Dimensionality Reduction and Feature Selection for Object-Based Land Cover Classification based on Sentinel-1 and Sentinel-2 Time Series Using Google Earth Engine
"> Figure 1
<p>Stockholm (left) and Beijing (right) study site extents.</p> "> Figure 2
<p>Generated segments: (left) high-density and low-density built-up, transportation, water, and urban green space land cover/land use classes in Stockholm’s urban center; (right) agricultural, forest, water, and wetland land cover/land use classes in Stockholm’s periphery.</p> "> Figure 3
<p>General workflow with the three-step approach of feature computation, classifier optimization, and classification.</p> "> Figure 4
<p>Features selection and classifier optimization workflow.</p> "> Figure 5
<p>Heat map of the search space in the grid search. Left: Highest accuracy at the edge of the search space. Right: Adjusted search space.</p> "> Figure 6
<p>Overall accuracy as function of the number of features used (Stockholm).</p> "> Figure 7
<p>Normalized confusion matrices of the predictions on testing data (Stockholm).</p> "> Figure 8
<p>Prediction times as function of the number of features used for different methods (Stockholm).</p> "> Figure 9
<p>Learning curves showing the influence of training set sizes (Stockholm).</p> "> Figure 10
<p>Display of input images and a resulting land cover map.</p> "> Figure 11
<p>Overall accuracy as function of the numbers of features (Beijing).</p> "> Figure 12
<p>Normalized confusion matrices of the predictions on testing data (Beijing).</p> "> Figure 13
<p>Averaged feature importance ranking grouped by S1 and S2 features (Stockholm left, Beijing right).</p> "> Figure 14
<p>Classes histograms for two geometric features (Stockholm).</p> "> Figure 15
<p>Scatterplots (S1, S2, S1&S2) of differently combined high-ranking features (Stockholm).</p> "> Figure 15 Cont.
<p>Scatterplots (S1, S2, S1&S2) of differently combined high-ranking features (Stockholm).</p> ">
Abstract
:1. Introduction
2. Study Areas and Data Description
3. Methodology
3.1. Feature Set and Classifier Optimization
- The first step is to scale the input features. Scaling ensures that all features have the similar ranges of value, which are beneficial or even essential for some dimensionality reduction methods as well as for the SVM classifier. We applied scaler that uses the second and third quantiles of the data to rescale the data linearly in the range [0,1].
- The second step is a reduction of the feature set dimensionality either through the feature extraction or feature selection methods.
- Given the selected feature set, the last step is to find, via grid search, the optimum set of hyperparameters for the SVM classifier.
3.1.1. Data Sets and Cross-Validation
3.1.2. Dimensionality Reduction Step
3.1.3. SVM Hyperparameters Estimation
3.1.4. Non-Inferiority Test
3.2. Classification
4. Results and Discussion
4.1. Stockholm Study Area
4.2. Beijing Study Area
4.3. Comparison of the Study Areas
5. Conclusions
Author Contributions
Funding
Acknowledgments
Conflicts of Interest
References
