Mapping wetland using the object-based stacked generalization method based on multi-temporal optical and SAR data

Abstract Wetland ecosystems have experienced dramatic challenges in the past few decades due to natural and human factors. Wetland maps are essential for the conservation and management of terrestrial ecosystems. This study is to obtain an accurate wetland map using an object-based stacked generalization (Stacking) method on the basis of multi-temporal Sentinel-1 and Sentinel-2 data. Firstly, the Robust Adaptive Spatial Temporal Fusion Model (RASTFM) is used to get time series Sentinel-2 NDVI, from which the vegetation phenology variables are derived by the threshold method. Subsequently, both vertical transmit-vertical receive (VV) and vertical transmit-horizontal receive (VH) polarization backscatters (σ0 VV, σ0 VH) are obtained using the time series Sentinel-1 images. Speckle noise inherent in SAR data, resulting in over-segmentation or under-segmentation, can affect image segmentation and degrade the accuracies of wetland classification. Therefore, we segment Sentinel-2 multispectral images to delineate meaningful objects in this study. Then, in order to reduce data redundancy and computation time, we analyze the optimal feature combination using the Sentinel-2 multispectral images, Sentinel-2 NDVI time series, phenological variables and other vegetation index derived from Sentinel-2 multispectral images, as well as time series Sentinel-1 backscatters at the object level. Finally, the stacked generalization algorithm is utilized to extract the wetland information based on the optimal feature combination in the Dongting Lake wetland. The overall accuracy and Kappa coefficient of the object-based stacked generalization method are 92.46% and 0.92, which are 3.88% and 0.04 higher than that using the pixel-based method. Moreover, the object-based stacked generalization algorithm is superior to single classifiers in classifying vegetation of high heterogeneity areas.

[1]  Avik Bhattacharya,et al.  A Novel Phenology Based Feature Subset Selection Technique Using Random Forest for Multitemporal PolSAR Crop Classification , 2018, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[2]  Christopher J. Post,et al.  Geospatial analysis of land use change in the Savannah River Basin using Google Earth Engine , 2018, Int. J. Appl. Earth Obs. Geoinformation.

[3]  Mengmeng Li,et al.  Vegetable classification in Indonesia using Dynamic Time Warping of Sentinel-1A dual polarization SAR time series , 2019, Int. J. Appl. Earth Obs. Geoinformation.

[4]  M. Mahdianpari,et al.  Random forest wetland classification using ALOS-2 L-band, RADARSAT-2 C-band, and TerraSAR-X imagery , 2017 .

[5]  Li Ma,et al.  Local Manifold Learning-Based $k$ -Nearest-Neighbor for Hyperspectral Image Classification , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[6]  Jie Wang,et al.  Mapping paddy rice planting area in rice-wetland coexistent areas through analysis of Landsat 8 OLI and MODIS images , 2016, Int. J. Appl. Earth Obs. Geoinformation.

[7]  B. Gao.  NDWI—A normalized difference water index for remote sensing of vegetation liquid water from space , 1996 .

[8]  Rajiv Sinha,et al.  Distribution, diversity, and geomorphic evolution of floodplain wetlands and wetland complexes in the Ganga plains of north Bihar, India , 2020 .

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

[10]  Per Jönsson,et al.  TIMESAT - a program for analyzing time-series of satellite sensor data , 2004, Comput. Geosci..

[11]  David H. Wolpert,et al.  No free lunch theorems for optimization , 1997, IEEE Trans. Evol. Comput..

[12]  Yaying Lou,et al.  Hydromorphological processes of Dongting Lake in China between 1951 and 2014 , 2018, Journal of Hydrology.

[13]  Juha Hyyppä,et al.  Comparison of two-dimensional multitemporal Sentinel-2 data with three-dimensional remote sensing data sources for forest inventory parameter estimation over a boreal forest , 2019, Int. J. Appl. Earth Obs. Geoinformation.

