Estimation of Fractional Snow Cover from MODIS Data in Ilgaz Forest District Region by Support Vector Machines

Yıl 2019, Cilt: 21 Sayı: 3, 911 - 926, 15.12.2019

Bora Berkay Çiftçi Semih Kuter

https://doi.org/10.24011/barofd.595462

Öz

This study is focused on the assessment of
support vector machines (SVM) in order to estimate the fractional snow cover
(FSC) from coarse spatial resolution moderate resolution imaging
spectroradiometer (MODIS) imagery in Ilgaz Forest District area located within
the cities of Çankırı and Kastamonu.  SVM
model training is carried out by employing 10 predictor variables obtained from
MODIS images taken between March 2000 and April 2016, namely, MODIS
top-of-atmospheric reflectance values of bands 1-7, normalized difference snow
index, normalized difference vegetation index and land cover class. Higher
resolution Landsat 7/8 images are used to generate the corresponding reference
FSC maps. Accuracy of SVM models are assessed with respect to the size of the
training data and the sampling type. The impact of the kernel type on the
accuracy of the SVM models is also investigated. According to the results, all
SVM models trained with linear, 2^nd, 3^rd and 4^th
order polynomials as well as radial basis function (RBF) kernels give high
correlation rates with the associated reference FSC maps (R ≥ 0,91). On the other hand, MOD10A1, the standard FSC product of
MODIS, exhibits slightly poorer performance with average R = 0,77. In terms of computational efficiency with respect to CPU
times spent during the training stage, RBF kernel is found to be superior with
average model building times of 279, 2300 and 8457 seconds for small-, medium-
and large-sized training data sets, respectively.

Anahtar Kelimeler

Remote sensing of snow, MODIS, Landsat, support vector regression

Proje Numarası

OF090316L04

Destek Vektör Makineleri ile MODIS Verisinden Fraksiyonel Kar Örtüsünün Ilgaz Orman İşletme Müdürlüğü Bölgesinde Belirlenmesi

Öz

Bu çalışmada, Çankırı ve
Kastamonu il sınırları içinde yer alan Ilgaz Orman İşletme Müdürlüğü
bölgesinde, orta çözünürlüklü görüntüleme spektroradyometresi (MODIS) verisinden
etkili kar kaplı alan (EKKA) haritalaması amacıyla destek vektör makineleri
(DVM) tasarımı araştırılmıştır. DVM modellerin eğitilmesinde, Mart 2000 ve
Nisan 2016 tarihleri arasında alınan MODIS görüntülerinden elde edilen toplam
10 bağımsız değişken; MODIS bant 1-7 atmosfer üstü reflektans değerleri,
normalize fark kar indisi, normalize fark vejetasyon indisi ve arazi sınıfı kullanılmıştır.
Referans EKKA haritaları daha yüksek mekânsal çözünürlüğe sahip ilgili Landsat
7/8 görüntülerinden üretilmiştir. DVM modellerinin doğruluğu, eğitim
verilerinin boyutuna ve örneklem türüne göre değerlendirilmiştir. Kernel
türünün DVM modellerinin doğruluğu üzerindeki etkisi de incelenmiştir.
Sonuçlara göre, doğrusal, 2., 3. ve 4. dereceden polinomların yanı sıra radyal
temel fonksiyonu (RBF) kernelleri ile eğitilmiş tüm DVM modelleri, ilgili
referans EKKA haritaları ile yüksek korelasyon oranları vermektedir (R ≥ 0,91). Öte yandan, MODIS'in standart
EKKA ürünü olan MOD10A1, ortalama R =
0,77 ile biraz daha zayıf performans sergilemektedir. Eğitim aşamasında
harcanan CPU zamanlarına göre hesaplama etkinliği bakımından, RBF kernelinin, küçük,
orta ve büyük boyutlu eğitim veri setleri için sırasıyla 279, 2300 ve 8457
saniyelik ortalama model oluşturma süreleriyle daha üstün olduğu görülmüştür.

