Mapping Paddy Rice Fields by Combining Multi-Temporal Vegetation Index and Synthetic Aperture Radar Remote Sensing Data Using Google Earth Engine Machine Learning Platform

The knowledge of the area and spatial distribution of paddy rice fields is important for water resource management. However, accurate map of paddy rice is a long-term challenge because of its spatiotemporal discontinuity and short duration. To solve this problem, this study proposed a paddy rice area extraction approach by using the combination of optical vegetation indices and synthetic aperture radar (SAR) data. This method is designed to overcome the data-missing problem due to cloud contamination and spatiotemporal discontinuities of the traditional optical remote sensing method. More specifically, the Sentinel-1A SAR and the Sentinel-2 multispectral imager (MSI) Level-2A imagery are used to identify paddy rice with a high temporal and spatial resolution. Three vegetation indices, namely normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), and land surface water index (LSWI), are estimated from optical bands. Two polarization bands (VH (vertical-horizontal) and VV (vertical-vertical)) are used to overcome the cloud contamination problem. This approach was applied with the random forest machine learning algorithm on the Google Earth Engine platform for the Jianghan Plain in China as an experimental area. The results of 39 experiments uncovered the effect of different factors. The results indicated that the combination of VV and VH band showed a better performance compared with other polarization bands; the average producer’s accuracy of paddy rice (PA) is 72.79%, 1.58% higher than the second one VH. Secondly, the combination of three indices also showed a better result than others, with average PA 73.82%, 1.42% higher than using NDVI alone. The classification result presented the best combination is EVI, VV, and VH polarization band. The producer’s accuracy of paddy rice was 76.67%, with the overall accuracy (OA) of 66.07%, and Kappa statistics of 0.45. However, NDVI, EVI, and VH showed better performance in mapping the morphology. The results demonstrated the method developed in this study can be successfully applied to the cloud-prone area for mapping paddy rice to overcome the data missing caused by cloud and rain during the paddy growing season.

Download Full-text

Assessing the Effect of Training Sampling Design on the Performance of Machine Learning Classifiers for Land Cover Mapping Using Multi-Temporal Remote Sensing Data and Google Earth Engine

Remote Sensing ◽

10.3390/rs13081433 ◽

2021 ◽

Vol 13 (8) ◽

pp. 1433

Author(s):

Shobitha Shetty ◽

Prasun Kumar Gupta ◽

Mariana Belgiu ◽

S. K. Srivastav

Keyword(s):

Machine Learning ◽

Remote Sensing ◽

Random Sampling ◽

Sampling Design ◽

Remote Sensing Data ◽

Google Earth ◽

Machine Learning Classifiers ◽

Learning Classifiers ◽

Multi Temporal ◽

Google Earth Engine

Machine learning classifiers are being increasingly used nowadays for Land Use and Land Cover (LULC) mapping from remote sensing images. However, arriving at the right choice of classifier requires understanding the main factors influencing their performance. The present study investigated firstly the effect of training sampling design on the classification results obtained by Random Forest (RF) classifier and, secondly, it compared its performance with other machine learning classifiers for LULC mapping using multi-temporal satellite remote sensing data and the Google Earth Engine (GEE) platform. We evaluated the impact of three sampling methods, namely Stratified Equal Random Sampling (SRS(Eq)), Stratified Proportional Random Sampling (SRS(Prop)), and Stratified Systematic Sampling (SSS) upon the classification results obtained by the RF trained LULC model. Our results showed that the SRS(Prop) method favors major classes while achieving good overall accuracy. The SRS(Eq) method provides good class-level accuracies, even for minority classes, whereas the SSS method performs well for areas with large intra-class variability. Toward evaluating the performance of machine learning classifiers, RF outperformed Classification and Regression Trees (CART), Support Vector Machine (SVM), and Relevance Vector Machine (RVM) with a >95% confidence level. The performance of CART and SVM classifiers were found to be similar. RVM achieved good classification results with a limited number of training samples.

Download Full-text

Machine Learning Comparison and Parameter Setting Methods for the Detection of Dump Sites for Construction and Demolition Waste Using the Google Earth Engine

Remote Sensing ◽

10.3390/rs13040787 ◽

2021 ◽

Vol 13 (4) ◽

pp. 787

Author(s):

Lei Zhou ◽

Ting Luo ◽

Mingyi Du ◽

Qiang Chen ◽

Yang Liu ◽

...

