scholarly journals Fine Classification of Rice Paddy Based on RHSI-DT Method Using Multi-Temporal Compact Polarimetric SAR Data

2021 ◽  
Vol 13 (24) ◽  
pp. 5060
Author(s):  
Xianyu Guo ◽  
Junjun Yin ◽  
Kun Li ◽  
Jian Yang
Keyword(s):  
Rice Paddy ◽  
Surface Scattering ◽  
Support Vector ◽  
Growth Monitoring ◽  
Japonica Rice ◽  
Effective Parameters ◽  
Multi Temporal ◽  
Sar Data ◽  
Fine Classification

In recent years, the compact polarimetric (CP) synthetic aperture radar (SAR) has become a hotspot of SAR Earth observation. Meanwhile, CP SAR provides both relatively rich polarization information and large swath-width for rice mapping. Fine classification of rice paddy plays an important role in growth monitoring, pest prevention and yield estimation of rice. In this study, the multi-temporal CP SAR data were firstly simulated by fully polarimetric RADARSAT-2 data, and 22 CP parameters from each of the six temporal CP SAR data were extracted. Then we built a rice height-sensitive index (RHSI). Furthermore, a decision tree (DT) method was established by using the optimal CP parameters based on RHSI. Finally, the classification results of rice paddy based on DT and support vector machine (SVM) methods were compared. Results showed that the RHSI-DT method could obtain better results, with an overall accuracy of 97.94% and a kappa coefficient of 0.973, which was 2% higher and 0.03 larger than those of the SVM method. Besides, we found that the surface scattering of m-χ decomposition (m-χ_s (0627)) and ΔShannon entropy intensity Hi(Hi (1015)-Hi (0627)) were highly effective parameters to distinguish paddies of transplanting hybrid rice (T-H) and direct-sown japonica rice (D-J).

scholarly journals Classification of agricultural fields using time series of dual polarimetry TerraSAR-X images

2014 ◽  
Vol XL-2/W3 ◽  
pp. 191-196 ◽  
Author(s):  
S. Mirzaee ◽  
M. Motagh ◽  
H. Arefi ◽  
M. Nooryazdan
Keyword(s):  
Environmental Changes ◽  
Support Vector ◽  
Svm Classifier ◽  
Sensing Applications ◽  
Imaging Characteristics ◽  
Training Samples ◽  
Study Results ◽  
Phenological Changes ◽  
Multi Temporal

Due to its special imaging characteristics, Synthetic Aperture Radar (SAR) has become an important source of information for a variety of remote sensing applications dealing with environmental changes. SAR images contain information about both phase and intensity in different polarization modes, making them sensitive to geometrical structure and physical properties of the targets such as dielectric and plant water content. In this study we investigate multi temporal changes occurring to different crop types due to phenological changes using high-resolution TerraSAR-X imagers. The dataset includes 17 dual-polarimetry TSX data acquired from June 2012 to August 2013 in Lorestan province, Iran. Several features are extracted from polarized data and classified using support vector machine (SVM) classifier. Training samples and different features employed in classification are also assessed in the study. Results show a satisfactory accuracy for classification which is about 0.91 in kappa coefficient.

scholarly journals Irrigation Mapping Using Sentinel-1 Time Series at Field Scale

2018 ◽  
Vol 10 (9) ◽  
pp. 1495 ◽  
Author(s):  
Qi Gao ◽  
Mehrez Zribi ◽  
Maria Escorihuela ◽  
Nicolas Baghdadi ◽  
Pere Segui
Keyword(s):  
Time Series ◽  
Weather Conditions ◽  
Mean Value ◽  
Support Vector ◽  
Irrigated Crops ◽  
The Mean ◽  
Sar Data ◽  
Ground Truthing ◽  
Agricultural Site

The recently launched Sentinel-1 satellite with a Synthetic Aperture Radar (SAR) sensor onboard offers a powerful tool for irrigation monitoring under various weather conditions, with high spatial and temporal resolution. This research discusses the potential of different metrics calculated from the Sentinel-1 time series for mapping irrigated fields. A methodology for irrigation mapping using SAR data is proposed. The study is performed using VV (vertical–vertical) and VH (vertical–horizontal) polarizations over an agricultural site in Urgell, Catalunya (Spain). With field segmentation information from SIGPAC (the Geographic Information System for Agricultural Parcels), the backscatter intensities are averaged within each field. From the Sentinel-1 time series for each field, the statistics and metrics, including the mean value, the variance of the signal, the correlation length, and the fractal dimension, are analyzed. With the Support Vector Machine (SVM), the classification of irrigated crops, irrigated trees, and non-irrigated fields is performed with the metrics vector. The results derived from the SVM are validated with ground truthing from SIGPAC over the whole study area, with a good overall accuracy of 81.08%. Random Forest (RF) machine classification is also tested in this study, which gives an accuracy of around 82.2% when setting the tree depth at three. The methodology is based only on SAR data, which makes it applicable to all areas, even with frequent cloud cover, but this method may be less robust when irrigation is less dominated to soil moisture change.

