Rice identification and change detection using TerraSAR-X data

Pei, Zhiyuan; Zhang, Songling; Guo, Lin; McNairn, Heather; Shang, Jiali; Jiao, Xianfeng

doi:10.5589/m11-025

Cited by 11 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different SAR sensors have been used to map rice-growing regions. Considering the X-band (8-12 GHz) radar, several types of research used the sensors: TerraSAR-X [31][32][33][34] and COnstellation of small Satellites for the Mediterranean basin Observation (COSMO)-SkyMed [33,[35][36][37][38]. Among the C-Band radar sensors (4-8 GHz) used to map the rice crops were: European Remote Sensing Satellite (ERS) [12,[39][40][41], Radarsat [42][43][44][45][46][47], and Advanced Synthetic Aperture Radar (ASAR) [48][49][50][51][52].…”

Section: Introductionmentioning

confidence: 99%

Rice Crop Detection Using LSTM, Bi-LSTM, and Machine Learning Models from Sentinel-1 Time Series

et al. 2020

View full text Add to dashboard Cite

The Synthetic Aperture Radar (SAR) time series allows describing the rice phenological cycle by the backscattering time signature. Therefore, the advent of the Copernicus Sentinel-1 program expands studies of radar data (C-band) for rice monitoring at regional scales, due to the high temporal resolution and free data distribution. Recurrent Neural Network (RNN) model has reached state-of-the-art in the pattern recognition of time-sequenced data, obtaining a significant advantage at crop classification on the remote sensing images. One of the most used approaches in the RNN model is the Long Short-Term Memory (LSTM) model and its improvements, such as Bidirectional LSTM (Bi-LSTM). Bi-LSTM models are more effective as their output depends on the previous and the next segment, in contrast to the unidirectional LSTM models. The present research aims to map rice crops from Sentinel-1 time series (band C) using LSTM and Bi-LSTM models in West Rio Grande do Sul (Brazil). We compared the results with traditional Machine Learning techniques: Support Vector Machines (SVM), Random Forest (RF), k-Nearest Neighbors (k-NN), and Normal Bayes (NB). The developed methodology can be subdivided into the following steps: (a) acquisition of the Sentinel time series over two years; (b) data pre-processing and minimizing noise from 3D spatial-temporal filters and smoothing with Savitzky-Golay filter; (c) time series classification procedures; (d) accuracy analysis and comparison among the methods. The results show high overall accuracy and Kappa (>97% for all methods and metrics). Bi-LSTM was the best model, presenting statistical differences in the McNemar test with a significance of 0.05. However, LSTM and Traditional Machine Learning models also achieved high accuracy values. The study establishes an adequate methodology for mapping the rice crops in West Rio Grande do Sul.

show abstract

Section: Introductionmentioning

confidence: 99%

Rice Crop Detection Using LSTM, Bi-LSTM, and Machine Learning Models from Sentinel-1 Time Series

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Many studies have proposed methods which help to extract rice regions by using multi-temporal SAR images (Nguyen Ba Duy et al, 2015;Yuan et al, 2009;Zhiyuan et al, 2011). Figure 12 shows the morphology of water rice crop with eight main stages.…”

Section: Double-cropped Rice Phenology By Multitemporal Sar Imagesmentioning

confidence: 99%

Mapping land cover using multi-temporal sentinel-1a data: A case study in Hanoi

Hang¹,

Vu²,

Duong³

et al. 2017

TCCKHVTD

View full text Add to dashboard Cite

Land cover mapping is one of the most important applications of both optical and microwave remote sensing. The optical remote sensing recognizes land cover objects using spectral reflectance of the material constituting the land cover. The microwave remote sensing recognizes ground objects using backscatter, of which the intensity depends on the roughness of the ground's surface. Therefore, the multi temporal SAR images owning a lot of phenology information of land cover are the potential ideal data source for land cover mapping, in particular in the urban area. In this article, the authors present a new approach to the classification of land cover by using multi-temporal Sentinel-1A data. The experience data are single-pole (VV) in Interferometric Wide Swath mode (IW) collected from December 2014 to October 2015 along descending orbit over Hanoi, Vietnam. Decision tree method is applied base on analyzing threshold of standard deviation, mean backscatter value of land cover patterns, and combining double-crop rice classification image. The double-crop rice image is classified by rice phenology using multi-temporal Sentinel-1A images. The threshold in decision tree method is analyzed by field surveying data. The resulting classified image has been assessed using the test points in high-resolution images of Google Earth and field data. The accuracy of proposed method achieved 84.7%.Optical satellite imagery plays an important part in mapping land cover. Recognition of the land cover is based on spectral reflectance characteristics of land cover categories (Abdalla and Abdulaziz, 2012;Nguyen Dinh Duong et al., 2014;Li et al., 2004) or NDVI time series (Lambin et al., 1999;Myneni et al., 1995). However, optical imagery has many disadvantages due to weather condition and cloudiness. This is apparent

