparse Embedded Preserving Projection
Algorithm 1: Supervised Sparse Embedded Preserving Projection (SSEPP) |
Input: The set of sample , error value , and dimensionality reduction . Output: The set of feature after dimension reduction . (1) Standardize the sample set; (2) Use the SPGL1 [17] algorithm to solve the sparse reconstruction coefficients using Equation (5); (3) Calculate the weight matrix of the labeled sample using Equation (11); (4) Use Equation (15) to calculate the objective function, which minimizes the error of samples of the same class. Then, obtain the multi-objective function according to Equation (16); (5) Obtain the projection matrix by solving the eigenvectors of the generalized eigenvalue decomposition using Equation (17); (6) Use Equation (18) to obtain the reduced-dimension sample feature set . End For |
4. Experimental Results and Analysis
4.1. Indian Pines Dataset
4.2. Zhangjiangkou Mangrove Nature Reserve HJ1A-HSI Dataset
5. Conclusions
Author Contributions
Funding
Acknowledgments
Conflicts of Interest
References
- Dalponte, M.; Bruzzone, L.; Vescovo, L.; Gianelle, D. The role of spectral resolution and classifier complexity in the analysis of hyperspectral images of forest areas. Remote Sens. Environ. 2009, 113, 2345–2355. [Google Scholar] [CrossRef]
- Gevaert, C.M.; Suomalainen, J.; Tang, J.; Kooistra, L. Generation of Spectral–Temporal Response Surfaces by Combining Multispectral Satellite and Hyperspectral UAV Imagery for Precision Agriculture Applications. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2015, 8, 3140–3146. [Google Scholar] [CrossRef]
- Manolakis, D.; Shaw, G. Detection algorithms for hyperspectral imaging applications. IEEE Signal Process. Mag. 2002, 19, 29–43. [Google Scholar] [CrossRef]
- Wei, H.; Zhang, H.; Zhang, L.; Philips, W.; Liao, W. Weighted Sparse Graph Based Dimensionality Reduction for Hyperspectral Images. IEEE Geosci. Remote Sens. Lett. 2017, 13, 686–690. [Google Scholar] [CrossRef]
- Zhang, X.; He, Y.; Jiao, L.; Liu, R.; Feng, J.; Zhou, S. Scaling cut criterion-based discriminant analysis for supervised dimension reduction. Knowl. Inf. Syst. 2015, 43, 633–655. [Google Scholar] [CrossRef]
- Abdi, H.; Williams, L.J. Principal component analysis. Wiley Interdiscip. Rev. Comput. Stat. 2010, 2, 433–459. [Google Scholar] [CrossRef]
- Fisher, R.A. The use of multiple measurements in taxonomic problems. Ann. Hum. Genet. 1936, 7, 179–188. [Google Scholar] [CrossRef]
- Bachmann, C.M.; Ainsworth, T.L.; Fusina, R.A. Exploiting manifold geometry in hyperspectral imagery. IEEE Trans. Geosci. Remote Sens. 2005, 43, 441–454. [Google Scholar] [CrossRef]
- Zhang, Y.; Du, B.; Zhang, L. A Sparse Representation-Based Binary Hypothesis Model for Target Detection in Hyperspectral Images. IEEE Trans. Geosci. Remote Sens. 2015, 53, 1346–1354. [Google Scholar] [CrossRef]
- Chen, P.; Jiao, L.; Liu, F.; Gou, S.; Zhao, J.; Zhao, Z. Dimensionality Reduction of Hyperspectral Imagery Using Sparse Graph Learning. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2017, 10, 1165–1181. [Google Scholar] [CrossRef]
- Sun, X.; Wang, J.; She, M.F.H.; Kong, L. Scale invariant texture classification via sparse representation. Neurocomputing 2013, 122, 338–348. [Google Scholar] [CrossRef]
- Zou, H.; Hastie, T.; Tibshirani, R. Sparse Principal Component Analysis. J. Comput. Graph. Stat. 2006, 15, 265–286. [Google Scholar] [CrossRef]
- Siddiqui, S.; Robila, S.; Peng, J.; Wang, D. Sparse Representations for Hyperspectral Data Classification. In Proceedings of the 2008 IEEE International Geoscience and Remote Sensing Symposium, Boston, MA, USA, 7–11 July 2008; pp. 577–580. [Google Scholar]
- Qiao, L.; Chen, S.; Tan, X. Sparsity preserving projections with applications to face recognition. Pattern Recognit. 2010, 43, 331–341. [Google Scholar] [CrossRef]
- Huang, H.; Yang, M. Dimensionality Reduction of Hyperspectral Images with Sparse Discriminant Embedding. IEEE Trans. Geosci. Remote Sens. 2015, 53, 5160–5169. [Google Scholar] [CrossRef]
- Yan, S.; Xu, D.; Zhang, B.; Zhang, H.-J.; Yang, Q.; Lin, S. Graph Embedding and Extensions: A General Framework for Dimensionality Reduction. IEEE Trans. Pattern Anal. Mach. Intell. 2007, 29, 40–51. [Google Scholar] [CrossRef] [PubMed]
- Van den Ewout, B.; Friedlander, M.P. SPGL1: A Solver for Large-Scale Sparse Reconstruction. 2007. Available online: https://friedlander.io/software/spgl1/ (accessed on 30 August 2019).