- Haack, B.; Bryant, N.; Adams, S. An assessment of landsat MSS and TM data for urban and near-urban land-cover digital classification. Remote Sens. Environ. 1987, 21, 201–213. [Google Scholar] [CrossRef]
- Quarmby, N.A.; Cushnie, J.L. Monitoring urban land cover changes at the urban fringe from SPOT HRV imagery in south-east England. Int. J. Remote Sens. 1989, 10, 953–963. [Google Scholar] [CrossRef]
- Zha, Y.; Gao, J.; Ni, S. Use of normalized difference built-up index in automatically mapping urban areas from TM imagery. Int. J. Remote Sens. 2003, 24, 583–594. [Google Scholar] [CrossRef]
- Choi, J. A hybrid approach to urban land use/cover mapping using Landsat 7 Enhanced Thematic Mapper Plus (ETM+) images. Int. J. Remote Sens. 2004, 25, 2687–2700. [Google Scholar] [CrossRef]
- Yuan, F.; Sawaya, K.E.; Loeffelholz, B.C.; Bauer, M.E. Land cover classification and change analysis of the Twin Cities (Minnesota) Metropolitan Area by multitemporal Landsat remote sensing. Remote Sens. Environ. 2005, 98, 317–328. [Google Scholar] [CrossRef]
- Zhong, P.; Wang, R. A Multiple Conditional Random Fields Ensemble Model for Urban Area Detection in Remote Sensing Optical Images. IEEE Trans. Geosci. Remote Sens. 2007, 45, 3978–3988. [Google Scholar] [CrossRef]
- Pierce, L.E.; Ulaby, F.T.; Sarabandi, K.; Dobson, M.C. Knowledge-based classification of polarimetric SAR images. IEEE Trans. Geosci. Remote Sens. 1994, 32, 1081–1086. [Google Scholar] [CrossRef]
- Dell’Acqua, F.; Gamba, P. Discriminating urban environments using multiscale texture and multiple SAR images. Int. J. Remote Sens. 2006, 27, 3797–3812. [Google Scholar] [CrossRef]
- Dell’Acqua, F.; Gamba, P. Texture-based characterization of urban environments on satellite SAR images. IEEE Trans. Geosci. Remote Sens. 2003, 41, 153–159. [Google Scholar] [CrossRef]
- Brenner, A.R.; Roessing, L. Radar Imaging of Urban Areas by Means of Very High-Resolution SAR and Interferometric SAR. IEEE Trans. Geosci. Remote Sens. 2008, 46, 2971–2982. [Google Scholar] [CrossRef]
- Niu, X.; Ban, Y. Multi-temporal RADARSAT-2 polarimetric SAR data for urban land-cover classification using an object-based support vector machine and a rule-based approach. Int. J. Remote Sens. 2013, 34, 1–26. [Google Scholar] [CrossRef]
- Salehi, M.; Sahebi, M.R.; Maghsoudi, Y. Improving the Accuracy of Urban Land Cover Classification Using Radarsat-2 PolSAR Data. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2014, 7, 1394–1401. [Google Scholar] [CrossRef]
- Zhou, Y.; Wang, H.; Xu, F.; Jin, Y. Polarimetric SAR Image Classification Using Deep Convolutional Neural Networks. IEEE Geosci. Remote Sens. Lett. 2016, 13, 1935–1939. [Google Scholar] [CrossRef]
- Dekker, R.J. Texture analysis and classification of ERS SAR images for map updating of urban areas in The Netherlands. IEEE Trans. Geosci. Remote Sens. 2003, 41, 1950–1958. [Google Scholar] [CrossRef]
- Su, W.; Li, J.; Chen, Y.; Liu, Z.; Zhang, J.; Low, T.M.; Suppiah, I.; Hashim, S.A.M. Textural and local spatial statistics for the object-oriented classification of urban areas using high resolution imagery. Int. J. Remote Sens. 2008, 29, 3105–3117. [Google Scholar] [CrossRef]
- Engdahl, M.E.; Hyyppa, J.M. Land-cover classification using multitemporal ERS-1/2 InSAR data. IEEE Trans. Geosci. Remote Sens. 2003, 41, 1620–1628. [Google Scholar] [CrossRef]
