Synthetic aperture radar (SAR) data-driven approach for rice mapping and monitoring | |
| Author | Pathakorn Usaha |
| Call Number | AIT Thesis no.RS-23-11 |
| Subject(s) | Rice--Thailand--Classification Synthetic aperture radar |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Remote Sensing and Geographic Information Systems |
| Publisher | Asian Institute of Technology |
| Abstract | Rice plays a vital role as the most important food crop in Thailand. A significant portion of rice cultivation takes place during the rainy season, which is in the country's monsoon period. However, cloud cover during this season poses a significant challenge for optical remote sensing. Synthetic aperture radar (SAR) provides a viable altermative for rice crop monitoring due to its ability to operate effectively in various weather conditions and valuable tool for rice crop analysis. The study presents Multi temporal of Sentinel-1A data in VV and VH polarizations were used in 2020-2022 for rice area classification and 2016-2021 for rice yield prediction to predict rice yield in 2022 by using Machine learning model like Random Forest classification and Random Forest regression, respectively. The integration of multi-temporal Sentinel-1A SAR data. The classification of rice cultivated areas achieved impressive overall accuracy rates. VH polarization demonstrated the highest accuracy of 87%, while VV polarization achieved the highest accuracy of 82%. This can show that VH polarization has better performance than VV Polarization. While the highest accuracy of rice yield prediction using SAR time series data in this study area is VH polarization, the result demonstrates that VH polarization achieved a RMSE of 0.1043 ha-1 and a MAE of 0.076 t ha-1, while the vV polarization achieved an RMSE of 0.1163 t ha-1 and an MAE of 0.089 t ha-1 Therefore,the effective combination of SAR data proves to be highly effective in precisely identifying and mapping rice cultivation areas. Additionally, the findings highlight the capability of SAR time series data for accurately predicting rice yields. The utilization of SAR data empowers farmers and stakeholders to make informed decisions regarding crop management, leading to enhanced agricultural productivity and improved outcomes. |
| Year | 2023 |
| Type | Thesis |
| School | School of Engineering and Technology |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Remote Sensing and Geographic Information Systems (RS) |
| Chairperson(s) | Virdis, Salvatore G.P.;Noppadon Khiripet (Co-chairperson) |
| Examination Committee(s) | Sarawut Ninsawat;Mozumder, Chitrini |
| Scholarship Donor(s) | Royal Thai Government Fellowships |
| Degree | Thesis (M. Sc.) - Asian Institute of Technology, 2023 |