Satellite-based system of area estimation for main agricultural crops of Ukraine
The article studies one of the most important issues of agricultural production maintenance – development of a system of crops area estimation in Ukraine. The objective of this paper is to describe the similar system that uses high resolution satellite data and operational agrometeorological data from the network of the Hydrometeorological Centre of Ukraine as input information. The system is based on step-by-step solving of the following tasks: obtaining geoinformation data for individual agricultural crops; development of methods for multispectral satellite images classification; development of software applications to automate the process of these images classification with subsequent classification of crop areas. The research uses the following algorithms (or classifiers) to classify the agricultural land: SVM (support vector machine), RF ("random forest") and NN (neural networks). The choice of the most accurate of them formed the basis of the general method of classification. The values of spectral characteristics of red and infrared channels of a complete set of cloudless satellite images during the growing period were used as input data (features). As a result, in 2018 some test calculations were conducted to estimate the area of agricultural crops in Kyiv Region. The results of evaluation of accuracy of the satellite-based agricultural crops area estimation using the statistical data showed that the lowest accuracy is typical for winter wheat and corn. The accuracy of soybeans and spring barley classification is quite low for most of the tested fields. Sunflower and rapeseed crops showed the highest accuracy. In order to improve the accuracy of classification, it is necessary to introduce more classification features (in a temporary aspect) by processing more satellite images during the growing period, and to increase the number of test samples through systematic sampling of ground data across the regions in Ukraine. We suggest using the scheme of main agricultural crops area estimation satellite-based system by the Hydrometeorological Centre of Ukraine.
Kleshchenko, A. (1986). Otsenka sostoyaniya zernovykh kul’tur s primeneniem distantsionnykh metodov. [Assessment of grain crops state using remote sensing methods]. Leningrad: Gidrometeoizdat (in Russ.)
Kleshchenko, A. & Virchenko, O. (1989). K voprosu ob ispol’zovanii sputnikovykh dannykh pri operativnom agrometeoobsluzhivanii narodnogo khozyaystva [On the use of satellite data in the operational agrometeorological service of the national economy.]. Trudy VNIISKHM [Proceedings of All-Russian Research Institute of Agricultural Meteorology], №25. (in Russ.)
Kyvobok, O., Kryvoshein, O. & Adamenko, T. (2018). [Peculiarities of technological adaptation of the cgms system for agricultural crops monitoring in Ukraine]. Ukr. gìdrometeorol. ž. [Ukrainian hydrometeorological journal], 22, pp. 64-79. https://doi.org/10.31481/ uhmj.22.2018.07 (in Russ.)
Fernandez, J. (2015). Analysis of Classification Algorithms for Crop Detection using LANDSAT 8 images. Universidade nova de Lisboa.
Oruc, M., Marangoz, A. & Buyuksalih, G. (2004). Comparison of pixel-based and object-oriented classification approaches using Landsat-7 ETM spectral bands. Proceedings of the ISRPS 2004 Annual Conference, pp. 19–23.
Whiteside, T. & Ahmad, W. (2005). A comparison of object-oriented and pixelbasedclassification methods for mapping land cover in northern Australia. Proceedings of SSC2005 Spatial intelligence, innovation and praxis: The nationalbiennial Conference of the Spatial Sciences Institute, pp. 1225–1231.
Platt, R.V. & Rapoza, L. (2008). An Evaluation of an Object-Oriented Paradigm for Land Use/Land Cover Classification. The Professional Geographer, 60(1), pp. 87–100.
Zhou, W., Troy, A. & Grove, M. (2008). Object-based land cover classification and change analysis in the Baltimore metropolitan area using multitemporaligh resolution remote sensing data. Sensors, 8(3), pp. 1613–1636.
Mohan, B.K., Ladha, S.I. (2009). CSRE. Comparison of object based and pixel based classification of high resolution satellite images using artificial neural networks. IIT Bombay, Mumbai.
Duro, D.C., Franklin, S.E., & Dubé, M.G. (2012). A comparison of pixel-based and object-based image analysis with selected machine learning algorithms for the classification of agricultural landscapes using SPOT-5 HRG imagery. Remote Sensing of Environment, 118, pp. 259–272.
Jebur, M.N., Mohd Shafri, H.Z. et al. (2013). Perpixel and object-oriented classification methods for mapping urban land cover extraction using SPOT 5 imagery. Geocarto International (ahead-of-print), pp. 1–15.
Huang, C, Davis L., & Townshend, J. (2002). An assessment of support vector machines for land cover classification. International Journal of Remote Sensing, 23(4), pp. 725–749.
Pal, M. & Mather, P.M. (2003). An assessment of the effectiveness of decision tree methods for land cover classification. Remote sensing of environment, 86(4), pp. 554–565.
Otukei, J. & Blaschke, T. (2010). Land cover change assessment using decision trees, support vector machines and maximum likelihood classification algorithms. International Journal of Applied Earth Observation and Geoinformation, 12, pp. S27–S31.
Rodriguez-Galiano, V. et al. (2012). An assessment of the effectiveness of a random forest classifier for landcover classification. ISPRS Journal of Photogrammetry and Remote Sensing, 67, pp. 93–104.
Dingle Robertson, L. & King, D.J. (2011). Comparison of pixel-and object-based classification in land cover change mapping. International Journal of Remote Sensing, 32(6).
Kussul, N., Skakun, S. & Shelestov, A. (2014) Analiz riskov chrezvychaynykh situatsiy na osnove sputnikovykh dannykh [Satellite based emergency risk analysis]. Kiev “Modeli i tehnologii”. (in Russ).
Haykin, S. (1994). Neural Networks. A comprehensive Foundation. New Jersey: Prentice Hall.
Bishop, C.M. (2006). Pattern Recognition and Machine Learning. New York: Springer Science + Business Media.
Rumelhart, D.E., Hinton, G.E. & Williams, R.J. (1988). Learning representations by back-propagating errors. Neurocomputing: foundations of research. Cambridge, MA, USA: MIT Press, pp. 696-699.
This work is licensed under a Creative Commons Attribution 4.0 International License.