Anahtar Kelimeler

Karın uzaktan algılanması, MODIS, Landsat, destek vektör regresyonu

Destekleyen Kurum

ÇANKIRI KARATEKİN ÜNİVERSİTESİ BAP BİRİMİ

Proje Numarası

OF090316L04

Teşekkür

Bu çalışma, Bora Berkay ÇİFTÇİ’nin Çankırı Karatekin Üniversitesi BAP biriminin OF090316L04 nolu projesince desteklenen yüksek lisans tezinden üretilmiştir.

Kaynakça

• Akyürek, Z., Hall D. K., Riggs, G.A., Sensoy, A. (2010). Evaluating the utility of the ANSA blended snow cover product in the mountains of eastern Turkey. International Journal of Remote Sensing, 31(14), 3727-3744.
• Aydinozu, D., Ibret, U., Aydin, M. (2011). Analysis of Terrain Usage in Kastamonu-Ilgaz Mountain Natural Park. International Symposium on Environmental Protection and Planning: Geographic Information Systems (GIS) and Remote Sensing (RS) Applications (ISEPP), 28-29 June 2011, İzmir - TURKEY.
• Bruzzone, L., Melgani, F. (2005). Robust multiple estimator systems for the analysis of biophysical parameters from remotely sensed data. IEEE Transactions on Geoscience and Remote Sensing, 43(1), 159-174.
• Chen, J., Zhu, X., Vogelmann, J. E., Gao, F., Jin, S. (2011). A simple and effective method for filling gaps in Landsat ETM+ SLC-off images. Remote Sensing of Environment, 115(4): 1053-1064.
• Chi, M., Feng, R., Bruzzone, L. (2008). Classification of hyperspectral remote-sensing data with primal SVM for small-sized training dataset problem. Advances in Space Research, 41(11), 1793-1799.
• Davis CS (2002). Statistical methods for the analysis of repeated measurements: Springer Science & Business Media.
• Dietz, A. J., Kuenzer, C., Gessner, U., Dech, S. (2012). Remote sensing of snow – a review of available methods. International Journal of Remote Sensing, 33(13), 4094-4134.
• Dobreva, I. D., Klein, A. G. (2011). Fractional snow cover mapping through artificial neural network analysis of MODIS surface reflectance. Remote Sensing of Environment, 115(12), 3355-3366.
• Dodge, Y. (2008). The Concise Encyclopedia of Statistics. New York: Springer.
• Dwyer, J., Schmidt, G. (2006). The MODIS Reprojection Tool. In J. J. Qu, W. Gao, M. Kafatos, R. E. Murphy ve V. V. Salomonson (Eds.), Earth Science Satellite Remote Sensing: Vol. 2: Data, Computational Processing, and Tools (162-177). Berlin, Heidelberg: Springer Berlin Heidelberg.
• Fisher, W. D., Camp, T. K., Krzhizhanovskaya, V. V. (2017). Anomaly detection in earth dam and levee passive seismic data using support vector machines and automatic feature selection. Journal of Computational Science, 20, 143-153.
• Foody, G. M., Mathur, A. (2004). Toward intelligent training of supervised image classifications: directing training data acquisition for SVM classification. Remote Sensing of Environment, 93(1-2), 107-117.
• Foody, G. M., Mathur, A. (2006). The use of small training sets containing mixed pixels for accurate hard image classification: Training on mixed spectral responses for classification by a SVM. Remote Sensing of Environment, 103(2), 179-189.
• Frei, A., Tedesco, M., Lee, S., Foster, J., Hall, D. K., Kelly, R., Robinson, D. A. (2012). A review of global satellite-derived snow products. Advances in Space Research, 50(8), 1007-1029.
• Friedl, M. A., Sulla-Menashe, D., Tan, B., Schneider, A., Ramankutty, N., Sibley, A., Huang, X. (2010). MODIS Collection 5 global land cover: Algorithm refinements and characterization of new datasets. Remote Sensing of Environment, 114(1), 168-182.