- Guerschman, J.P.; Paruelo, J.M.; Bella, C.D.; Giallorenzi, M.C.; Pacin, F. Land cover classification in the Argentine Pampas using multi-temporal Landsat TM data. Int. J. Remote Sens. 2003, 24, 3381–3402. [Google Scholar] [CrossRef]
- Camps-Valls, G.; Gomez-Chova, L.; Munoz-Mari, J.; Rojo-Alvarez, J.L.; Martinez-Ramon, M. Kernel-Based Framework for Multitemporal and Multisource Remote Sensing Data Classification and Change Detection. IEEE Trans. Geosci. Remote Sens. 2008, 46, 1822–1835. [Google Scholar] [CrossRef]
- Tigges, J.; Lakes, T.; Hostert, P. Urban vegetation classification: Benefits of multitemporal RapidEye satellite data. Remote Sens. Environ. 2013, 136, 66–75. [Google Scholar] [CrossRef]
- Waske, B.; van der Linden, S. Classifying Multilevel Imagery from SAR and Optical Sensors by Decision Fusion. IEEE Trans. Geosci. Remote Sens. 2008, 46, 1457–1466. [Google Scholar] [CrossRef]
- Amarsaikhan, D.; Blotevogel, H.H.; Genderen, J.L.; van Ganzorig, M.; Gantuya, R.; Nergui, B. Fusing high-resolution SAR and optical imagery for improved urban land cover study and classification. Int. J. Image Data Fusion 2010, 1, 83–97. [Google Scholar] [CrossRef]
- Corbane, C.; Faure, J.-F.; Baghdadi, N.; Villeneuve, N.; Petit, M. Rapid Urban Mapping Using SAR/Optical Imagery Synergy. Sensors 2008, 8, 7125–7143. [Google Scholar] [CrossRef] [PubMed]
- Pacifici, F.; Frate, F.D.; Emery, W.J.; Gamba, P.; Chanussot, J. Urban Mapping Using Coarse SAR and Optical Data: Outcome of the 2007 GRSS Data Fusion Contest. IEEE Geosci. Remote Sens. Lett. 2008, 5, 331–335. [Google Scholar] [CrossRef] [Green Version]
- Ban, Y.; Yousif, O.; Hu, H. Fusion of SAR and Optical Data for Urban Land Cover Mapping and Change Detection. In Global Urban Monitoring and Assessment through Earth Observation; Weng, Q., Ed.; CRC Press: Boca Raton, FL, USA, 2014; pp. 353–386. [Google Scholar] [CrossRef]
- Makarau, A.; Palubinskas, G.; Reinartz, P. Multi-sensor data fusion for urban area classification. In Proceedings of the 2011 Joint Urban Remote Sensing Event, Munich, Germany, 11–13 April 2011; pp. 21–24. [Google Scholar] [CrossRef] [Green Version]
- Zhang, H.; Lin, H.; Li, Y. Impacts of Feature Normalization on Optical and SAR Data Fusion for Land Use/Land Cover Classification. IEEE Geosci. Remote Sens. Lett. 2015, 12, 1061–1065. [Google Scholar] [CrossRef]
- Zhang, Y.; Zhang, H.; Lin, H. Improving the impervious surface estimation with combined use of optical and SAR remote sensing images. Remote Sens. Environ. 2014, 141, 155–167. [Google Scholar] [CrossRef]
- Zhu, Z.; Woodcock, C.E.; Rogan, J.; Kellndorfer, J. Assessment of spectral, polarimetric, temporal, and spatial dimensions for urban and peri-urban land cover classification using Landsat and SAR data. Remote Sens. Environ. Remote Sens. Urban Environ. 2012, 117, 72–82. [Google Scholar] [CrossRef]
- Ban, Y.; Hu, H.; Rangel, I.M. Fusion of Quickbird MS and RADARSAT SAR data for urban land-cover mapping: object-based and knowledge-based approach. Int. J. Remote Sens. 2010, 31, 1391–1410. [Google Scholar] [CrossRef]
