Tropical Mangrove Species Classification Using Random Forest Algorithm and Very High-Resolution Satellite Imagery
Abstract
The spectral mixing was a challenge that principally found in the pixel-based classification method, particularly in the species level. The objective of this study was to evaluate the effectiveness of the random forest (RF) algorithm in order to improve the accuracy of the tropical mangrove species classification in Pak Phanang mangrove conservation, Nakhon Si Thammarat Province. The study utilized the very high-resolution, the Quickbird image, which was pre-calibrated using radiometric and geometric correction to incorporate with the field observation data. The process divided the input data into the training and the validation sets. The training process adjusted the input parameters for instances, the tree depth, the number of sample node, and the number of trees to acquire the best RF classification model. The validation compared the classified result with the conventional pixel-based maximum likelihood classification (MLC). The overall accuracy (OA), the kappa statistic, and the Z-statistic were indications of the RF classification evaluation. The result revealed that the RF algorithm achieved higher efficiency with the overall accuracy of 78.00% and 0.72 for the kappa statistic. Meanwhile, for MLC, the OA and the kappa statistic presented 56.00% and 0.44, respectively. The Z statistic (Z = 3.68) result also significantly confirmed the difference between RF and MLC at the 95% confidence level. Keywords : random forest, very high resolution satellite imagery, classification, tropical mangrove speciesReferences
Anthony, J. V., & Joanne, M. G. (2005). Understanding Interobserver Agreement: The Kappa Statistics. Family Medicine, 375(5), 3.
Breiman, L. (1996). Heuristics of instability and stabilization in model selection. Annals of Statistics, 24(6), 2350-2383.
Breiman, L., Friedman, J., Stone, C. J., & Olshen, R. A. (1984). Classification and Regression Trees: Taylor & Francis.
Chan, J. C. W., & Paelinckx, D. (2008). Evaluation of random forest and Adaboost tree-based ensemble classification and spectral band selection for ecotope mapping using airborne hyperspectral imagery. Remote Sensing Environment, 112, 2999-3011.
Congalton, R. G., & Green, K. (2009). Assessing the accuracy of remotely sensed data. [electronic resource] : principles and practices: Boca Raton : CRC Press/Taylor & Francis, c2009. 2nd ed.
Dubeau, P., King, D. J., Unbushe, D. G., & Rebelo, L. M. (2017). Mapping the Dabus Wetlands, Ethiopia, Using Random Forest Classification of Landsat, PALSAR and Topographic Data. Remote Sensing, 9(10). doi:10.3390/rs9101056
Hamdan, O., Khairunnisa, M. R., Ammar, A. A., Hasmadi, I. M., & Aziz, H. K. (2013). MANGROVE CARBON STOCK ASSESSMENT BY OPTICAL SATELLITE IMAGERY. Penilaian stok karbon hutan paya laut menggunakan imej satelit optik., 25(4), 554-565.
Heumann, B. W. (2011). Satellite remote sensing of mangrove forests: Recent advances and future opportunities. Progress in Physical Geography, 35(1), 87-108. doi:10.1177/0309133310385371
Jhonnerie, R., Siregar, V. P., Nababan, B., Prasetyo, L. B., & Wouthuyzen, S. (2015). Random Forest Classification for Mangrove Land Cover Mapping Using Landsat 5 TM and Alos Palsar Imageries. Procedia Environmental Sciences, 24, 215-221. doi:https://doi.org/10.1016/j.proenv.2015.03.028
Koedsin, W., & Vaiphasa, C. (2013). Discrimination of Tropical Mangroves at the Species Level with EO-1 Hyperion Data. Remote Sensing, 5(7), 3562-3582. doi:10.3390/rs5073562
Kuenzer, C., Bluemel, A., Gebhardt, S., Quoc, T. V., & Dech, S. (2011). Remote Sensing of Mangrove Ecosystems: A Review. Remote Sensing, 3(5), 878-928. doi:10.3390/rs3050878
Lee, S. Y., Primavera, J. H., Dahdouh-Guebas, F., McKee, K., Bosire, J. O., Cannicci, S., Record, S. (2014). Ecological role and services of tropical mangrove ecosystems: a reassessment. Global Ecology and Biogeography, 23(7), 726-743. doi:10.1111/geb.12155
Luo, Y. M., Huang, D. T., Liu, P. Z., & Feng, H. M. (2016). An novel random forests and its application to the classification of mangroves remote sensing image. Multimedia Tools and Applications, 75(16),
9707-9722. doi:10.1007/s11042-015-2906-9
Ramo, R., & Chuvieco, E. (2017). Developing a Random Forest Algorithm for MODIS Global Burned Area Classification. Remote Sensing, 9(11). doi:10.3390/rs9111193
Rodriguez-Galiano, V. F., Ghimire, B., Rogan, J., Chica-Olmo, M., & Rigol-Sanchez, J. P. (2012).
An assessment of the effectiveness of a random forest classifier for land-cover classification. Isprs Journal of Photogrammetry and Remote Sensing, 67, 93-104. doi:10.1016/j.isprsjprs.2011.11.002
Sasmito, S. D., Murdiyarso, D., Friess, D. A., & Kurnianto, S. (2016). Can mangroves keep pace with contemporary sea level rise? A global data review. Wetlands Ecology and Management, 24(2),
263-278. doi:10.1007/s11273-015-9466-7
Wang, D. Z., Wan, B., Qiu, P. H., Su, Y. J., Guo, Q. H., & Wu, X. C. (2018). Artificial Mangrove Species Mapping Using Pleiades-1: An Evaluation of Pixel-Based and Object-Based Classifications with Selected Machine Learning Algorithms. Remote Sensing, 10(2). doi:10.3390/rs10020294
Watanakij, N., & Vaiphasa, C. (2016). Improving the accuracy of mangrove species discrimination using object based and high spatial resolution imagery: A case study in Pak Phanang, Thailand. International Journal of Geoinformatics, 12(3), 41-49.