Keyword(s):

Machine Learning ◽

Remote Sensing ◽

Google Earth ◽

Construction And Demolition Waste ◽

Parameterization Scheme ◽

Classification Methods ◽

Demolition Waste ◽

Optimal Method ◽

Identification Method ◽

Google Earth Engine

Machine learning has been successfully used for object recognition within images. Due to the complexity of the spectrum and texture of construction and demolition waste (C&DW), it is difficult to construct an automatic identification method for C&DW based on machine learning and remote sensing data sources. Machine learning includes many types of algorithms; however, different algorithms and parameters have different identification effects on C&DW. Exploring the optimal method for automatic remote sensing identification of C&DW is an important approach for the intelligent supervision of C&DW. This study investigates the megacity of Beijing, which is facing high risk of C&DW pollution. To improve the classification accuracy of C&DW, buildings, vegetation, water, and crops were selected as comparative training samples based on the Google Earth Engine (GEE), and Sentinel-2 was used as the data source. Three classification methods of typical machine learning algorithms (classification and regression trees (CART), random forest (RF), and support vector machine (SVM)) were selected to classify the C&DW from remote sensing images. Using empirical methods, the experimental trial method, and the grid search method, the optimal parameterization scheme of the three classification methods was studied to determine the optimal method of remote sensing identification of C&DW based on machine learning. Through accuracy evaluation and ground verification, the overall recognition accuracies of CART, RF, and SVM for C&DW were 73.12%, 98.05%, and 85.62%, respectively, under the optimal parameterization scheme determined in this study. Among these algorithms, RF was a better C&DW identification method than were CART and SVM when the number of decision trees was 50. This study explores the robust machine learning method for automatic remote sensing identification of C&DW and provides a scientific basis for intelligent supervision and resource utilization of C&DW.

Download Full-text

Assessment of the State of the Territories of Closed Waste Disposal Facilities by Vegetation Indices

Ecology and Industry of Russia ◽

10.18412/1816-0395-2021-9-30-37 ◽

2021 ◽

Vol 25 (9) ◽

pp. 30-37

Author(s):

N.N. Sliusar ◽

A.P. Belousova ◽

G.M. Batrakova ◽

R.D. Garifzyanov ◽

M. Huber-Humer ◽

...

Keyword(s):

Remote Sensing ◽

Waste Disposal ◽

Vegetation Cover ◽

Vegetation Index ◽

Vegetation Indices ◽

Cost Effective ◽

Google Earth ◽

The State ◽

Landsat 8 ◽

The Earth

The possibilities of using remote sensing of the Earth data to assess the formation of phytocenoses at reclaimed dumps and landfills are presented. The objects of study are landfills and dumps in the Perm Territory, which differed from each other in the types and timing of reclamation work. The state of the vegetation cover on the reclaimed and self-overgrowing objects was compared with the reference plots with naturally formed herbage of zonal meadow vegetation. The process of reclamation of the territory of closed landfills was assessed by the presence and homogeneity of the vegetation layer and by the values of the vegetation index NDVI. To identify the dynamics of changes in the vegetation cover, we used multi-temporal satellite images from the open resources of Google Earth and images in the visible and infrared ranges of the Landsat-5/TM and Landsat-8/OLI satellites. It is shown that the data of remote sensing of the Earth, in particular the analysis of vegetation indices, can be used to assess the dynamics of overgrowing of territories of reclaimed waste disposal facilities, as well as an additional and cost-effective method for monitoring the restoration of previously disturbed territories.

Download Full-text

Synthetic sampling for spatio-temporal land cover mapping with machine learning and the Google Earth Engine in Andalusia, Spain

10.5194/egusphere-egu2020-1153 ◽

2020 ◽

Author(s):

Laura Bindereif ◽

Tobias Rentschler ◽

Martin Batelheim ◽

Marta Díaz-Zorita Bonilla ◽

Philipp Gries ◽

...

