Çoklu Zamanlı Sentinel-2 Görüntülerinden Tarımsal Ürün Tespiti: Mardin – Kızıltepe Örneği

Bu çalışmada, Mardin İli, Artuklu, Kızıltepe ve Derik İlçelerinde tarımsal arazilerden oluşan bir bölgede, 2018 yılına ait çoklu tarihli Sentinel-2 uydu görüntülerinden sınıflandırma yöntemi ile ürün tespiti yapılmıştır. Sınıflandırmada, Rastgele Orman (RO) algoritması parsel-tabanlı yaklaşımla kullanılmıştır. Tespit edilen ürünler mısır, buğday, pamuk, nohut, mercimek ve diğerleridir. Görüntü olarak altı farklı tarihte (8 Nisan, 23 Mayıs, 12 Temmuz, 11 Ağustos, 5 Eylül ve 5 Ekim) çekilmiş görüntüler seçilmiştir. Sınıflandırmada 10m konumsal çözünürlüklü Mavi (M), Yeşil (Y), Kırmızı (K) ve Yakın Kızıl Ötesi (YKÖ) bantlar kullanılmıştır. Ayrıca, her bir görüntü tarihi için Normalize Edilmiş Fark Bitki Örtüsü İndeksi (NFBİ) bandı hesaplanmış ve sınıflandırmada ek bant olarak kullanılmıştır. RO algoritması ile sınıflandırma işlemi, her bir görüntü tarihine ait beş bant (M, Y, K, YKÖ ve NFBİ) olmak üzere, toplam 30 bantlı görüntü yığınının tek seferde sınıflandırmaya dâhil edilmesi şeklinde gerçekleştirilmiştir. Eğitim alanı örnekleri ve sonuçların doğruluk analizleri için, mevcut Çiftçi Kayıt Sistemi (ÇKS) verilerinden yararlanılmıştır. Sınıflandırma neticesinde % 96.35 genel doğruluk ve % 93.13 kappa katsayısı değerlerine ulaşılmıştır.

Anahtar Kelimeler:

Sentinel 2, Makine Öğrenme Algoritması, Rastgele Orman (RO) Sınıflandırma Tekniği, Çiftçi Kayıt Sistemi (ÇKS)

Agricultural Crop Detection from Multi-Temporal Sentinel 2 Images: A Case Study of Mardin - Kızıltepe

In this study, crop detection was carried out in an agricultural region in the Artuklu, Kızıltepe and Derik districts of the city of Mardin through classification of multi-temporal Sentinel-2 images from 2018. In the classification, the Random Forest (RF) algorithm was used through a parcel-based approach. The detected crops are corn, wheat, cotton, chickpeas, lentils and the others. The images acquired on six different dates (April 8, May 23, July 12, August 11, September 5 and October 5) were selected as the images. In the classification, the 10m spatial resolution bands Blue (B), Green (G), Red (R) and Near Infrared (NIR) were used. Furthermore, for each image date a Normalized Difference Vegetation Index (NDVI) band was computed and used as additional band in classification. The classification process through the RF algorithm was carried out using the stack of 30 bands that includes five bands (B, G, R, NIR, NDVI) for each image. For training samples and the accuracy assessment of the results, the existing Farmers’ Registry System (FRS) was utilised. As a result of the classification process, the overall accuracy of 96.35% and the Kappa value of 93.13% were achieved.