Zhu, Y. H., Liu, K., Liu, L., Myint, S. W., Wang, S. G., Liu, H. X., & He, Z. (2017). Exploring the Potential of WorldView-2 Red-Edge Band-Based Vegetation Indices for Estimation of Mangrove Leaf Area Index with Machine Learning Algorithms. Remote Sensing, 9(10). doi:10.3390/rs9101060
Breiman, L. (1996). Heuristics of instability and stabilization in model selection. Annals of Statistics, 24(6), 2350-2383.
Breiman, L., Friedman, J., Stone, C. J., & Olshen, R. A. (1984). Classification and Regression Trees: Taylor & Francis.
Chan, J. C. W., & Paelinckx, D. (2008). Evaluation of random forest and Adaboost tree-based ensemble classification and spectral band selection for ecotope mapping using airborne hyperspectral imagery. Remote Sensing Environment, 112, 2999-3011.
Congalton, R. G., & Green, K. (2009). Assessing the accuracy of remotely sensed data. [electronic resource] : principles and practices: Boca Raton : CRC Press/Taylor & Francis, c2009. 2nd ed.
Dubeau, P., King, D. J., Unbushe, D. G., & Rebelo, L. M. (2017). Mapping the Dabus Wetlands, Ethiopia, Using Random Forest Classification of Landsat, PALSAR and Topographic Data. Remote Sensing, 9(10). doi:10.3390/rs9101056
Hamdan, O., Khairunnisa, M. R., Ammar, A. A., Hasmadi, I. M., & Aziz, H. K. (2013). MANGROVE CARBON STOCK ASSESSMENT BY OPTICAL SATELLITE IMAGERY. Penilaian stok karbon hutan paya laut menggunakan imej satelit optik., 25(4), 554-565.
Heumann, B. W. (2011). Satellite remote sensing of mangrove forests: Recent advances and future opportunities. Progress in Physical Geography, 35(1), 87-108. doi:10.1177/0309133310385371
Jhonnerie, R., Siregar, V. P., Nababan, B., Prasetyo, L. B., & Wouthuyzen, S. (2015). Random Forest Classification for Mangrove Land Cover Mapping Using Landsat 5 TM and Alos Palsar Imageries. Procedia Environmental Sciences, 24, 215-221. doi:https://doi.org/10.1016/j.proenv.2015.03.028
Koedsin, W., & Vaiphasa, C. (2013). Discrimination of Tropical Mangroves at the Species Level with EO-1 Hyperion Data. Remote Sensing, 5(7), 3562-3582. doi:10.3390/rs5073562
Kuenzer, C., Bluemel, A., Gebhardt, S., Quoc, T. V., & Dech, S. (2011). Remote Sensing of Mangrove Ecosystems: A Review. Remote Sensing, 3(5), 878-928. doi:10.3390/rs3050878
Lee, S. Y., Primavera, J. H., Dahdouh-Guebas, F., McKee, K., Bosire, J. O., Cannicci, S., Record, S. (2014). Ecological role and services of tropical mangrove ecosystems: a reassessment. Global Ecology and Biogeography, 23(7), 726-743. doi:10.1111/geb.12155
Luo, Y. M., Huang, D. T., Liu, P. Z., & Feng, H. M. (2016). An novel random forests and its application to the classification of mangroves remote sensing image. Multimedia Tools and Applications, 75(16),
9707-9722. doi:10.1007/s11042-015-2906-9
Ramo, R., & Chuvieco, E. (2017). Developing a Random Forest Algorithm for MODIS Global Burned Area Classification. Remote Sensing, 9(11). doi:10.3390/rs9111193
Rodriguez-Galiano, V. F., Ghimire, B., Rogan, J., Chica-Olmo, M., & Rigol-Sanchez, J. P. (2012).
An assessment of the effectiveness of a random forest classifier for land-cover classification. Isprs Journal of Photogrammetry and Remote Sensing, 67, 93-104. doi:10.1016/j.isprsjprs.2011.11.002
Sasmito, S. D., Murdiyarso, D., Friess, D. A., & Kurnianto, S. (2016). Can mangroves keep pace with contemporary sea level rise? A global data review. Wetlands Ecology and Management, 24(2),
263-278. doi:10.1007/s11273-015-9466-7
Wang, D. Z., Wan, B., Qiu, P. H., Su, Y. J., Guo, Q. H., & Wu, X. C. (2018). Artificial Mangrove Species Mapping Using Pleiades-1: An Evaluation of Pixel-Based and Object-Based Classifications with Selected Machine Learning Algorithms. Remote Sensing, 10(2). doi:10.3390/rs10020294
Watanakij, N., & Vaiphasa, C. (2016). Improving the accuracy of mangrove species discrimination using object based and high spatial resolution imagery: A case study in Pak Phanang, Thailand. International Journal of Geoinformatics, 12(3), 41-49.
Zhu, Y. H., Liu, K., Liu, L., Myint, S. W., Wang, S. G., Liu, H. X., & He, Z. (2017). Exploring the Potential of WorldView-2 Red-Edge Band-Based Vegetation Indices for Estimation of Mangrove Leaf Area Index with Machine Learning Algorithms. Remote Sensing, 9(10). doi:10.3390/rs9101060
Downloads
Published
2019-06-18
Issue
Section
Research Article