Keyword(s):

Machine Learning ◽

Remote Sensing ◽

Land Cover ◽

Classification Accuracy ◽

Sampling Methods ◽

Google Earth ◽

Land Cover Mapping ◽

Machine Learning Applications ◽

Google Earth Engine ◽

Spatio Temporal

<p>Land cover information plays an essential role for resource development, environmental monitoring and protection. Amongst other natural resources, soils and soil properties are strongly affected by land cover and land cover change, which can lead to soil degradation. Remote sensing techniques are very suitable for spatio-temporal mapping of land cover mapping and change detection. With remote sensing programs vast data archives were established. Machine learning applications provide appropriate algorithms to analyse such amounts of data efficiently and with accurate results. However, machine learning methods require specific sampling techniques and are usually made for balanced datasets with an even training sample frequency. Though, most real-world datasets are imbalanced and methods to reduce the imbalance of datasets with synthetic sampling are required. Synthetic sampling methods increase the number of samples in the minority class and/or decrease the number in the majority class to achieve higher model accuracy. The Synthetic Minority Over-Sampling Technique (SMOTE) is a method to generate synthetic samples and balance the dataset used in many machine learning applications. In the middle Guadalquivir basin, Andalusia, Spain, we used random forests with Landsat images from 1984 to 2018 as covariates to map the land cover change with the Google Earth Engine. The sampling design was based on stratified random sampling according to the CORINE land cover classification of 2012. The land cover classes in our study were arable land, permanent crops (plantations), pastures/grassland, forest and shrub. Artificial surfaces and water bodies were excluded from modelling. However, the number of the 130 training samples was imbalanced. The classes pasture (7&#160;samples) and shrub (13&#160;samples) show a lower number than the other classes (48, 47 and 16&#160;samples). This led to misclassifications and negatively affected the classification accuracy. Therefore, we applied SMOTE to increase the number of samples and the classification accuracy of the model. Preliminary results are promising and show an increase of the classification accuracy, especially the accuracy of the previously underrepresented classes pasture and shrub. This corresponds to the results of studies with other objectives which also see the use of synthetic sampling methods as an improvement for the performance of classification frameworks.</p>

Download Full-text

The NDVI algorithm utilization on the google earth engine platform to monitor changes in forest density in mining area

IOP Conference Series Earth and Environmental Science ◽

10.1088/1755-1315/886/1/012100 ◽

2021 ◽

Vol 886 (1) ◽

pp. 012100

Author(s):

Munajat Nursaputra ◽

Siti Halimah Larekeng ◽

Nasri ◽

Andi Siady Hamzah

Keyword(s):

Remote Sensing ◽

Time Series ◽

Large Scale ◽

Vegetation Index ◽

Forest Degradation ◽

Google Earth ◽

Aerial Images ◽

Forest Monitoring ◽

Forest Density ◽

Google Earth Engine

Abstract Periodic forest monitoring needs to be done to avoid forest degradation. In general, forest monitoring can be conducted manually (field surveys) or using technological innovations such as remote sensing data derived from aerial images (drone results) or cloud computing-based image processing. Currently, remote sensing technology provides large-scale forest monitoring using multispectral sensors and various vegetation index processing algorithms. This study aimed to evaluate the use of the Google Earth Engine (GEE) platform, a geospatial dataset platform, in the Vale Indonesia mining concession area to improve accountable forest monitoring. This platform integrates a set of programming methods with a publicly accessible time-series database of satellite imaging services. The method used is NDVI processing on Landsat multispectral images in time series format, which allows for the description of changes in forest density levels over time. The results of this NDVI study conducted on the GEE platform have the potential to be used as a tool and additional supporting data for monitoring forest conditions and improvement in mining regions.

Download Full-text

Temporal Patterns in Illumination Conditions and Its Effect on Vegetation Indices Using Landsat on Google Earth Engine

Remote Sensing ◽

10.3390/rs12020211 ◽

2020 ◽

Vol 12 (2) ◽

pp. 211 ◽

Cited By ~ 2

Author(s):

Pablo Martín-Ortega ◽

Luis Gonzaga García-Montero ◽

Nicole Sibelet

Keyword(s):

Vegetation Index ◽

Normalized Difference Vegetation Index ◽

Vegetation Indices ◽

Temporal Patterns ◽

Google Earth ◽

Surface Reflectance ◽

Enhanced Vegetation Index ◽

Google Earth Engine ◽

Level 1 ◽

Irregular Topography

Vegetation indices (VI) describe vegetation structure and functioning but they are affected by illumination conditions (IC). Moreover, the fact that the effect of the IC on VI can be stronger than other biophysical or seasonal processes is under debate. Using Google Earth Engine and the latest Landsat Surface Reflectance level 1 data, we evaluated the temporal patterns of IC and two VI, the Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegetation Index (EVI) in a mountainous tropical forest during the years 1984–2017. We evaluated IC and VI at different times, their relationship with the topography and the correlations between them. We show that IC is useful for understanding the patterns of variation between VI and IC at the pixel level using Landsat sensors. Our findings confirmed a strong correlation between EVI and IC and less between NDVI and IC. We found a significant increase in IC, EVI, and NDVI throughout time due to an improvement in the position of all Landsat sensors. Our results reinforce the need to consider IC to interpret VI over long periods using Landsat data in order to increase the precision of monitoring VI in irregular topography.