- Myint, S.W.; Gober, P.; Brazel, A.; Grossman-Clarke, S.; Weng, Q. Per-pixel vs. object-based classification of urban land cover extraction using high spatial resolution imagery. Remote Sens. Environ. 2011, 115, 1145–1161. [Google Scholar] [CrossRef]
- Bruce, L.M.; Koger, C.H.; Li, J. Dimensionality reduction of hyperspectral data using discrete wavelet transform feature extraction. IEEE Trans. Geosci. Remote Sens. 2002, 40, 2331–2338. [Google Scholar] [CrossRef]
- Harsanyi, J.C.; Chang, C.-I. Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach. IEEE Trans. Geosci. Remote Sens. 1994, 32, 779–785. [Google Scholar] [CrossRef] [Green Version]
- Laliberte, A.S.; Browning, D.M.; Rango, A. A comparison of three feature selection methods for object-based classification of sub-decimeter resolution UltraCam-L imagery. Int. J. Appl. Earth Obs. Geoinf. 2012, 15, 70–78. [Google Scholar] [CrossRef]
- Lennon, M.; Mercier, G.; Mouchot, M.C.; Hubert-Moy, L. Independent component analysis as a tool for the dimensionality reduction and the representation of hyperspectral images. In Proceedings of the IGARSS 2001. Scanning the Present and Resolving the Future. IEEE 2001 International Geoscience and Remote Sensing Symposium (Cat. No.01CH37217), Sydney, Australia, 9–13 July 2001; pp. 2893–2895. [Google Scholar] [CrossRef] [Green Version]
- Ren, J.; Zabalza, J.; Marshall, S.; Zheng, J. Effective Feature Extraction and Data Reduction in Remote Sensing Using Hyperspectral Imaging [Applications Corner]. IEEE Signal Process. Mag. 2014, 31, 149–154. [Google Scholar] [CrossRef] [Green Version]
- Van Coillie, F.M.B.; Verbeke, L.P.C.; De Wulf, R.R. Feature selection by genetic algorithms in object-based classification of IKONOS imagery for forest mapping in Flanders, Belgium. Remote Sens. Environ. For. Spec. Issue 2007, 110, 476–487. [Google Scholar] [CrossRef]
- Guyon, I.; Elisseeff, A. An Introduction to Variable and Feature Selection. J. Mach. Learn. Res. 2003, 3, 1157–1182. [Google Scholar]
- Eklundh, L.; Singh, A. A comparative analysis of standardised and unstandardised Principal Components Analysis in remote sensing. Int. J. Remote Sens. 1993, 14, 1359–1370. [Google Scholar] [CrossRef]
- Du, Q. Modified Fisher’s Linear Discriminant Analysis for Hyperspectral Imagery. IEEE Geosci. Remote Sens. Lett. 2007, 4, 503–507. [Google Scholar] [CrossRef]
- Martínez, A.M.; Kak, A.C. PCA versus LDA. IEEE Trans. Pattern Anal. Mach. Intell. 2001, 23, 228–233. [Google Scholar] [CrossRef] [Green Version]
- Cao, L.J.; Chua, K.S.; Chong, W.K.; Lee, H.P.; Gu, Q.M. A comparison of PCA, KPCA and ICA for dimensionality reduction in support vector machine. Neurocomputing 2003, 55, 321–336. [Google Scholar] [CrossRef]
- Li, J.; Cheng, K.; Wang, S.; Morstatter, F.; Trevino, R.P.; Tang, J.; Liu, H. Feature Selection: A Data Perspective. ACM Comput. Surv. 2017, 50, 1–45. [Google Scholar] [CrossRef] [Green Version]
- Gorelick, N.; Hancher, M.; Dixon, M.; Ilyushchenko, S.; Thau, D.; Moore, R. Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sens. Environ. 2017, 202, 18–27. [Google Scholar] [CrossRef]
- Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Müller, A.; Nothman, J.; Louppe, G.; et al. Scikit-learn: Machine Learning in Python. arXiv 2012, arXiv:1201.0490. [Google Scholar]
- Buitinck, L.; Louppe, G.; Blondel, M.; Pedregosa, F.; Mueller, A.; Grisel, O.; Niculae, V.; Prettenhofer, P.; Gramfort, A.; Grobler, J.; et al. API design for machine learning software: experiences from the scikit-learn project. arXiv 2013, arXiv:1309.0238. [Google Scholar]
- Mountrakis, G.; Im, J.; Ogole, C. Support vector machines in remote sensing: A review. ISPRS J. Photogramm. Remote Sens. 2011, 66, 247–259. [Google Scholar] [CrossRef]
- Melgani, F.; Bruzzone, L. Classification of hyperspectral remote sensing images with support vector machines. IEEE Trans. Geosci. Remote Sens. 2004, 42, 1778–1790. [Google Scholar] [CrossRef] [Green Version]
- Tarabalka, Y.; Fauvel, M.; Chanussot, J.; Benediktsson, J.A. SVM- and MRF-Based Method for Accurate Classification of Hyperspectral Images. IEEE Geosci. Remote Sens. Lett. 2010, 7, 736–740. [Google Scholar] [CrossRef] [Green Version]
- Huang, X.; Zhang, L. An SVM Ensemble Approach Combining Spectral, Structural, and Semantic Features for the Classification of High-Resolution Remotely Sensed Imagery. IEEE Trans. Geosci. Remote Sens. 2013, 51, 257–272. [Google Scholar] [CrossRef]
- Nemmour, H.; Chibani, Y. Multiple support vector machines for land cover change detection: An application for mapping urban extensions. ISPRS J. Photogramm. Remote Sens. 2006, 61, 125–133. [Google Scholar] [CrossRef]
- Bovolo, F.; Bruzzone, L.; Marconcini, M. A Novel Approach to Unsupervised Change Detection Based on a Semisupervised SVM and a Similarity Measure. IEEE Trans. Geosci. Remote Sens. 2008, 46, 2070–2082. [Google Scholar] [CrossRef] [Green Version]
- Volpi, M.; Tuia, D.; Bovolo, F.; Kanevski, M.; Bruzzone, L. Supervised change detection in VHR images using contextual information and support vector machines. Int. J. Appl. Earth Obs. Geoinf. 2013, 20, 77–85. [Google Scholar] [CrossRef]
- Zheng, S.; Shi, W.; Liu, J.; Tian, J. Remote Sensing Image Fusion Using Multiscale Mapped LS-SVM. IEEE Trans. Geosci. Remote Sens. 2008, 46, 1313–1322. [Google Scholar] [CrossRef]
- Hay, G.J.; Castilla, G. Geographic Object-Based Image Analysis (GEOBIA): A new name for a new discipline. In Object-Based Image Analysis; Spatial Concepts for Knowledge-driven Remote Sensing; Blaschke, T., Lang, S., Hay, G.J., Eds.; Springer: Berlin/Heidelnberg, Germany, 2008; Chapter 1.4. [Google Scholar] [CrossRef]
- Blaschke, T.; Lang, S.; Hay, G.J. Object-Based Image Analysis: Spatial Concepts for Knowledge-Driven Remote Sensing Applications, Lecture Notes in Geoinformation and Cartography; Springer: Berlin/Heidelberg, Germany, 2008; pp. 75–89. [Google Scholar] [CrossRef]
- Ma, L.; Li, M.; Ma, X.; Cheng, L.; Du, P.; Liu, Y. A review of supervised object-based land cover classification. ISPRS J. Photogramm. Remote Sens. 2017, 130, 277–293. [Google Scholar] [CrossRef]
- Peña-Barragán, J.M.; Ngugi, M.K.; Plant, R.E.; Six, J. Object-based crop identification using multiple vegetation indices, textural features and crop phenology. Remote Sens. Environ. 2011, 115, 1301–1316. [Google Scholar] [CrossRef]