• Hall, D., Foster, J., Verbyla, D., Klein, A., Benson, C. (1998). Assessment of snow-cover mapping accuracy in a variety of vegetation-cover densities in central Alaska. Remote Sensing of Environment, 66(2), 129-137.
• Hall, D. K., Riggs, G. A., Salomonson, V. V. (1995). Development of Methods for Mapping Global Snow Cover Using Moderate Resolution Imaging Spectroradiometer Data. Remote Sensing of Environment, 54, 127-140.
• Hall, D. K., Riggs, G. A., Salomonson, V. V. (2006). MODIS Snow and Sea Ice Products. In J. J. Qu, W. Gao, M. Kafatos, R. E. Murphy ve V. V. Salomonson (Eds.), Earth Science Satellite Remote Sensing Vol. 1: Science and Instruments (154-181). Berlin, Heidelberg: Springer.
• Hall, D. K., Riggs, G. A., Salomonson, V. V., DiGirolamo, N. E., Bayr, K. J. (2002). MODIS snow-cover products. Remote Sensing of Environment, 83, 181-194. Haykin, S. (2009). Neural Networks and Learning Machines (3rd ed.). Upper Saddle River, NJ, USA: Pearson.
• Kaheil, Y. H., Rosero, E., Gill, M. K., McKee, M., Bastidas, L. A. (2008). Downscaling and forecasting of evapotranspiration using a synthetic model of wavelets and support vector machines. IEEE Transactions on Geoscience and Remote Sensing, 46(9), 2692-2707.
• Kavzoglu, T., Colkesen, I. (2009). A kernel functions analysis for support vector machines for land cover classification. International Journal of Applied Earth Observation and Geoinformation, 11(5), 352-359.
• Klein, A. G., Barnett, A. C. (2003). Validation of daily MODIS snow cover maps of the Upper Rio Grande River Basin for the 2000–2001 snow year. Remote Sensing of Environment, 86(2), 162-176.
• Kumar, D., Meghwani, S. S., Thakur, M. (2016). Proximal support vector machine based hybrid prediction models for trend forecasting in financial markets. Journal of Computational Science, 17, 1-13.
• Kumar, K. K., Shelokar, P. S. (2008). An SVM method using evolutionary information for the identification of allergenic proteins. Bioinformation, 2(6), 253.
• Kuter, N. (2008). Evaluation of Ilgaz Mountain National Park in Terms of Forest Landscape and Aesthetics. Turkish Journal of Forestry, 1, 36-47.
• Kuter, S., Akyurek, Z., Weber, G.W. (2018). Retrieval of fractional snow covered area from MODIS data by multivariate adaptive regression splines. Remote Sensing of Environment, 205, 236-252.
• Lehning, M., Völksch, I., Gustafsson, D., Nguyen, T. A., Stähli, M., Zappa, M. (2006). ALPINE3D: a detailed model of mountain surface processes and its application to snow hydrology. Hydrological Processes, 20(10), 2111-2128.
• Liu, Y., Chen, Y. (2007). Face Recognition Using Total Margin-Based Adaptive Fuzzy Support Vector Machines. IEEE Transactions on Neural Networks, 18(1), 178-192.
• Luojus, K. P., Pulliainen, J. T., Metsamaki, S. J., Hallikainen, M. T. (2007). Snow-covered area estimation using satellite radar wide-swath images. IEEE Transactions on Geoscience and Remote Sensing, 45(4), 978-989.
• Mattera, D., Haykin, S. (1999). Support vector machines for dynamic reconstruction of a chaotic system. In S. Bernhard, J. C. B. Christopher ve J. S. Alexander (Eds.), Advances in Kernel Methods - Support Vector Learning (211-241). Cambridge, MA: MIT Press.
