Comparative Analysis of MODIS Time-Series Classification Using Support Vector Machines and Methods Based upon Distance and Similarity Measures in the Brazilian Cerrado-Caatinga Boundary

We have mapped the primary native and exotic vegetation that occurs in the Cerrado-Caatinga transition zone in Central Brazil using MODIS-NDVI time series (product MOD09Q1) data over a two-year period (2011–2013). Our methodology consists of the following steps: (a) the development of a three-dimensional cube composed of the NDVI-MODIS time series; (b) the removal of noise; (c) the selection of reference temporal curves and classification using similarity and distance measures; and (d) classification using support vector machines (SVMs). We evaluated different temporal classifications using similarity and distance measures of land use and land cover considering several combinations of attributes. Among the classification using distance and similarity measures, the best result employed the Euclidean distance with the NDVI-MODIS data by considering more than one reference temporal curve per class and adopting six mapping classes. In the majority of tests, the SVM classifications yielded better results than other methods. The best result among all the tested methods was obtained using the SVM classifier with a fourth-degree polynomial kernel; an overall accuracy of 80.75% and a Kappa coefficient of 0.76 were obtained. Our results demonstrate the potential of vegetation studies in semiarid ecosystems using time-series data.

[1]  P. R. Meneses,et al.  Spectral Correlation Mapper ( SCM ) : An Improvement on the Spectral Angle Mapper ( SAM ) , 2000 .

[2]  Leila Maria Garcia Fonseca,et al.  Mapeamento da cobertura vegetal em escala regional do Estado de Minas Gerais utilizando imagens MODIS , 2010 .

[3]  Renato Fontes Guimarães,et al.  Spatial Patterns of Fire Recurrence Using Remote Sensing and GIS in the Brazilian Savanna: Serra do Tombador Nature Reserve, Brazil , 2014, Remote. Sens..

[4]  Clement Atzberger,et al.  Exploiting the Classification Performance of Support Vector Machines with Multi-Temporal Moderate-Resolution Imaging Spectroradiometer (MODIS) Data in Areas of Agreement and Disagreement of Existing Land Cover Products , 2012, Remote. Sens..

[5]  Chong-Yung Chi,et al.  A convex analysis-based minimum-volume enclosing simplex algorithm for hyperspectral unmixing , 2009, IEEE Trans. Signal Process..

[6]  R. Lunetta,et al.  Land-cover change detection using multi-temporal MODIS NDVI data , 2006 .

[7]  Fred A. Kruse,et al.  Comparison of airborne hyperspectral data and EO-1 Hyperion for mineral mapping , 2003, IEEE Trans. Geosci. Remote. Sens..

[8]  Geoffrey M. Henebry,et al.  Spatio-Temporal Statistical Methods for Modelling Land Surface Phenology , 2010 .

[9]  Yosio Edemir Shimabukuro,et al.  Combining noise-adjusted principal components transform and median filter techniques for denoising modis temporal signatures , 2012 .

[10]  A. Huete,et al.  Overview of the radiometric and biophysical performance of the MODIS vegetation indices , 2002 .

[11]  Damien Sulla-Menashe,et al.  Winter wheat area estimation from MODIS-EVI time series data using the Crop Proportion Phenology Index , 2012 .

[12]  Alan R. Gillespie,et al.  Remote Sensing of Landscapes with Spectral Images , 2006 .

[13]  Renato Fontes Guimarães,et al.  Standardized Time-Series and Interannual Phenological Deviation: New Techniques for Burned-Area Detection Using Long-Term MODIS-NBR Dataset , 2015, Remote. Sens..

[14]  Andreas Christmann,et al.  Support vector machines , 2008, Data Mining and Knowledge Discovery Handbook.

[15]  Veraldo Liesenberg,et al.  Análise da dinâmica sazonal e separabilidade espectral de algumas fitofisionomias do cerrado com índices de vegetação dos sensores MODIS/TERRA e AQUA , 2007 .

[16]  Gregory Asner,et al.  Endmember bundles: a new approach to incorporating endmember variability into spectral mixture analysis , 2000, IEEE Trans. Geosci. Remote. Sens..

[17]  P. Atkinson,et al.  Inter-comparison of four models for smoothing satellite sensor time-series data to estimate vegetation phenology , 2012 .

