Implementation of Support Vector Machines Algorithm to Classify Burn Area of Peat
Abstract
Forest fires have become annual disasters in Indonesia and have an impact on peatland degradation. Many forest fires occur on peatlands. In August 2019, 810 hotspots were detected in Jambi Province. Burned area of peatland information is needed so that the goverment can determine policy on the effectiveness and efficiency of forest management. Information about burned area of peatland is difficult to obtain from field measurements because the area is large and not easily accessible. Landsat data is a type of image from remote sensing technology that can be used to map this area. One method that is often used to estimate the burned area is visual on-screen interpretation. However, this technique requires experienced interpreters. For this reason, this study used digital interpretation techniques using the Support Vector Machine (SVM) algorithm to classify burned area, vegetation, and bare soil from remote sensed data of protected area of peat in Muaro Jambi District, Jambi Province. This study obtained a classifier with the accuracy of 99.8%. The estimated area of peat based on the SVM classifier on August 15th 2019 are 1,396.89 hectares at the burned area class, 7,069.5 hectares at the vegetation class, and 1,089.54 hectares at the bare soil class.
References
(2019) “Angka Deforestasi Sebagai ‘Alarm’ Memburuknya Hutan, [Online], https://fwi.or.id/wp-content/uploads/2019/10/FS_Deforestasi _FWI_small.pdf, tanggal akses: 15-Sep-2020.
L. Syaufina, “Forest and Land Fires in Indonesia: Assessment and Mitigation,” dalam Integrating Disaster Science and Management, P. Samui, D. Kim, dan C. Ghosh, Eds., Amsterdam, Netherlands: Elsevier, 2018, hal. 109–121.
W.C. Adinugroho, I.N.N. Suryadiputra, B.H. Saharjo, dan L. Siboro, Panduan Pengendalian Kebakaran Hutan dan Lahan Gambut. Bogor, Indonesia: Wetlands International - Indonesia Programme dan Wildlife Habitat Canada, 2005.
Suwarsono, Rokhmatuloh., dan T. Waryono, “Pengembangan model identifikasi daerah bekas kebakaran hutan dan lahan (burned area) menggunakan citra MODIS di Kalimantan,” J. Penginderaan Jauh dan Pengolahan Data Citra Digital, Vol. 10, No. 2, hal. 93-112, 2013.
D. Kosasih, M.B. Saleh, dan L.B. Prasetyo, “Visual and Digital Interpretations for Land Cover Classification in Kuningan District, West Java,” J. Ilmu Pertan. Indones., Vol. 24, No. 2, hal. 101–108, 2019.
U. Khaira, I.S. Sitanggang, dan L. Syaufina, “Detection and Prediction of Peatland Cover Changes Using Support Vector Machine and Markov Chain Model,” TELKOMNIKA (Telecommunication Comput. Electron. Control), Vol. 14, No. 1, hal. 294-301, 2016.
P. Thariqa, I.S. Sitanggang, dan L. Syaufina, “Comparative Analysis of Spatial Decision Tree Algorithms for Burned Area of Peatland in Rokan Hilir Riau,” TELKOMNIKA (Telecommunication Comput. Electron. Control), Vol. 14, No.2, hal. 684-691, 2016.
C. Pennington, “Burn Scar Mapping in the Sabine National Wildlife Refuge Using Landsat TM and ETM+ Imagery,” Thesis, Louisiana State University, Baton Rouge, USA, 2006.
I. Prasasti, R. Boer, M. Ardiansyah, A. Buono, dan L. Syaufina, “Analisis Hubungan Kode-Kode SPBK (Sistem Peringkat Bahaya Kebakaran) dan Hotspot dengan Kebakaran Hutan dan Lahan di Kalimantan Tengah,” J. Nat. Resour. Environ. Manag., Vol. 2, No. 2, hal. 91-101, 2012.
C. Huang, L.S. Davis, dan J.R.G. Townshend, “An Assessment of Support Vector Machines for Land Cover Classification,” Int. J. Remote Sens., Vol. 23, No. 4, hal. 725-749, 2002.
A.T. Azar dan S.A. El-Said, “Performance Analysis of Support Vector Machines Classifiers in Breast Cancer Mammography Recognition,” Neural Comput. Appl., Vol. 24, No. 5, hal. 1163-1177, 2014.
C. Pitoy, “Metode Support Vector Machines pada Klasterisasi K-Means Data Nonlinear Separable,” Frontiers: Jurnal Sains Dan Teknologi., Vol. 2, No. 1, hal. 71–77, 2010.
J. Weston, S. Mukherjee, O. Chapelle, M. Pontil, T. Poggio, dan V. Vapnik, “Feature Selection for SVMs,” Proc. Advances in Neural Information Processing Systems, 2001, hal. 668-674.
L.I. Kuncheva, Combining Pattern Classifiers. Hoboken, USA: John Wiley & Sons, Inc., 2014.
M.Z. Lubis, H.M. Taki, W. Anurogo, D.S. Pamungkas, P. Wicaksono, dan T. Aprilliyanti, “Mapping the Distribution of Potential Land Drought in Batam Island Using the Integration of Remote Sensing and Geographic Information Systems (GIS),” IOP Conference Series: Earth and Environmental Science, Vol. 98, The 5th Geoinformation Science Symposium, 2017, hal. 1-8.
L.S. Wijedasa, S. Sloan, D.G. Michelakis, dan G.R. Clements, “Overcoming Limitations with Landsat Imagery for Mapping of Peat Swamp Forests in Sundaland,” Remote Sens., Vol. 4, No. 9, hal. 2595-2618, 2012.
I.H. Witten, E. Frank, M.A. Hall, dan C.J. Pal, Data Mining: Practical Machine Learning Tools and Techniques, Burlington, USA: Morgan Kaufmann, 2011.
P. Lameski, E. Zdravevski, R. Mingov, dan A. Kulakov, “SVM Parameter Tuning with Grid Search and Its Impact on Reduction of Model Over-fitting,” dalam Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing. Lecture Notes in Computer Science, Vol. 9437, Y. Yao, Q. Hu, H. Yu, J.W. Grzymala-Busse, Eds., Cham, Switzerland: Springer, 2015, hal. 464–474.
F. Yuan, K.E. Sawaya, B.C. Loeffelholz, dan M.E. Bauer, “Land Cover Classification and Change Analysis of the Twin Cities (Minnesota) Metropolitan Area by Multitemporal Landsat Remote Sensing,” Remote Sens. Environ., Vol. 98, No. 2, hal. 317–328, Okt. 2005.
© Jurnal Nasional Teknik Elektro dan Teknologi Informasi, under the terms of the Creative Commons Attribution-ShareAlike 4.0 International License.
1.png)