• Melgani, F., Bruzzone, L. (2004). Classification of hyperspectral remote sensing images with support vector machines. Geoscience and Remote Sensing, IEEE Transactions on, 42(8), 1778-1790.
• Mountrakis, G., Im, J., Ogole, C. (2011). Support vector machines in remote sensing: A review. ISPRS Journal of Photogrammetry and Remote Sensing, 66(3), 247-259.
• Müller, K.-R., Smola, A. J., Rätsch, G., Schölkopf, B., Kohlmorgen, J., Vapnik, V. (1997, October 8–10, 1997 ). Predicting time series with support vector machines. International Conference on Artificial Neural Networks - ICANN'97, Lausanne, Switzerland.
• Painter, T. H., Rittger, K., McKenzie, C., Slaughter, P., Davis, R. E., Dozier, J. (2009). Retrieval of subpixel snow covered area, grain size, and albedo from MODIS. Remote Sensing of Environment, 113(4), 868-879.
• Pal, M., Mather, P. M. (2005). Support vector machines for classification in remote sensing. International Journal of Remote Sensing, 26(5), 1007-1011.
• Piazzi, G., Tanis, C. M., Kuter, S., Simsek, B., Puca, S., Toniazzo, A., Takala, M., Akyürek, Z., Gabellani, S., Arslan, A. N. (2019). Cross-Country Assessment of H-SAF Snow Products by Sentinel-2 Imagery Validated against In-Situ Observations and Webcam Photography. Geosciences, 9(3), 129.
• Richards, J. A., Jia, X. (2006). Remote sensing digital image analysis: An introduction (4th ed.). Germany: Springer.
• Romanov, P., Tarpley, D., Gutman, G., Carroll, T. (2003). Mapping and monitoring of the snow cover fraction over North America. Journal of Geophysical Research, Atmospheres, 108(D16).
• Scaramuzza, P., Micijevic, E., Chander, G. (2004). SLC-off Gap-Filled Products Gap-Fill Algorithm Methodology Phase 2 Gap-Fill Algorithm. US Geological Survey Earth Resources Observation and Science (EROS) Center.
• Shanthi, N., Duraiswamy, K. (2010). A novel SVM-based handwritten Tamil character recognition system. Pattern Analysis and Applications, 13(2), 173-180.
• Siljamo, N., Hyvärinen, O. (2011). New Geostationary Satellite–Based Snow-Cover Algorithm. Journal of Applied Meteorology and Climatology, 50(6), 1275-1290.
• Smola, A. J., Schölkopf, B. (2004). A tutorial on support vector regression. Statistics and Computing, 14(3), 199-222.
• Vapnik, V. N. (1995). The Nature of Statistical Learning Theory: Springer Heidelberg.
• Vermote, E. F., Kotchenova, S. Y., Ray, J. P. (2011). MODIS surface reflectance user’s guide - Version 1.3. MODIS Land Surface Reflectance Science Computing Facility
• Wolfe, R. E. (2006). MODIS Geolocation. In J. J. Qu, W. Gao, M. Kafatos, R. E. Murphy ve V. V. Salomonson (Eds.), Earth Science Satellite Remote Sensing Vol. 1: Science and Instruments (50-73). Berlin, Heidelberg: Springer.
• Xiong, X., Isaacman, A., Barnes, W. (2006). MODIS Level-1B Products. In J. J. Qu, W. Gao, M. Kafatos, R. E. Murphy ve V. V. Salomonson (Eds.), Earth Science Satellite Remote Sensing (33-49): Springer Berlin Heidelberg.
• Yuan, F.-C., Lee, C.-H. (2015). Using least square support vector regression with genetic algorithm to forecast beta systematic risk. Journal of Computational Science, 11, 26-33.
• Zhang, T. (2005). Influence of the seasonal snow cover on the ground thermal regime: An overview. Reviews of Geophysics, 43(4) 1-23.
• Zheng, S., Shi, W.-z., Liu, J., Tian, J. (2008). Remote sensing image fusion using multiscale mapped LS-SVM. IEEE Transactions on Geoscience and Remote Sensing, 46(5), 1313-1322.