- Furberg, D.; Ban, Y.; Nascetti, A. Monitoring of Urbanization and Analysis of Environmental Impact in Stockholm with Sentinel-2A and SPOT-5 Multispectral Data. Remote Sens. 2019, 11, 2048. [Google Scholar] [CrossRef] [Green Version]
- Ban, Y.; Webber, L.; Gamba, P.; Paganini, M. EO4Urban: Sentinel-1A SAR and Sentinel-2A MSI data for global urban services. In Proceedings of the 2017 Joint Urban Remote Sensing Event (JURSE), Dubai, United Arab Emirates, 6–8 March 2017; pp. 1–4. [Google Scholar] [CrossRef]
- Benz, U.C.; Hofmann, P.; Willhauck, G.; Lingenfelder, I.; Heynen, M. Multi-resolution, object-oriented fuzzy analysis of remote sensing data for GIS-ready information. ISPRS J. Photogramm. Remote Sens. 2004, 58, 239–258. [Google Scholar] [CrossRef]
- Smith, A. Image segmentation scale parameter optimization and land cover classification using the Random Forest algorithm. J. Spat. Sci. 2010, 55, 69–79. [Google Scholar] [CrossRef]
- Johnson, B.; Xie, Z. Unsupervised image segmentation evaluation and refinement using a multi-scale approach. ISPRS J. Photogramm. Remote Sens. 2011, 66, 473–483. [Google Scholar] [CrossRef]
- Yang, J.; Li, P.; He, Y. A multi-band approach to unsupervised scale parameter selection for multi-scale image segmentation. ISPRS J. Photogramm. Remote Sens. 2014, 94, 13–24. [Google Scholar] [CrossRef]
- Lameski, P.; Zdravevski, E.; Mingov, R.; Kulakov, A. SVM Parameter Tuning with Grid Search and Its Impact on Reduction of Model Over-fitting. In Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing, Lecture Notes in Computer Science; Yao, Y., Hu, Q., Yu, H., Grzymala-Busse, J.W., Eds.; Springer International Publishing: Cham, Switzerland, 2015; pp. 464–474. [Google Scholar]
- Hastie, T.; Tibshirani, R.; Friedman, J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd ed.; Springer Series in Statistics; Springer: New York, NY, USA, 2009. [Google Scholar]
- Xanthopoulos, P.; Pardalos, P.M.; Trafalis, T.B. Robust Data Mining, SpringerBriefs in Optimization; Springer: New York, NY, USA, 2013. [Google Scholar]
- Gu, Q.; Li, Z.; Han, J. Linear Discriminant Dimensionality Reduction. In Machine Learning and Knowledge Discovery in Databases, Lecture Notes in Computer Science; Gunopulos, D., Hofmann, T., Malerba, D., Vazirgiannis, M., Eds.; Springer: Berlin/Heidelberg, Germany, 2011; pp. 549–564. [Google Scholar]
- Kraskov, A.; Stoegbauer, H.; Grassberger, P. Estimating Mutual Information. arXiv 2003, arXiv:cond-mat/0305641. [Google Scholar] [CrossRef] [Green Version]
- Weston, J.; Mukherjee, S.; Chapelle, O.; Pontil, M.; Poggio, T.; Vapnik, V. Feature Selection for SVMs. In Advances in Neural Information Processing Systems 13; Leen, T.K., Dietterich, T.G., Tresp, V., Eds.; MIT Press: Cambridge, MA, USA, 2001; pp. 668–674. [Google Scholar]
- Huang, R.; Liu, Q.; Lu, H.; Ma, S. Solving the small sample size problem of LDA. In Proceedings of the Object Recognition Supported by User Interaction for Service Robots, Quebec, QC, Canada, 11–15 August 2002; Volume 3, pp. 29–32. [Google Scholar] [CrossRef]
- Stromann, O. GitHub Repository. Available online: https://github.com/ostromann/GEE-LandCoverClass (accessed on 21 December 2019).
