ASSESSING THE TRANSFERABILITY OF MACHINE LEARNING ALGORITHMS USING CLOUD COMPUTING AND EARTH OBSERVATION DATASETS FOR AGRICULTURAL LAND USE/COVER MAPPING

Abstract. Mapping of agricultural land use/cover was initiated since the past several decades for land use planning, change detection analysis, crop yield monitoring etc. using earth observation datasets and traditional parametric classifiers. Recently, machine learning, cloud computing, Google Earth Engine (GEE) and open source earth observation datasets widely used for fast, cost-efficient and precise agricultural land use/cover mapping and change detection analysis. Main objective of this study was to assess the transferability of the machine learning algorithms for land use/cover mapping using cloud computing and open source earth observation datasets. In this study, the Landsat TM (L5, L8) of 2018, 2009 and 1998 were selected and median reflectance of spectral bands in Kharif and Rabi season were used for the classification. In addition, three important machine learning algorithms such as Support Vector Machine with Radial Basis Function (SVM-RBF), Random forest (RF) and Classification and Regression Tree (CART) were selected to evaluate the performance in transferability for agricultural land use classification using GEE. Seven land use/cover classes such as built-up, cropland, fallow land, vegetation etc. were selected based on literature review and local land use classification scheme. In this classification, several strategies were employed such as feature extraction, feature selection, parameter tuning, sensitivity analysis on size of training samples, transferability analysis to assess the performance of the selected machine learning algorithms for land use/cover classification. The result shows that SVM-RBF outperforms the RF and CART for both spatial and temporal transferability analysis. This result is very helpful for agriculture and remote sensing scientist to suggest promising guideline to land use planner and policy-makers for efficient land use mapping, change detection analysis, land use planning and natural resource management.

[1]  Michael J. Garay,et al.  An operational MISR pixel classifier using support vector machines , 2007 .

[2]  Prosper Basommi Laari,et al.  Modelling of land use land cover change using earth observation data-sets of Tons River Basin, Madhya Pradesh, India , 2018 .

[3]  S. Wu,et al.  Per‐field urban land use classification based on tax parcel boundaries , 2007 .

[4]  Tiho Ancev,et al.  Improving the Accuracy of Land Use and Land Cover Classification of Landsat Data Using Post-Classification Enhancement , 2009, Remote. Sens..

[5]  P. Srivastava,et al.  Delineation and classification of rural–urban fringe using geospatial technique and onboard DMSP–Operational Linescan System , 2018 .

[6]  Bernhard Schölkopf,et al.  Remote Sensing Feature Selection by Kernel Dependence Measures , 2010, IEEE Geoscience and Remote Sensing Letters.

[7]  Justin Morgenroth,et al.  Developments in Landsat Land Cover Classification Methods: A Review , 2017, Remote. Sens..

[8]  Christian Thiel,et al.  Mapping CORINE Land Cover from Sentinel-1A SAR and SRTM Digital Elevation Model Data using Random Forests , 2015, Remote. Sens..

[9]  Cailan Gong,et al.  The Hughes phenomenon in hyperspectral classification based on the ground spectrum of grasslands in the region around Qinghai Lake , 2013, Other Conferences.

[10]  Giles M. Foody,et al.  Toward intelligent training of supervised image classifications: directing training data acquisition for SVM classification , 2004 .

[11]  Lei Ma,et al.  Training set size, scale, and features in Geographic Object-Based Image Analysis of very high resolution unmanned aerial vehicle imagery , 2015 .

[12]  Nirbhow Jap Singh,et al.  Cropping pattern of Uttar Pradesh using IRS-P6 (AWiFS) data , 2011 .

[13]  Onisimo Mutanga,et al.  Land-use/cover classification in a heterogeneous coastal landscape using RapidEye imagery: evaluating the performance of random forest and support vector machines classifiers , 2014 .

[14]  Eduarda Martiniano de Oliveira Silveira,et al.  Assessment of geostatistical features for object-based image classification of contrasted landscape vegetation cover , 2017 .

[15]  Cláudia Maria de Almeida,et al.  Evaluating Sentinel-2 and Landsat-8 Data to Map Sucessional Forest Stages in a Subtropical Forest in Southern Brazil , 2017, Remote. Sens..