[14]  Bruno Basso,et al.  Multi-temporal RADARSAT-2 polarimetric SAR for maize mapping supported by segmentations from high-resolution optical image , 2019, Int. J. Appl. Earth Obs. Geoinformation.

[15]  Paul D. Wagner,et al.  Combining Sentinel-1 and Sentinel-2 data for improved land use and land cover mapping of monsoon regions , 2018, Int. J. Appl. Earth Obs. Geoinformation.

[16]  Dino Ienco,et al.  Combining Sentinel-1 and Sentinel-2 Satellite Image Time Series for land cover mapping via a multi-source deep learning architecture , 2019 .

[17]  Yuri A. Gritz,et al.  Relationships between leaf chlorophyll content and spectral reflectance and algorithms for non-destructive chlorophyll assessment in higher plant leaves. , 2003, Journal of plant physiology.

[18]  A. Ghulam,et al.  Detecting subcanopy invasive plant species in tropical rainforest by integrating optical and microwave (InSAR/PolInSAR) remote sensing data, and a decision tree algorithm , 2014 .

[19]  Jahan Kariyeva,et al.  Comparing Deep Learning and Shallow Learning for Large-Scale Wetland Classification in Alberta, Canada , 2019, Remote. Sens..

[20]  Jiao Qu,et al.  Effects of conversion from boreal natural wetlands to rice paddy fields on the dynamics of total dissolved iron during extreme precipitation events. , 2020, Chemosphere.

[21]  Maycira Costa,et al.  Landcover classification of the Lower Nhecolândia subregion of the Brazilian Pantanal Wetlands using ALOS/PALSAR, RADARSAT-2 and ENVISAT/ASAR imagery , 2013 .

[22]  Xiaoling Chen,et al.  Four decades of winter wetland changes in Poyang Lake based on Landsat observations between 1973 and 2013 , 2015 .

[23]  Eric W. Gill,et al.  A new fully convolutional neural network for semantic segmentation of polarimetric SAR imagery in complex land cover ecosystem , 2019, ISPRS Journal of Photogrammetry and Remote Sensing.

[24]  A. Gitelson,et al.  Remote estimation of chlorophyll content in higher plant leaves , 1997 .

[25]  Zhe Zhu,et al.  Mapping forest change using stacked generalization: An ensemble approach , 2018 .

[26]  Cong Xu,et al.  Effects of pre-processing methods on Landsat OLI-8 land cover classification using OBIA and random forests classifier , 2018, Int. J. Appl. Earth Obs. Geoinformation.

[27]  Yun Zhang,et al.  Wetland Classification Using Deep Convolutional Neural Network , 2018, IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium.

[28]  Limin Wang,et al.  Optimal Segmentation Scale Parameter, Feature Subset and Classification Algorithm for Geographic Object-Based Crop Recognition Using Multisource Satellite Imagery , 2019, Remote. Sens..

[29]  Xiaoling Chen,et al.  Dynamics of the wetland vegetation in large lakes of the Yangtze Plain in response to both fertilizer consumption and climatic changes , 2018, ISPRS Journal of Photogrammetry and Remote Sensing.

[30]  Martin Brandt,et al.  From woody cover to woody canopies: How Sentinel-1 and Sentinel-2 data advance the mapping of woody plants in savannas , 2019 .

[31]  Meng Zhang,et al.  Mapping paddy rice by the object-based random forest method using time series Sentinel-1/Sentinel-2 data , 2019 .

[32]  Brian Brisco,et al.  An efficient feature optimization for wetland mapping by synergistic use of SAR intensity, interferometry, and polarimetry data , 2018, Int. J. Appl. Earth Obs. Geoinformation.

[33]  Hongsheng Zhang,et al.  Exploring the optimal integration levels between SAR and optical data for better urban land cover mapping in the Pearl River Delta , 2018, Int. J. Appl. Earth Obs. Geoinformation.

[34]  F. Javier García-Haro,et al.  A comparison of STARFM and an unmixing-based algorithm for Landsat and MODIS data fusion , 2015 .