- Baumgardner, M.F.; Biehl, L.L.; Landgrebe, D.A. 220 Band AVIRIS Hyperspectral Image Data Set: June 12, 1992 Indian Pine Test Site 3. Available online: https://purr.purdue.edu/publications/1947/1 (accessed on 30 September 2015).
- He, W.; Zhang, H.; Zhang, L.; Shen, H. Total-Variation-Regularized Low-Rank Matrix Factorization for Hyperspectral Image Restoration. IEEE Trans. Geosci. Remote Sens. 2016, 54, 178–188. [Google Scholar] [CrossRef]
- China Centre for Resources Satellite Data and Application. HJ-1A/B/C. Available online: http://www.cresda.com/EN/satellite/7117.shtml (accessed on 3 November 2015).
- Soh, L.K.; Tsatsoulis, C. Texture analysis of SAR sea ice imagery using gray level co-occurrence matrices. IEEE Trans. Geosci. Remote Sens. 1999, 37, 780–795. [Google Scholar] [CrossRef]
- Deering, D.W. Rangeland Reflectance Characteristics Measured by Aircraft and Spacecraft Sensors. Ph.D. Thesis, Texas A&M University, College Station, TX, USA, 1978. [Google Scholar]
# | Class | Samples | Color (R,G,B) |
---|---|---|---|
C1 | Alfalfa | 46 | 255,254,137 |
C2 | Corn-notill | 1428 | 3,28,241 |
C3 | Corn-min | 830 | 255,89,1 |
C4 | Corn | 237 | 5,255,133 |
C5 | Grass/pasture | 483 | 255,5,251 |
C6 | Grass/trees | 730 | 89,1,255 |
C7 | Grass/pasture-mowed | 28 | 3,171,255 |
C8 | Hay-windrowed | 478 | 12,255,7 |
C9 | Oats | 20 | 172,175,84 |
C10 | Soybean-notill | 972 | 160,78,158 |
C11 | Soybean-min | 2455 | 101,173,255 |
C12 | Soybean-clean | 593 | 60,91,112 |
C13 | Wheat | 205 | 104,192,63 |
C14 | Woods | 1265 | 139,69,46 |
C15 | Bldg-Grass-Trees-Drives | 386 | 119,255,172 |
C16 | Stone-Steel towers | 93 | 254,255,3 |
# | Samples | DR+1-NN Classifier | ||||
---|---|---|---|---|---|---|
Train | Test | SPP | PCA | SDE | SSEPP | |
C1 | 5 | 41 | 14.63 | 68.29 | 70.73 | 65.85 |
C2 | 143 | 1285 | 43.97 | 45.21 | 62.02 | 68.72 |
C3 | 83 | 747 | 38.29 | 45.92 | 44.44 | 50.87 |
C4 | 24 | 213 | 12.68 | 30.99 | 21.13 | 23.00 |
C5 | 48 | 435 | 67.82 | 78.16 | 85.06 | 81.38 |
C6 | 73 | 657 | 86.00 | 88.43 | 92.69 | 95.74 |
C7 | 3 | 25 | 36.00 | 72.00 | 60.00 | 72.00 |
C8 | 48 | 430 | 61.16 | 95.58 | 95.35 | 99.77 |
C9 | 2 | 18 | 11.11 | 0.00 | 11.11 | 22.22 |
C10 | 97 | 875 | 42.17 | 60.46 | 56.34 | 57.83 |
C11 | 246 | 2209 | 59.48 | 67.41 | 66.82 | 75.69 |
C12 | 59 | 534 | 26.97 | 37.27 | 60.86 | 57.87 |
C13 | 21 | 184 | 86.96 | 91.85 | 95.65 | 98.91 |
C14 | 127 | 1138 | 74.87 | 84.27 | 92.62 | 93.76 |
C15 | 39 | 347 | 32.56 | 27.38 | 50.14 | 53.03 |
C16 | 9 | 84 | 28.57 | 82.14 | 73.81 | 80.95 |
Overall Accuracy (OA) /% | 54.15 | 63.73 | 69.06 | 73.31 |
Class Name | Land Cover Class | Description |
---|---|---|
C1 | Mangroves | Mangrove forests |
C2 | Upland vegetation | Deciduous or evergreen forest land, orchards, and tree groves |
C3 | Urban areas | Residential, commercial, industrial, and other developed land |