[16]  S. Mustak.  CORRECTION OF ATMOSPHERIC HAZE IN RESOURCESAT-1 LISS-4 MX DATA FOR URBAN ANALYSIS: AN IMPROVED DARK OBJECT SUBTRACTION APPROACH , 2013 .

[17]  Chris Bacon,et al.  Using Google Earth Engine to Map Complex Shade-Grown Coffee Landscapes in Northern Nicaragua , 2018, Remote. Sens..

[18]  José A. Sobrino,et al.  Land use classification from multitemporal Landsat imagery using the Yearly Land Cover Dynamics (YLCD) method , 2011, Int. J. Appl. Earth Obs. Geoinformation.

[19]  Xin Lu,et al.  A random forest of combined features in the classification of cut tobacco based on gas chromatography fingerprinting. , 2010, Talanta.

[20]  Gabriele Bitelli,et al.  Preliminary Comparison of Sentinel-2 and Landsat 8 Imagery for a Combined Use , 2016, Remote. Sens..

[21]  Huadong Guo,et al.  Pixel- and feature-level fusion of hyperspectral and lidar data for urban land-use classification , 2015 .

[22]  Vladimir Vapnik,et al.  Chervonenkis: On the uniform convergence of relative frequencies of events to their probabilities , 1971 .

[23]  Yi-Shiang Shiu,et al.  A Comparative Analysis of Machine Learning with WorldView-2 Pan-Sharpened Imagery for Tea Crop Mapping , 2016, Sensors.

[24]  Paul M. Mather,et al.  Assessment of the effectiveness of support vector machines for hyperspectral data , 2004, Future Gener. Comput. Syst..

[25]  Albert Rango,et al.  Texture and Scale in Object-Based Analysis of Subdecimeter Resolution Unmanned Aerial Vehicle (UAV) Imagery , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[26]  Prashant K. Srivastava,et al.  Prioritization of Malesari mini-watersheds through morphometric analysis: a remote sensing and GIS perspective , 2013, Environmental Earth Sciences.

[27]  Vladimir Vapnik.  Introduction: Four Periods in the Research of the Learning Problem , 1995 .

[28]  Debra P. C. Peters,et al.  The changing landscape : ecosystem responses to urbanization and pollution across climatic and societal gradients , 2008 .

[29]  Karen C. Seto,et al.  Using logit models to classify land cover and land‐cover change from Landsat Thematic Mapper , 2005 .

[30]  Q. Henry Wu,et al.  Local prediction of non-linear time series using support vector regression , 2008, Pattern Recognit..

[31]  Massimiliano Pittore,et al.  Performance Evaluation of Machine Learning Algorithms for Urban Pattern Recognition from Multi-spectral Satellite Images , 2014, Remote. Sens..

[32]  Martin Kappas,et al.  Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery , 2017, Sensors.

[33]  Giles M. Foody,et al.  A relative evaluation of multiclass image classification by support vector machines , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[34]  Dawei Han,et al.  Artificial intelligence techniques for clutter identification with polarimetric radar signatures , 2012 .

[35]  Sung-Bae Cho,et al.  Fingerprint classification using one-vs-all support vector machines dynamically ordered with naive Bayes classifiers , 2008, Pattern Recognit..

[36]  Giles M. Foody,et al.  Sample size determination for image classification accuracy assessment and comparison , 2009 .

[37]  Hannah Jacobson,et al.  A novel approach to mapping land conversion using Google Earth with an application to East Africa , 2015, Environ. Model. Softw..

[38]  P. Srivastava,et al.  Predicting Spatial and Decadal LULC Changes Through Cellular Automata Markov Chain Models Using Earth Observation Datasets and Geo-information , 2015, Environmental Processes.

[39]  Lei Shi,et al.  Mapping Vegetation and Land Use Types in Fanjingshan National Nature Reserve Using Google Earth Engine , 2018, Remote. Sens..

[40]  Edzer Pebesma,et al.  Using Google Earth Engine to detect land cover change: Singapore as a use case , 2018 .

[41]  Yang Shao,et al.  Comparison of support vector machine, neural network, and CART algorithms for the land-cover classification using limited training data points , 2012 .

[42]  Nicolas Courty,et al.  Sparse Hilbert Schmidt Independence Criterion and Surrogate-Kernel-Based Feature Selection for Hyperspectral Image Classification , 2017, IEEE Transactions on Geoscience and Remote Sensing.