[18]  Bernardo Rudorff,et al.  Monitoring biennial bearing effect on coffee yield using modis remote sensing imagery , 2012, 2012 IEEE International Geoscience and Remote Sensing Symposium.

[19]  R. Alves,et al.  Can campo rupestre vegetation be floristically delimited based on vascular plant genera? , 2010, Plant Ecology.

[20]  Anastasios N. Venetsanopoulos,et al.  Kernel Discriminant Learning with Application to Face Recognition , 2005 .

[21]  Jan Verbesselt,et al.  Assessing intra-annual vegetation regrowth after fire using the pixel based regeneration index , 2011 .

[22]  K. M. Wong,et al.  Some statistical properties of median filters , 1981 .

[23]  V. F. Dutra,et al.  Three New Species of Mimosa (Leguminosae) from Minas Gerais, Brazil , 2013 .

[24]  V. Vapnik.  Estimation of Dependences Based on Empirical Data , 2006 .

[25]  Rubens Manoel dos Santos,et al.  Riqueza e similaridade florística de oito remanescentes florestais no norte de Minas Gerais, Brasil , 2007 .

[26]  Hugo Carrão,et al.  Contribution of multispectral and multitemporal information from MODIS images to land cover classification , 2008 .

[27]  Russell G. Congalton,et al.  Assessing the accuracy of remotely sensed data : principles and practices , 1998 .

[28]  Jai Singh Parihar,et al.  Comparison of Two Data Smoothing Techniques for Vegetation Spectra Derived From EO-1 Hyperion , 2011 .

[29]  Xiang Zhao,et al.  Distribution and Variation of Forests in China from 2001 to 2011: A Study Based on Remotely Sensed Data , 2013 .

[30]  Sander Veraverbeke,et al.  The temporal dimension of differenced Normalized Burn Ratio (dNBR) fire/burn severity studies: the case of the large 2007 Peloponnese wildfires in Greece. , 2010 .

[31]  Mark A. Friedl,et al.  Mapping Crop Cycles in China Using MODIS-EVI Time Series , 2014, Remote. Sens..

[32]  Jennifer N. Hird,et al.  Noise reduction of NDVI time series: An empirical comparison of selected techniques , 2009 .

[33]  T. Sakamoto,et al.  Detecting temporal changes in the extent of annual flooding within the cambodia and the vietnamese mekong delta from MODIS time-series imagery , 2007 .

[34]  V. Radeloff,et al.  Author's Personal Copy Mapping Abandoned Agriculture with Multi-temporal Modis Satellite Data , 2022 .

[35]  R. J. Scholes,et al.  Leaf green-up in a semi-arid African savanna –separating tree and grass responses to environmental cues , 2007 .

[36]  Peter E. Hart,et al.  Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.

[37]  J. A. Schell,et al.  Monitoring vegetation systems in the great plains with ERTS , 1973 .

[38]  A. P. Magalhães,et al.  A GEOMORFOLOGIA DO PLANALTO DO ESPINHAÇO SETENTRIONAL AVALIADA PARA A IMPLANTAÇÃO DE BARRAGEM: A HUE DE IRAPÉ – MG , 1997 .

[39]  Alan R. Gillespie,et al.  A New Approach to Change Vector Analysis Using Distance and Similarity Measures , 2011, Remote. Sens..

[40]  Alan R. Gillespie,et al.  Radiometric Normalization of Temporal Images Combining Automatic Detection of Pseudo-Invariant Features from the Distance and Similarity Spectral Measures, Density Scatterplot Analysis, and Robust Regression , 2013, Remote. Sens..

[41]  Mario Winter,et al.  N-FINDR: an algorithm for fast autonomous spectral end-member determination in hyperspectral data , 1999, Optics & Photonics.

[42]  T. Sakamoto,et al.  A crop phenology detection method using time-series MODIS data , 2005 .

[43]  Vojislav Kecman,et al.  Support Vector Machines – An Introduction , 2005 .

[44]  Renato Fontes Guimarães,et al.  IDENTIFICAÇÃO REGIONAL DA FLORESTA ESTACIONAL DECIDUAL NA BACIA DO RIO PARANÃ A PARTIR DA ANÁLISE MULTITEMPORAL DE IMAGENS MODIS , 2006 .