show abstract

“…Active microwave sensors are much less affected by cloud cover and due to their all-weather, day and night imaging capabilities and have been used since the 1990s to map rice areas. Studies using time series data from C-and X-band sensors have shown their potential for rice mapping, the European Remote Sensing satellites (ERS) 1 and 2 (Aschbacher et al, 1995;Kurosu et al, 1995;Patel et al, 1995;Chakraborty et al, 1997;Le Toan et al, 1997;Panigrahy et al, 1997;Liew et al, 1998b;McNairn & Brisco, 2004;Diuk-Wasser et al, 2006), Radarsat (Liew et al, 1998a;Panigrahy et al, 1999;Ribbes, 1999;Shao et al, 2001;Li et al, 2003;Choudhury & Chakraborty, 2006;Yonezawa et al, 2012;Yang et al, 2016;Zhang et al, 2016), Envisat ASAR (Advanced Synthetic Aperture Radar) (Bouvet et al, 2009;Bouvet & Le Toan, 2011;Karila et al, 2014;Nguyen et al, 2015), TerraSAR-X (Lopez-Sanchez et al, 2011;Pei et al, 2011;Asilo et al, 2014;Nelson et al, 2014), COSMO-SkyMed (Asilo et al, 2014;Nelson et al, 2014;Corcione et al, 2016;Busetto et al, 2017;Boschetti et al, 2017;Phan et al, 2018) and more recently Sentinel-1 (Clauss et al, 2017;Torbick et al, 2017;Onojeghuo et al, 2018;Nguyen et al, 2016;Son et al, 2017). These studies focussed on mapping r...…”

Section: Introductionmentioning

confidence: 99%

Estimating rice production in the Mekong Delta, Vietnam, utilizing time series of Sentinel-1 SAR data

Clauss

Ottinger

Leinenkugel

et al. 2018

International Journal of Applied Earth Observation and Geoinfor

View full text Add to dashboard Cite

Rice is the most important food crop in Asia and rice exports can significantly contribute to a country's GDP. Vietnam is the third largest exporter and fifth largest producer of rice, the majority of which is grown in the Mekong Delta. The cultivation of rice plants is important, not only in the context of food security, but also contributes to greenhouse gas emissions, provides man-made wetlands as an ecosystem, sustains smallholders in Asia and influences water resource planning and runoff water management. Rice growth can be monitored with Synthethic Aperture Radar (SAR) time series due to the agronomic flooding followed by rapid biomass increase affecting the backscatter signal. With the advent of Sentinel-1 a wealth of free and open SAR data is available to monitor rice on regional or larger scales and limited data availability should not be an issue from 2015 onwards. We used Sentinel-1 SAR time series to estimate rice production in the Mekong Delta, Vietnam, for three rice seasons centered on the year 2015. Rice production for each growing season was estimated by first classifying paddy rice area using superpixel segmentation and a phenology based decision tree, followed by yield estimation using random forest regression models trained on in-situ yield data collected by surveying 357 rice farms. The estimated rice production for the three rice growing seasons 2015 correlates well with data at the district level collected from the province statistics offices with R 2 s of 0.93 for the Winter-Spring, 0.86 for the Summer-Autumn and 0.87 for the Autumn-Winter season.

show abstract

Rice identification and change detection using TerraSAR-X data

Cited by 11 publications

References 18 publications

Rice Crop Detection Using LSTM, Bi-LSTM, and Machine Learning Models from Sentinel-1 Time Series

Rice Crop Detection Using LSTM, Bi-LSTM, and Machine Learning Models from Sentinel-1 Time Series

Mapping land cover using multi-temporal sentinel-1a data: A case study in Hanoi

Estimating rice production in the Mekong Delta, Vietnam, utilizing time series of Sentinel-1 SAR data

Contact Info

Product

Resources

About