Class | Description | Stockholm | Beijing | ||
---|---|---|---|---|---|
No. Reference Points | No. Sample Segments | No. Reference Points | No. Sample Segments | ||
1-HDB | High-density built-up | 1000 | 986 | 150 | 134 |
2-LDB | Low-density built-up | 1000 | 926 | 150 | 54 |
3-R | Roads and Railroads | 1009 | 166 | 153 | 89 |
4-UGS | Urban green spaces | 1045 | 571 | 70 | 63 |
5-GC | Golf courses | 1013 | 275 | 80 | 73 |
6-AG | Agriculture | 1045 | 866 | 160 | 133 |
7-F | Forests | 1000 | 908 | 218 | 175 |
8-W | Water | 1000 | 780 | 161 | 93 |
9-BR | Bare rock | 503 | 172 | None | None |
10-WL | Wetlands | 500 | 105 | None | None |
Sum: | 9115 | 5755 | 1142 | 814 |
Stockholm | Beijing |
---|---|
S1 imagery (15 Ascending, 12 Descending) | S1 imagery (11 Ascending, 18 Descending) |
Period: 01-05-2015-–30-09-2015 | Period: 01-05-2016–30-09-2016 |
Polarizations: VV, VH | Polarizations: VV, VH |
S2 Imagery (8 Images) | S2 Imagery (11 Images) |
Period: 01-06-2015–30-08-2015- | Period: 01-06-2016–30-09-2016 |
Cloudy pixels: <15% | Cloudy pixels: <15% |
Band | Description | Band | Description |
---|---|---|---|
asm | angular second moment | contrast | contrast |
corr | correlation | var | sum of squares: variance |
idm | inverse difference moment | savg | sum average |
svar | sum variance | sent | sum entropy |
ent | entropy | dvar | difference variance |
dent | difference entropy | imcorr1/imcorr2 | information measures of correlation 1/2 |
diss | dissimilarity | inertia | inertia |
Sensor | Input layers | Segment Statistics (Mean, Min, Max, Std.Dev) | No. of Features | |
---|---|---|---|---|
S1 | temporal mean | 4 images (Asc/Desc VV/VH) | 16 | |
temporal standard deviation | 4 images (Asc/Desc VV/VH) | 16 | ||
texture features | 8 images 17 GLCM features | 544 | ||
S2 | cloud-free composite | 12 spectral bands | 48 | |
2 spectral indices | 8 | |||
texture features NIR | 10 m bands | 1 band 17 GLCM features | 68 | |
geometry | minimum enclosing circle | radius, areal difference | 2 | |
minimal area rotated rectangle | height, width, angle, aspect ratio, aereal difference | 5 | ||
least-squares-fitted ellipse | major-axis, minor-axis, angle, aspect ratio, areal difference | 5 | ||
Sum | 712 |
Method | Feature Range | -Parameter | -Parameter | ||||
---|---|---|---|---|---|---|---|
Start | Stop | Num | Start | Stop | Num | ||
Def.SVM | 712 | 0 | 8 | 9 | −10 | −2 | 9 |
LDA | 1–9 | −1 | 5 | 7 | −6 | −1 | 6 |
MI | 1–712 | 1 | 6 | 6 | −9 | −3 | 7 |
F-Score | 1–712 | 1 | 7 | 7 | −8 | −2 | 7 |
Mean Overall Accuracy [%] | Std. Dev. of Accuracy [%] | Accuracy Test Data [%] | Number of Features | Dimensionality Compression | Decrease Factor Training Time | Decrease Factor Pred. Time | |
---|---|---|---|---|---|---|---|
Method | Best-performing classifier | ||||||
Def.SVM | 88.0 | 0.78 | 87.9 | 712 | 1 | 1 | 1 |
LDA | 93.1 | 0.93 | 94.4 | 9 | 79 | 16 | 17 |
MI | 94.4 | 0.64 | 94.7 | 185 | 4 | 6 | 10 |