crops planting area identification and analysis based on multi-source high resolution remote sensing data

2020 ◽  
Author(s):  
Lei Wang ◽  
Haoran Sun ◽  
Wenjun Li ◽  
Liang Zhou
Keyword(s):  
Remote Sensing ◽  
Random Forest ◽  
Classification Accuracy ◽  
Kappa Coefficient ◽  
Optical Data ◽  
Accurate Estimation ◽  
Support Vector ◽  
Random Forest Method ◽  
Multi Temporal ◽  
Sar Data

<p>Crop planting structure is of great significance to the quantitative management of agricultural water and the accurate estimation of crop yield. With the increasing spatial and temporal resolution of remote sensing optical and SAR(Synthetic Aperture Radar) images,  efficient crop mapping in large area becomes possible and the accuracy is improved. In this study, Qingyijiang Irrigation District in southwest of China is selected for crop identification methods comparison, which has heterogeneous terrain and complex crop structure . Multi-temporal optical (Sentinel-2) and SAR (Sentinel-1) data were used to calculate NDVI and backscattering coefficient as the main classification indexes. The multi-spectral and SAR data showed significant change in different stages of the whole crop growth period and varied with different crop types. Spatial distribution and texture analysis was also made. Classification using different combinations of indexes were performed using neural network, support vector machine and random forest method. The results showed that, the use of multi-temporal optical data and SAR data in the key growing periods of main crops can both provide satisfactory classification accuracy. The overall classification accuracy was greater than 82% and Kappa coefficient was greater than 0.8. SAR data has high accuracy and much potential in rice identification. However optical data had more accuracy in upland crops classification. In addition, the classification accuracy can be effectively improved by combination of classification indexes from optical and SAR data, the overall accuracy was up to 91.47%. The random forest method was superior to the other two methods in terms of the overall accuracy and the kappa coefficient.</p>

scholarly journals Feature Extraction and Classification of Multi-temporal SAR Data Using 3D Wavelet Transform

2013 ◽  
Vol 29 (5) ◽  
pp. 569-579 ◽  
Author(s):  
Hee Young Yoo ◽  
No-Wook Park ◽  
Sukyoung Hong ◽  
Kyungdo Lee ◽  
Yihyun Kim
Keyword(s):  
Feature Extraction ◽  
Wavelet Transform ◽  
3D Wavelet Transform ◽  
Multi Temporal ◽  
Sar Data

scholarly journals Study of Discrimination Between Plantation and Dense Scrub Based on Backscattering Behavior of C Band SAR Data

2014 ◽  
Vol XL-8 ◽  
pp. 755-760 ◽  
Author(s):  
M. Iyyappan ◽  
S. S. Ramakrishnan ◽  
K. Srinivasa Raju
Keyword(s):  
Classification Accuracy ◽  
Confusion Matrix ◽  
Training Sample ◽  
Support Vector ◽  
Dual Polarization ◽  
Envisat Asar ◽  
Band Combination ◽  
Training Samples ◽  
Multi Temporal ◽  
Sar Data

The study about on landuse and landcover classification using multi polarization and multi temporal C-band Synthetic Aperture Radar (SAR) data of recently launched multi-mode of RISAT-1 (Radar Imaging Satellite) by Indian Space Research Organization (ISRO) and European satellite, Envisat ASAR data. The backscattering coefficient were extracted for various land features from Cband SAR data. The training sample collecting from satellite optical imagery of study and field visit for verification. The training samples are used for the supervised classification technique of maximum Likelihood (ML) algorithms, Neural Network (NN) and Support Vector Machine (SVM) algorithms were applied for fourteen different polarizations combination of multi temporal and multiple polarizations. The previous study was carried only four band combination of RISAT 1 data, the continuation of work both SAR data were used in this study. The Classification results are verified with confusion matrix. The pixel based classification gives the good results in the dual polarization of CRS – HH and HV of RISAT −1 compared to dual polarization Envisat ASAR data. Meanwhile the quad Polarization combination of Envisat ASAR data got better classification accuracy. The SVM classifiers has given better classification results for all band combination followed by ML and NN. The Scrub are better identified in EnviSat ASAR – VV & VH Polarization and Plantation are better identified in EnviSat ASAR – HH, HH-HV & HV Polarization. The classification accuracy of both Scrub and Plantation is about 80 % in EnviSat ASAR – HH, VH & VV Polarization combination.

scholarly journals CLASSIFICATION OF MULTI-TEMPORAL SAR IMAGES FOR RICE CROPS USING COMBINED ENTROPY DECOMPOSITION AND SUPPORT VECTOR MACHINE TECHNIQUE

2007 ◽  
Vol 71 ◽  
pp. 19-39 ◽  
Author(s):  
Chue-Poh Tan ◽  
Jun-Yi Koay ◽  
Ka-Sing Lim ◽  
Hong-Tat Ewe ◽  
Hean-Teik Chuah
Keyword(s):  
Support Vector Machine ◽  
Support Vector ◽  
Sar Images ◽  
Multi Temporal
Sign in / Sign up

Export Citation Format

Share Document