[35]  Fouad Khelifi,et al.  Automatic classification of colorectal and prostatic histologic tumor images using multiscale multispectral local binary pattern texture features and stacked generalization , 2018, Neurocomputing.

[36]  Huihui Song,et al.  A robust adaptive spatial and temporal image fusion model for complex land surface changes , 2018 .

[37]  Onisimo Mutanga,et al.  Multi-season RapidEye imagery improves the classification of wetland and dryland communities in a subtropical coastal region , 2019, ISPRS Journal of Photogrammetry and Remote Sensing.

[38]  G. Krieger,et al.  The global forest/non-forest map from TanDEM-X interferometric SAR data , 2018 .

[39]  Chaoyang Wu,et al.  Estimating chlorophyll content from hyperspectral vegetation indices : Modeling and validation , 2008 .

[40]  S. Kanmani,et al.  Optimization of stacking ensemble configurations through Artificial Bee Colony algorithm , 2013, Swarm Evol. Comput..

[41]  Grzegorz Ziomek,et al.  Random search optimization approach for highly multi-modal nonlinear problems , 2005, Adv. Eng. Softw..

[42]  Michael A. Wulder,et al.  Discriminating treed and non-treed wetlands in boreal ecosystems using time series Sentinel-1 data , 2020, Int. J. Appl. Earth Obs. Geoinformation.

[43]  Hui Lin,et al.  A new scheme for urban impervious surface classification from SAR images , 2018 .

[44]  Peter M. Atkinson,et al.  Spatio-temporal fusion for daily Sentinel-2 images , 2018 .

[45]  Ugur Avdan,et al.  Evaluating the utilization of the red edge and radar bands from sentinel sensors for wetland classification , 2019, CATENA.

[46]  Shoba Periasamy,et al.  Significance of dual polarimetric synthetic aperture radar in biomass retrieval: An attempt on Sentinel-1 , 2018, Remote Sensing of Environment.

[47]  Weimin Huang,et al.  A dynamic classification scheme for mapping spectrally similar classes: Application to wetland classification , 2019, Int. J. Appl. Earth Obs. Geoinformation.

[48]  Hui Lin,et al.  Mapping Paddy Rice Using a Convolutional Neural Network (CNN) with Landsat 8 Datasets in the Dongting Lake Area, China , 2018, Remote. Sens..

[49]  P. Gong,et al.  Object-based analysis and change detection of major wetland cover types and their classification uncertainty during the low water period at Poyang Lake, China , 2011 .

[50]  Swades Pal,et al.  Wetland delineation simulation and prediction in deltaic landscape , 2020 .

[51]  Martin Brandt,et al.  Trends of land surface phenology derived from passive microwave and optical remote sensing systems and associated drivers across the dry tropics 1992–2012 , 2019, Remote Sensing of Environment.

[52]  Dirk Tiede,et al.  ESP: a tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data , 2010, Int. J. Geogr. Inf. Sci..

[53]  Dongmei Chen,et al.  Segmentation for Object-Based Image Analysis (OBIA): A review of algorithms and challenges from remote sensing perspective , 2019, ISPRS Journal of Photogrammetry and Remote Sensing.

[54]  Brigitte Leblon,et al.  Mapping wetlands in Nova Scotia with multi-beam RADARSAT-2 Polarimetric SAR, optical satellite imagery, and Lidar data , 2018, International Journal of Applied Earth Observation and Geoinformation.

[55]  Graeme Newell,et al.  The utility of Random Forests for wildfire severity mapping , 2018, Remote Sensing of Environment.

[56]  J. Benediktsson,et al.  Nomination-favoured opinion pool for optical-SAR-synergistic rice mapping in face of weakened flooding signals , 2019, ISPRS Journal of Photogrammetry and Remote Sensing.

[57]  Yun Zhang,et al.  Deep Convolutional Neural Network for Complex Wetland Classification Using Optical Remote Sensing Imagery , 2018, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.