Download Full-text

Downscaling TRMM Monthly Precipitation Using Google Earth Engine and Google Cloud Computing

Remote Sensing ◽

10.3390/rs12233860 ◽

2020 ◽

Vol 12 (23) ◽

pp. 3860

Author(s):

Abdelrazek Elnashar ◽

Hongwei Zeng ◽

Bingfang Wu ◽

Ning Zhang ◽

Fuyou Tian ◽

...

Keyword(s):

Machine Learning ◽

Vegetation Index ◽

Google Earth ◽

Upstream Region ◽

Support Vector ◽

Monthly Precipitation ◽

Spatiotemporal Resolution ◽

High Spatiotemporal Resolution ◽

Trmm Precipitation ◽

Google Earth Engine

Accurate precipitation data at high spatiotemporal resolution are critical for land and water management at the basin scale. We proposed a downscaling framework for Tropical Rainfall Measuring Mission (TRMM) precipitation products through integrating Google Earth Engine (GEE) and Google Colaboratory (Colab). Three machine learning methods, including Gradient Boosting Regressor (GBR), Support Vector Regressor (SVR), and Artificial Neural Network (ANN) were compared in the framework. Three vegetation indices (Normalized Difference Vegetation Index, NDVI; Enhanced Vegetation Index, EVI; Leaf Area Index, LAI), topography, and geolocation are selected as geospatial predictors to perform the downscaling. This framework can automatically optimize the models’ parameters, estimate features’ importance, and downscale the TRMM product to 1 km. The spatial downscaling of TRMM from 25 km to 1 km was achieved by using the relationships between annual precipitations and annually-averaged vegetation index. The monthly precipitation maps derived from the annual downscaled precipitation by disaggregation. According to validation in the Great Mekong upstream region, the ANN yielded the best performance when simulating the annual TRMM precipitation. The most sensitive vegetation index for downscaling TRMM was LAI, followed by EVI. Compared with existing downscaling methods, the proposed framework for downscaling TRMM can be performed online for any given region using a wide range of machine learning tools and environmental variables to generate a precipitation product with high spatiotemporal resolution.

Download Full-text

Synthetic Aperture Radar (L band) and Optical Vegetation Indices for Discriminating the Brazilian Savanna Physiognomies: A Comparative Analysis

Earth Interactions ◽

10.1175/ei117.1 ◽

2005 ◽

Vol 9 (15) ◽

pp. 1-15 ◽

Cited By ~ 24

Author(s):

Edson E. Sano ◽

Laerte G. Ferreira ◽

Alfredo R. Huete

Keyword(s):

Comparative Analysis ◽

Synthetic Aperture Radar ◽

Large Scale ◽

Vegetation Index ◽

Vegetation Indices ◽

Primary Response ◽

Synthetic Aperture ◽

L Band ◽

Vegetation Structures ◽

Aperture Radar

Abstract The all-weather capability, signal independence to the solar illumination angle, and response to 3D vegetation structures are the highlights of active radar systems for natural vegetation mapping and monitoring. However, they may present significant soil background effects. This study addresses a comparative analysis of the performance of L-band synthetic aperture radar (SAR) data and optical vegetation indices (VIs) for discriminating the Brazilian cerrado physiognomies. The study area was the Brasilia National Park, Brazil, one of the test sites of the Large-Scale Biosphere–Atmosphere (LBA) experiment in Amazonia. Seasonal Japanese Earth Resources Satellite-1 (JERS-1) SAR backscatter coefficients (σ°) were compared with two vegetation indices [normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI)] over the five most dominant cerrados’ physiognomies plus gallery forest. In contrast to the VIs, σ° from dry and wet seasons did not change significantly, indicating primary response to vegetation structures. Discriminant analysis and analysis of variance (ANOVA) showed an overall higher performance of radar data. However, when both SAR and VIs are combined, the discrimination capability increased significantly, indicating that the fusion of the optical and radar backscatter observations provides overall improved classifications of the cerrado types. In addition, VIs showed good performance for monitoring the cerrado dynamics.