[45]  Nilton Correia da Silva,et al.  Classificação de padrões de savana usando assinaturas temporais NDVI do sensor MODLS no Parque Nacional Chapada dos Veadeiros , 2008 .

[46]  Sarah E. Metcalfe,et al.  Dynamic changes in savanna and seasonally dry vegetation through time , 2007 .

[47]  Steven I. Higgins,et al.  Is there a temporal niche separation in the leaf phenology of savanna trees and grasses? , 2011 .

[48]  Osmar Abílio de Carvalho Júnior,et al.  CHARACTERIZATION OF THE AGRICULTURE OCCUPATION IN THE CERRADO BIOME USING MODIS TIME-SERIES , 2013 .

[49]  M. H. O. Pinheiro,et al.  Contribution to the discussions on the origin of the cerrado biome: Brazilian savanna. , 2010, Brazilian journal of biology = Revista brasleira de biologia.

[50]  Fred A. Kruse,et al.  The Spectral Image Processing System (SIPS) - Interactive visualization and analysis of imaging spectrometer data , 1993 .

[51]  Renato Fontes Guimarães,et al.  Probability Density Components Analysis: A New Approach to Treatment and Classification of SAR Images , 2014, Remote. Sens..

[52]  G. Sánchez‐Azofeifa,et al.  Extent and conservation of tropical dry forests in the Americas , 2010 .

[53]  R. Pennington,et al.  Woody Plant Diversity, Evolution, and Ecology in the Tropics: Perspectives from Seasonally Dry Tropical Forests , 2009 .

[54]  Chein-I Chang,et al.  A New Growing Method for Simplex-Based Endmember Extraction Algorithm , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[55]  Anatoly A. Gitelson,et al.  An evaluation of MODIS 8- and 16-day composite products for monitoring maize green leaf area index , 2012 .

[56]  Esteban G. Jobbágy,et al.  Cultivating the dry forests of South America: Diversity of land users and imprints on ecosystem functioning , 2015 .

[57]  Alfredo Huete,et al.  Analysis of Cerrado Physiognomies and Conversion in the MODIS Seasonal-Temporal Domain , 2005 .

[58]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[59]  Armin Shmilovici,et al.  Support Vector Machines , 2005, Data Mining and Knowledge Discovery Handbook.

[60]  J. Boardman.  Automating spectral unmixing of AVIRIS data using convex geometry concepts , 1993 .

[61]  P. Switzer,et al.  A transformation for ordering multispectral data in terms of image quality with implications for noise removal , 1988 .

[62]  T. Ricketts,et al.  Confronting a biome crisis: global disparities of habitat loss and protection , 2004 .

[63]  Jin Chen,et al.  A simple method for reconstructing a high-quality NDVI time-series data set based on the Savitzky-Golay filter , 2004 .

[64]  Geraldo Wilson Fernandes,et al.  Economic Environmental Management Tools in the Serra Do Espinhaço Biosphere Reserve , 2012 .

[65]  Nilton Correia da Silva,et al.  AVALIAÇÃO DOS CLASSIFICADORES ESPECTRAIS DE MÍNIMA DISTÂNCIA EUCLIDIANA E SPECTRAL CORRELATION MAPPER EM SÉRIES TEMPORAIS NDVI-MODIS NO CAMPO DE INSTRUÇÃO MILITAR DE FORMOSA (GO) , 2009, Revista Brasileira de Cartografia.

[66]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[67]  Brian Curtiss,et al.  A method for manual endmember selection and spectral unmixing , 1996 .

[68]  Fernanda P. Werneck,et al.  The diversification of eastern South American open vegetation biomes: Historical biogeography and perspectives , 2011 .

[69]  Y. Nunes,et al.  Changes in tree and liana communities along a successional gradient in a tropical dry forest in south-eastern Brazil , 2009, Plant Ecology.

[70]  Fred A. Kruse,et al.  Analysis of Imaging Spectrometer Data Using $N$ -Dimensional Geometry and a Mixture-Tuned Matched Filtering Approach , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[71]  Dan Hammer,et al.  Alerts of forest disturbance from MODIS imagery , 2014, Int. J. Appl. Earth Obs. Geoinformation.