C4 | Water | Permanent open water, lakes reservoirs, bays, and estuaries |
C5 | Littoral zone | Land in the intertidal zone or the transitional zone |
C6 | Fallow land | Fields no longer under cultivation |
C7 | Agricultural land | Crop fields, paddy fields, and grasslands |
# | SPP | PCA | ||||||||||||||
C1 | C2 | C3 | C4 | C5 | C6 | C7 | Total | C1 | C2 | C3 | C4 | C5 | C6 | C7 | Total | |
C1 | 21 | 0 | 3 | 0 | 0 | 6 | 4 | 34 | 12 | 0 | 0 | 0 | 0 | 0 | 24 | 36 |
C2 | 4 | 50 | 11 | 0 | 0 | 10 | 6 | 81 | 0 | 50 | 0 | 0 | 0 | 0 | 0 | 50 |
C3 | 4 | 0 | 32 | 0 | 0 | 1 | 7 | 44 | 15 | 0 | 38 | 0 | 1 | 0 | 5 | 59 |
C4 | 0 | 0 | 0 | 47 | 1 | 0 | 0 | 48 | 0 | 0 | 0 | 47 | 0 | 0 | 0 | 47 |
C5 | 0 | 0 | 1 | 3 | 49 | 5 | 0 | 58 | 1 | 0 | 0 | 3 | 49 | 5 | 0 | 58 |
C6 | 0 | 0 | 0 | 0 | 0 | 26 | 6 | 32 | 0 | 0 | 0 | 0 | 0 | 37 | 0 | 37 |
C7 | 1 | 0 | 3 | 0 | 0 | 2 | 27 | 33 | 2 | 0 | 12 | 0 | 0 | 8 | 21 | 43 |
Total | 30 | 50 | 50 | 50 | 50 | 50 | 50 | 330 | 30 | 50 | 50 | 50 | 50 | 50 | 50 | 330 |
# | SDE | SSEPP | ||||||||||||||
C1 | C2 | C3 | C4 | C5 | C6 | C7 | Total | C1 | C2 | C3 | C4 | C5 | C6 | C7 | Total | |
C1 | 18 | 0 | 0 | 0 | 0 | 6 | 2 | 26 | 18 | 0 | 0 | 0 | 0 | 1 | 2 | 21 |
C2 | 0 | 50 | 11 | 0 | 0 | 10 | 0 | 71 | 0 | 50 | 11 | 0 | 0 | 10 | 0 | 71 |
C3 | 4 | 0 | 32 | 0 | 0 | 1 | 7 | 44 | 4 | 0 | 33 | 0 | 0 | 1 | 7 | 45 |
C4 | 0 | 0 | 0 | 47 | 1 | 0 | 0 | 48 | 0 | 0 | 0 | 47 | 1 | 0 | 0 | 48 |
C5 | 0 | 0 | 1 | 3 | 49 | 5 | 0 | 58 | 0 | 0 | 1 | 3 | 49 | 5 | 0 | 58 |
C6 | 3 | 0 | 0 | 0 | 0 | 26 | 6 | 35 | 3 | 0 | 0 | 0 | 0 | 31 | 6 | 40 |
C7 | 5 | 0 | 6 | 0 | 0 | 2 | 35 | 48 | 5 | 0 | 5 | 0 | 0 | 2 | 35 | 47 |
Total | 30 | 50 | 50 | 50 | 50 | 50 | 50 | 330 | 30 | 50 | 50 | 50 | 50 | 50 | 50 | 330 |
SPP | PCA | SDE | SSEPP | |
---|---|---|---|---|
OA/% | 76.36 | 76.97 | 77.88 | 79.70 |
Kappa | 0.7235 | 0.7307 | 0.7407 | 0.7618 |
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Cai, F.; Guo, M.-X.; Hong, L.-F.; Huang, Y.-Y. Classification of Hyperspectral Images Based on Supervised Sparse Embedded Preserving Projection. Appl. Sci. 2019, 9, 3583. https://doi.org/10.3390/app9173583
Cai F, Guo M-X, Hong L-F, Huang Y-Y. Classification of Hyperspectral Images Based on Supervised Sparse Embedded Preserving Projection. Applied Sciences. 2019; 9(17):3583. https://doi.org/10.3390/app9173583
Chicago/Turabian StyleCai, Fen, Miao-Xia Guo, Li-Fang Hong, and Ying-Yi Huang. 2019. "Classification of Hyperspectral Images Based on Supervised Sparse Embedded Preserving Projection" Applied Sciences 9, no. 17: 3583. https://doi.org/10.3390/app9173583
APA StyleCai, F., Guo, M.-X., Hong, L.-F., & Huang, Y.-Y. (2019). Classification of Hyperspectral Images Based on Supervised Sparse Embedded Preserving Projection. Applied Sciences, 9(17), 3583. https://doi.org/10.3390/app9173583