Keywords:

Sentinel 2, Machine Learning Algorithm, Random Forest (RF) Classification Technique, Farm Registration System (FRS),

PDF

___

Akar, Ö., 2013. Rastgele Orman Sınıflandırıcısına Doku Özellikleri Entegre Edilerek Benzer Spektral Özellikteki Tarımsal Ürünlerin Sınıflandırılması. Doktora Tezi, Karadeniz Teknik Üniversitesi, Fen Bilimleri Enstitüsü, Trabzon, 180.
Archer, K.J. and Kimes, R.V., 2008. Emprical characterization of random forest variable importance measure. Computational Statistics & Data Analysis, 52 (4), 2249-2260.
Arvor, D., Jonathan, M., Meirelles, M. P., Dubreuil V. and Durieux, L., 2011. Classification of MODIS EVI time series for crop mapping in the state of Mato Grosso, Brazil. International Journal of Remote Sensing, 32, 7847–7871.
Ban, Y., 2003. Synergy of multitemporal ERS-1 SAR and Landsat TM data for classification of agricultural crops. Canadian Journal of Remote Sensing, 29, 518-526.
Belgiu, M. and Csillik, O., 2018. Sentinel-2 cropland mapping using pixel-based and object-based time-weighted dynamic time warping analysis. Remote Sensing of Environment, 204, 509-523.
Breiman, L., 2018. Manual on setting up, using and understanding random forests. RColorBrewer MASS, 3 (1), 1-33.
Breiman, L., 2001. Random forests. Machine Learning Kluwer Academic Publishers, 45, 5-32.
Brisco, B. and Brown, R. J., 1995. Multidate SAR/TM synergism for crop classification in western Canada. Photogrammetric Engineering and Remote Sensing, 61 (8), 1009-1014.
Campbell, J.B. and Wynne, R.H. 1996. Introduction to Remote Sensing. The Guilford Press, 408-429.
Cohen, J., 1960. A coefficient of agreement for nominal scales, Educational and Psychological Measurement, 20 (1), 37-46.
Dheeravath, V., Thenkabail, P.S., Chandrakantha, G., Noojipady, P., Reddy, G.P.O., Biradar, C.M., Gumma, M.K. and Velpuri, M., 2010. Irrigated areas of India derived using MODIS 500 m time series for the years 2001-2003. ISPRS Journal of Photogrammetry and Remote Sensing, 65 (1), 42-59.
Gasparovic, M. and Jogun, T., 2017. The effect of using Sentinel-2 bands on land-cover classification. International Journal of Remote Sensing, 39 (3), 822-841.
Gömez, C., White J.C. and Wulder, M. A., 2016. Optical remotely sensed time series data for land cover classification: a review. ISPRS Journal of Photogrammetry and Remote Sensing, 116, 55-72.
Gumma, M. K., Nelson, A., Thenkabail P. S. and Singh, A. N., 2011. Mapping rice areas of South Asia using MODIS multitemporal data. Journal of Applied Remote Sensing, 26, 1-27.
Horning, N., 2011. Random forests: An algorithm for image classification and generation of continuous fields data sets. International Conference on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences (GIS-IDEAS). Immitzer, M., Vuolo, F. and Atzberger, C., 2016. First Experience with Sentinel-2 Data for Crop and Tree Species Classifications in Central Europe. Remote Sensing, 8, 166-193.
Inglada, J., Vincent, A., Arias M. and & Marais-Sicre, C., 2016. Improved early crop type identification by joint use of high temporal resolution sar and optical image time series. Remote Sensing, 8 (5), 362-383. İrfanoğlu, F., ve Balçık, F. B., 2018. Arazi örtüsü ve arazi kullanımı sınıflarının sentinel-2 görüntüsü ve nesne tabanlı sınıflandırma yöntemiyle belirlenmesi. VII. Uzaktan Algılama-CBS Sempozyumu.
Liaw, A. and Wiener, M., 2002. Classification and regression by random forest, R News, 2 (3), 18-22.
Long, J.A., Lawrance, R. L., Greenwood, M. C., Marshall, L. and Miller, P.R., 2013. Object-oriented crop classification using multitemporal ETM+ SLC-off imagery and random forest. Remote Sensing, 50, 418-436.