F-Score | 95.0 | 0.75 | 94.7 | 124 | 6 | 10 | 12 |
Best-performing within 0.5% non-inferiority margin | |||||||
LDA | - | - | - | - | - | - | - |
MI | 94.3 | 0.69 | 94.5 | 38 | 19 | 11 | 38 |
F-Score | 84.9 | 0.43 | 94.9 | 102 | 7 | 10 | 10 |
Best-performing within 1% non-inferiority margin | |||||||
LDA | 92.5 | 0.51 | 94.4 | 7 | 102 | 22 | 25 |
MI | 93.6 | 0.36 | 94.3 | 31 | 23 | 4 | 43 |
F-Score | 94.4 | 0.32 | 94.3 | 56 | 13 | 16 | 23 |
Best-performing within 3% non-inferiority margin | |||||||
LDA | 91.3 | 0.54 | 92.3 | 5 | 142 | 48 | 58 |
MI | 91.8 | 0.95 | 93.5 | 14 | 51 | 1 | 47 |
F-Score | 92.3 | 0.65 | 93.7 | 14 | 51 | 19 | 45 |
Best-performing within 5% non-inferiority margin | |||||||
LDA | 89.2 | 0.92 | 90.3 | 4 | 178 | 36 | 75 |
MI | 89.9 | 0.83 | 91.4 | 9 | 79 | 1 | 44 |
F-Score | 90.7 | 0.75 | 91.9 | 8 | 89 | 2 | 50 |
Method | Feature Range | -Parameter | -Parameter | ||||
---|---|---|---|---|---|---|---|
Start | Stop | Num | Start | Stop | Num | ||
Def.SVM | 712 | 3 | 11 | 9 | −12 | −4 | 9 |
LDA | 1–7 | −1 | 5 | 7 | −7 | −1 | 7 |
MI | 1–712 | 3 | 8 | 6 | −9 | −4 | 6 |
F-Score | 1–712 | 2 | 8 | 7 | −10 | −3 | 8 |
Mean Overall Accuracy [%] | Std. Dev. of Accuracy [%] | Accuracy Test Data [%] | Number of Features | Dimensionality Compression | Decrease Factor Training Time | Decrease Factor Pred. Time | |
---|---|---|---|---|---|---|---|
Method | Best-performing classifier | ||||||
Def.SVM | 76.0 | 6.97 | 81.9 | 712 | 1 | 1 | 1 |
LDA | 84.4 | 3.62 | 77.5 | 7 | 102 | 5 | 14 |
MI | 91.5 | 2.64 | 90 | 226 | 3 | 3 | 4 |
F-Score | 93.7 | 2.43 | 93.1 | 226 | 3 | 3 | 4 |
Best-performing within 3% non-inferiority margin | |||||||
LDA | - | - | - | - | - | - | - |
MI | 89.9 | 2.42 | 88.1 | 13 | 55 | 1 | 16 |
F-Score | 91.6 | 2.42 | 90 | 110 | 6 | 3 | 7 |
Best-performing within 5% non-inferiority margin | |||||||
LDA | 81.9 | 3.93 | 77.5 | 6 | 119 | 5 | 13 |
MI | 87.7 | 2.55 | 89.4 | 10 | 71 | 1 | 16 |
F-Score | 89.7 | 1.01 | 93.1 | 21 | 34 | 4 | 13 |
© 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
Stromann, O.; Nascetti, A.; Yousif, O.; Ban, Y. Dimensionality Reduction and Feature Selection for Object-Based Land Cover Classification based on Sentinel-1 and Sentinel-2 Time Series Using Google Earth Engine. Remote Sens. 2020, 12, 76. https://doi.org/10.3390/rs12010076
Stromann O, Nascetti A, Yousif O, Ban Y. Dimensionality Reduction and Feature Selection for Object-Based Land Cover Classification based on Sentinel-1 and Sentinel-2 Time Series Using Google Earth Engine. Remote Sensing. 2020; 12(1):76. https://doi.org/10.3390/rs12010076
Chicago/Turabian StyleStromann, Oliver, Andrea Nascetti, Osama Yousif, and Yifang Ban. 2020. "Dimensionality Reduction and Feature Selection for Object-Based Land Cover Classification based on Sentinel-1 and Sentinel-2 Time Series Using Google Earth Engine" Remote Sensing 12, no. 1: 76. https://doi.org/10.3390/rs12010076