Research on Machine Learning Described by Researchers at China University of Geo sciences (Crop Phenology Estimation in Rice Fields Using Sentinel-1 GRD SAR Data and Machine Learning- Aided Particle Filtering Approach)
Emerging Technologies;
China University of Geosciences;
Asia;
Wuhan;
People's Republic of China;
Machine Learni ng;
Cyborgs;
期刊名称：
Robotics & Machine Learning Daily News
i s s n：
年卷期：
2024 年
May 29 期
页 码：
16-17
页 码：
摘 要：
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Fresh data on artificial intelligence are presented in a new report. According to news originating from Wuhan, People' s Republic of China, by NewsRx editors, the research stated, "Monitoring crop ph enology is essential for managing field disasters, protecting the environment, a nd making decisions about agricultural productivity. Because of its high timelin ess, high resolution, great penetration, and sensitivity to specific structural elements, synthetic aperture radar (SAR) is a valuable technique for crop phenol ogy estimation." The news reporters obtained a quote from the research from China University of G eosciences: "Particle filtering (PF) belongs to the family of dynamical approach and has the ability to predict crop phenology with SAR data in real time. The o bservation equation is a key factor affecting the accuracy of particle filtering estimation and depends on fitting. Compared to the common polynomial fitting (P OLY), machine learning methods can automatically learn features and handle compl ex data structures, offering greater flexibility and generalization capabilities . Therefore, incorporating two ensemble learning algorithms consisting of suppor t vector machine regression (SVR), random forest regression (RFR), respectively, we proposed two machine learning-aided particle filtering approaches (PF-SVR, P F-RFR) to estimate crop phenology. One year of time-series Sentinel-1 GRD SAR da ta in 2017 covering rice fields in Sevilla region in Spain was used for establis hing the observation and prediction equations, and the other year of data in 201 8 was used for validating the prediction accuracy of PF methods. Four polarizati on features (VV, VH, VH/VV and Radar Vegetation Index (RVI)) were exploited as t he observations in modeling. Experimental results reveals that the machine learn ing-aided methods are superior than the PF-POLY method. The PF-SVR exhibited bet ter performance than the PF-RFR and PF-POLY methods. The optimal outcome from PF -SVR yielded a root-mean-square error (RMSE) of 7.79, compared to 7.94 for PF-RF R and 9.1 for PF-POLY.
分享
载入中，请稍后...