Noi, P.T. and Kappas, M., 2018. Comparison of random forest, k-nearest neighbor, and support vector machine classifiers for land cover classification using sentinel-2 imagery. Remote Sensing, 18 (1), 18-38.
Pal, M., 2005. Random forest classifier for remote sensing classification. International Journal of Remote Sensing, 26 (1), 217-222.
Peña-Barragán, J. M., Ngugi, M. K., Plant R. E. and Six, J., 2011. Object-based crop identification using multiple vegetation indices, textural features and crop phenology. Remote Sensing of Environment, 115, 1301-1316.
Pirotti, F., Sunar, F. and Piragnolo, M., 2016. Benchmark of machine learning methods for classification of a sentinel-2 image. The International Archives of the Photogrammetry and, Remote Sensing and Spatial Information Sciences, 41, 335-340.
Powell, S. L., Pflugmacher, D., Kirschbaum, A.A., Kim, Y. and Cohen, W.B., 2007. Moderate resolution remote sensing alternatives: a review of Landsat-like sensors and their applications. Journal of Applied Remote Sensing, 1 (1), 1-16. Rodriguez-Galiano, V. F., Ghimire, B., Rogan, J., Chica-Olmo, M. and Rigol-Sanchez, J. P., 2012. An assessment of the effectiveness of a random forest classifier for landcover classification. ISPRS Journal of Photogrammetry and Remote Sensing, 67, 93–104.
Sakamoto, T., Yokozawa, M., Toritani H. and Shibayama, M., 2005. A crop phenology detection method using time-series MODIS data. Remote Sensing of Environment, 96, 366-374.
Simonneaux, V., Duchemin, B., Helson, D., Er-Raki, S., Olioso, A. and Chehbouni, G. A., 2008. The use of high-resolution image time series for crop classification and evapotranspiration estimate over an irrigated area in central Morocco. International Journal of Remote Sensing, 29 (1), 95-116.
Sonobe, R., Yamaya, Y., Tani, H., Wang, X., Kobayashi N. and Mochizuki, K., 2017. Assessing the suitability of data from sentinel-1A and 2A for crop classification. GIScience & Remote Sensing, 54 (6), 918-938.
Sunar, F., Özkan, C. ve Osmanoğlu, B., 2011. Uzaktan Algılama. Prof. Dr. Filiz Sunar (Editör), Anadolu Üniversitesi Yayınları, 150-178.
Thenkabail, P. S., Hanjra, M. A., Dheeravath, V. and Gumma, M., 2010. A holistic view of global croplands and their water use for ensuring global food security in the 21st century through advanced remote sensing and non-remote sensing approaches. Remote Sensing, 2, 211- 261.
Thenkabail, P. S., Hanjra, M. A., Dheeravath, V. and Gumma, M., 2011. Advances in Environmental Remote Sensing: Sensors, Algorithms and Applications. Qihao Weng (Editor), CRC Press, 383-419.
Thenkabail, P. and Wu, Z., 2012. An automated cropland classification algorithm (ACCA) for Tajikistan by combining Landsat, MODIS, and secondary data. Remote Sensing, 4, 2890–2918. Topaloğlu, R. H., Sertel, E. and Musaoğlu, N., 2016. Assessment of classification accuracies of sentinel-2 and Landsat-8 data for land cover/use mapping. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 41, 1055-1059.
Turker, M. and Arikan, M., 2005. Sequential masking classification of multi-temporal Landsat7 ETM+ images for field-based crop mapping in Karacabey Turkey. International Journal of Remote Sensing, 26 (17), 3813-3830.
Vuolo, F., Neuwirth, M., Immitzer, M., Atzberger, C. and NgW, T., 2018. How much does multi-temporal Sentinel-2 data improve crop type classification?. International Journal Earth Observation Geoinformation, 72, 122-130.
Wesseling, J. G. and Fedes, R. A., 2006. Assessing crop water productivity from field to regional scale. Agricultural Water Management, 86 (1), 30-39.
1-https://sentinel.esa.int/web/sentinel/user-guides. (03.05.2020).
2-https://desktop.arcgis.com/en/documentation/ (03.05.2020).
3- https://www.mathworks.com/ (03.05.2020).