[72]  J. Pirani,et al.  Areas of endemism in the Espinhaco Range in Minas Gerais, Brazil , 2011 .

[73]  A. Savitzky,et al.  Smoothing and Differentiation of Data by Simplified Least Squares Procedures. , 1964 .

[74]  Jungho Im,et al.  ISPRS Journal of Photogrammetry and Remote Sensing , 2022 .

[75]  Martin Herold,et al.  On the Suitability of MODIS Time Series Metrics to Map Vegetation Types in Dry Savanna Ecosystems: A Case Study in the Kalahari of NE Namibia , 2009, Remote. Sens..

[76]  J. Mustard,et al.  Wavelet analysis of MODIS time series to detect expansion and intensification of row-crop agriculture in Brazil , 2008 .

[77]  A. Atiya,et al.  Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond , 2005, IEEE Transactions on Neural Networks.

[78]  Bruce Dickson,et al.  Maximum noise fraction method reveals detail in aerial gamma-ray surveys , 2000 .

[79]  G. Arturo Sánchez-Azofeifa,et al.  Sustainability of tropical dry forests: Two case studies in southeastern and central Brazil , 2009 .

[80]  C. Portillo-Quintero,et al.  Monitoring deforestation with MODIS Active Fires in Neotropical dry forests: An analysis of local-scale assessments in Mexico, Brazil and Bolivia , 2013 .

[81]  D. Roy,et al.  An overview of MODIS Land data processing and product status , 2002 .

[82]  Mingguo Ma,et al.  Comparison of Eight Techniques for Reconstructing Multi-Satellite Sensor Time-Series NDVI Data Sets in the Heihe River Basin, China , 2014, Remote. Sens..

[83]  R. Schafer,et al.  What Is a Savitzky-Golay Filter? , 2022 .

[84]  Margaret Kalacska,et al.  Research priorities for neotropical dry forests , 2005 .

[85]  Flávio Jorge Ponzoni,et al.  Calibration of a Species-Specific Spectral Vegetation Index for Leaf Area Index (LAI) Monitoring: Example with MODIS Reflectance Time-Series on Eucalyptus Plantations , 2012, Remote. Sens..

[86]  Mukesh Singh Boori,et al.  Land use change detection for environmental management: using multi-temporal, satellite data in the Apodi Valley of northeastern Brazil , 2010 .

[87]  Felipe Salvo Aires,et al.  Fires in the cerrado, the Brazilian savanna , 2009 .

[88]  Laerte Guimarães Ferreira,et al.  Distribution Patterns of Burned Areas in the Brazilian Biomes: An Analysis Based on Satellite Data for the 2002-2010 Period , 2012, Remote. Sens..

[89]  Ronald W. Schafer,et al.  What Is a Savitzky-Golay Filter? [Lecture Notes] , 2011, IEEE Signal Processing Magazine.

[90]  Laerte Guimarães Ferreira,et al.  Biophysical Properties of Cultivated Pastures in the Brazilian Savanna Biome: An Analysis in the Spatial-Temporal Domains Based on Ground and Satellite Data , 2013, Remote. Sens..

[91]  G. Colli,et al.  Revisiting the historical distribution of Seasonally Dry Tropical Forests: new insights based on palaeodistribution modelling and palynological evidencegeb , 2011 .

[92]  Stuart E. Marsh,et al.  Phenological Characterization of Desert Sky Island Vegetation Communities with Remotely Sensed and Climate Time Series Data , 2010, Remote. Sens..

[93]  Jan de Leeuw,et al.  Length of Growing Period over Africa: Variability and Trends from 30 Years of NDVI Time Series , 2013, Remote. Sens..

[94]  Christiane Schmullius,et al.  Assessing effects of temporal compositing and varying observation periods for large-area land-cover mapping in semi-arid ecosystems: Implications for global monitoring , 2011 .

[95]  Edson E. Sano,et al.  Land cover mapping of the tropical savanna region in Brazil , 2010, Environmental monitoring and assessment.

[96]  Maurice D. Craig,et al.  Minimum-volume transforms for remotely sensed data , 1994, IEEE Trans. Geosci. Remote. Sens..