Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2017, Vol. 14, No. 5, pp. 172-183
Evaluation of the possibilities to classify agricultural lands using multi-seasonal satellite data processing
1 Moscow State University for Geodesy and Cartography, Moscow, Russia
Accepted: 25.09.2017
DOI: 10.21046/2070-7401-2017-14-5-172-183
The article is devoted to evaluation of the possibilities to classify agricultural lands type (arable land, abandoned land, fodder lands) and their condition (agricultural lands overgrowing) using multi-seasonal satellite data classification method. The analysis was based on combining several Landsat-OLI multi-seasonal images acquired during one vegetation season (2-6 images). Satellite data for the study area were stacked into spectro-temporal images which were used in further processing. In the study, several spectro-temporal combinations were formed including single images for further comparison of the classification results. Automated processing includes the following steps: data preprocessing, stack multiple images into a single image file, supervised classification using maximum likelihood technique and quantitative evaluation of classification accuracy that includes error matrix construction, computation of generalized indicators of accuracy such as overall accuracy, Cohen’s kappa coefficient and others. Classification accuracy assessment showed that using combined spectro-temporal images improves overall and agricultural land classification accuracy compared to single images processing. Using spectro-temporal images also enables visual comparison of agricultural land conditions at different times that allows improving the quality of training sample selection.
Keywords: Landsat 8 OLI, spectro-temporal images, different-season images co-processing, classification, quantitative evaluation of classification accuracy, agricultural land, agricultural lands overgrowing
Full textReferences:
- Bartalev S.A., Egorov V.A., Zharko V.O., Loupian E.A., Plotnikov D.E., Khvostikov S.A., Shabanov N.V., Sputnikovoe kartografirovanie rastitel’nogo pokrova Rossii (Land cover mapping over Russia using Earth observation data), Moscow: IKI RAN, 2016, 208 p.
- Bartalev S.A., Egorov V.A., Loupian E.A., Plotnikov D.E., Uvarov I.A., Raspoznavanie pakhotnykh zemel’ na osnove mnogoletnikh sputnikovykh dannykh spektroradiometra MODIS i lokal’no-adaptivnoi klassifikatsii (Agricaltural lands recognition using multi-year satellite data Spectroradiometer MODIS and locally adaptive classification), Komp’yuternaya optika, 2011, Vol. 35, No. 1, pp. 103–116.
- Loupian E.A., Bartalev S.A., Tolpin V.A., Vozmozhnosti i opyt ispol’zovaniya sputnikovykh servisov semeistva Sozvezdie-Vega dlya resheniya zadach monitoringa sel’skokhozyaistvennykh zemel’ i posevov (Possibilities and experience in the use satellite services-family Constellation-VEGA for the decision of tasks of monitoring agricultural land and crops), Materialy Vserossiiskoi nauchnoi konferentsii (s mezhdunarodnym uchastiem) “Primenenie sredstv distantsionnogo zondirovaniya Zemli v sel’skom khozyaistve”, (Materials of all-Russia Conf. “Application of remote sensing in agriculture”), Saint Petersburg, 16–17 September 2015, Saint Petersburg: FGBNU AFI, 2015, pp. 41–46.
- Lur’e I.K., Geoinformatsionnoe kartografirovanie Metody geoinformatiki i tsifrovoi obrabotki kosmicheskikh snimkov: uchebnik (GIS mapping Methods of Geoinformatics and digital processing of satellite images: a tutorial), Moscow: KDU, 2008, 424 p.
- Marinina O.A., Terekhin E.A., Kirilenko Zh.A., Kurlovich D.M., Koval’chik N.V., Osobennosti distantsionnogo vyyavleniya zalezhnykh uchastkov i problemy tselevogo ispol’zovaniya zemel’ sel’skokhozyaistvennogo naznacheniya (Features remote detection of fallow areas and the problems of target use of lands of agricultural purpose), Sovremennye problemy nauki i obrazovaniya, 2013, No. 5, 535 p.
- Plotnikov D.E., Bartalev S.A., Loupian E.A., Priznaki raspoznavaniya pakhotnykh zemel’ na osnove mnogoletnikh ryadov dannykh sputnikovogo spektroradiometra MODIS (Signs of recognition of arable land on the basis of long data series of satellite MODIS), Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2010, Vol. 7, No. 1, pp. 330–341.
- Popov M.A., Metodologiya otsenki tochnosti klassifikatsii ob’’ektov na kosmicheskikh izobrazheniyakh (The methodology for assessing the classification’s accuracy of objects on space images), Problemy upravleniya i informatiki, 2007, No. 1, pp. 97–103.
- Samsonova V.P., Kondrashkina M.I., Krotov D.G., Chichieva O.A., Raspoznavanie zarastayushchikh zemel’ na snimkakh Landsat 8 (Recognition of the overgrown land on the images Landsat 8), Problemy agrokhimii i ekologii, 2015, No. 1, pp. 53–57.
- Stytsenko E.A., Opyt deshifrirovaniya rastitel’nogo pokrova zemnoi poverkhnosti s ispol’zovaniem raznosezonnykh zonal’nykh kosmicheskikh izobrazhenii (Experience in the interpretation of the vegetation cover of the earth’s surface using season zonal space images), Aktual’nye problemy prirodoobustroistva, kadastra i zemlepol’zovaniya: materialy mezhdunarodnoi nauchno-prakticheskoi konferentsii, posvyashchennoi 95-letiyu fakul’tetu zemleustroistva i kadastrov VGAU (Actual problems of environmental engineering, cadastre and land management: materials of international scientific-practical conference dedicated to the 95th anniversary of the faculty of land management and cadastre, VSAU), Part I, Voronezh: FGBOU VO Voronezhskij GAU, 2016, pp. 245–250.
- Chaban L.N., Avtomatizirovannaya obrabotka aerokosmicheskoi informatsii pri kartografirovanii geoprostranstvennykh dannykh (Automated processing of aerospace information in the mapping of geospatial data. Tutorial), Moscow: MIIGAiK, 2013, 96 p.
- Alcantara C, Kuemmerle T., Baumann M., Bragina E.V., Griffiths P., Knorn J., Muller D., Prishchepov A.V., Schierhorn F., Sieber A., Radeloff V.C., Mapping the Extent of Abandoned Farmland in Central and Eastern Europe Using MODIS Time Series Satellite Data, Environment research letters, 2013, No. 8, pp. 1–9.
- Cohen J., A Coefficient of agreement for nominal scales, Educational and Psychological Measurement, 1960, No. 20, pp. 37–46.
- Congalton R., A Review of Assessing the Accuracy of Classifications of Remotely Sensed Data, Remote Sensing of Environment, 1991, No. 37, pp. 35–46.
- de Beurs K.M., Use of Landsat and MODIS Data to Remotely Estimate Russia’s Sown Area, Journal of Land Use Science, 2014, pp. 1–25.
- Estel S., Kuemmerle T., Alcantara C., Levers C., Prishchepov A., Hostert P., Mapping Farmland Abandonment and Recultivation across Europe Using MODIS NDVI Time Series, Remote Sensing of Environment, 2015, No. 163, pp. 312–325.