Download Full-text

Tropical Peatland Identification using L-Band Full Polarimetric Synthetic Aperture Radar (SAR) Imagery (Study Case: Siak Regency, Riau Province)

HAYATI Journal of Biosciences ◽

10.4308/hjb.26.2.63 ◽

2019 ◽

Vol 26 (2) ◽

pp. 63

Author(s):

Desti Ayunda ◽

Ketut Wikantika ◽

Dandy A. Novresiandi ◽

Agung B. Harto ◽

Riantini Virtriana ◽

...

Keyword(s):

Remote Sensing ◽

Synthetic Aperture Radar ◽

Vegetation Index ◽

Synthetic Aperture ◽

Optical Remote Sensing ◽

Error Matrix ◽

Tropical Peatland ◽

Average Value ◽

Sar Imagery ◽

Aperture Radar

From previous research reported that tropical peatland is one of terrestrial carbon storage in Earth, and has contribution to climate change. Synthetic Aperture Radar (SAR) is one of remote sensing technology which is more efcient than optical remote sensing. Its ability to penetrate cloud makes it useful to monitor tropical environment. This research is conducted in a tropical peatland in Siak Regency, Riau Province. This research was conducted to identify tropical peatland in Siak Regency using polarimetric decomposition, unsupervised classifcation ISODATA, and Radar Vegetation Index (RVI) from SAR data that had been geometrically and radiometrically corrected. Polarimetric decomposition Freeman-Durden was performed to analyze radar backscattering mechanism in tropical peatland, which shows that volume and surface scattering was dominant because of the presence of vegetation and open area. Unsupervised classifcation ISODATA was then performed to extract “shrub class”. By assessing its accuracy, the class that represents shrub class in reference map was selected as the selected “shrub class”. RVI then was calculated using a certain formula. Spatial analysis was then conducted to acquire certain information that average value of RVI in tropical peatland tend to be higher than in non-tropical peatland. By integrating selected “shrub class” and RVI, peat classes were extracted. The best peat class was selected by comparing with peatland referenced map which is acquired from the Indonesian Agency for Agricultural Resources and Development (IAARD) using error matrix. In this research, the best peat class yielded 73.5 percent of Producer’s Accuracy (PA), 81.6 percent of User’s Accuracy (UA), 66.1 percent of Overall Accuracy (OA), and 0.1079 of Kappa coefcient (Ks).

Download Full-text

Strategies for landslide detection using open-access synthetic aperture radar backscatter change in Google Earth Engine

10.5194/nhess-2021-283 ◽

2021 ◽

Author(s):

Alexander L. Handwerger ◽

Shannan Y. Jones ◽

Pukar Amatya ◽

Hannah R. Kerner ◽

Dalia B. Kirschbaum ◽

...

Keyword(s):

Synthetic Aperture Radar ◽

Situational Awareness ◽

Remote Sensing Data ◽

Google Earth ◽

Disaster Mitigation ◽

Synthetic Aperture ◽

Landslide Event ◽

Combining Data ◽

Google Earth Engine ◽

Aperture Radar

Abstract. Rapid detection of landslides is critical for emergency response, disaster mitigation, and improving our understanding of landslide dynamics. Satellite-based synthetic aperture radar (SAR) can be used to detect landslides, often within days of a triggering event, because it penetrates clouds, operates day and night, and is regularly acquired worldwide. Here we present a SAR backscatter change detection approach that uses multi-temporal stacks of freely available data from the Copernicus Sentinel-1 satellites to detect areas with high landslide density using the cloud-based Google Earth Engine (GEE). Importantly, our approach does not require downloading a large volume of data to a local system or specialized processing software. We provide strategies, including a landslide density heatmap approach, that can aid in rapid response and landslide detection. We test our GEE-based approach on multiple recent rainfall- and earthquake-triggered landslide events. Our ability to detect surface change from landslides generally improves with the total number of SAR images acquired before and after a landslide event, by combining data from both ascending and descending satellite acquisition geometries, and applying topographic masks to remove flat areas unlikely to experience landslides. Importantly, our GEE approach allows the broader hazards and landslide community to utilize and advance these state-of-the-art remote sensing data for improved situational awareness of landslide hazards.

Download Full-text