Classification of SAR data. Primarily based on 25 Sentinel-1 pictures, they carried out crop classification in Camage, France. The experimental final results showed that LSTM and GRU classifiers had been drastically superior than the classical techniques [41]. Wang et.al combined 11 Sentinel-2 pictures and 23 Sentinel-1 GRD images covering the Tongxiang County of China’s Zhejiang Province then place them to the developed LSTM classifier to get a paddy rice map [60]. The overall accuracy was as much as 0.937. Filho et al. used 60 scenes of Sentinel-1 VH information from 2017 to 2018 and BiLSTM to classify rice in Rio Grande do Sul state of Brazil [39]. The results of your BiLSTM model were greater than the LSTM model. RNNs have achieved some success in the field of rice extraction, but these models give the same weight for the time dimension options with distinctive value inside the classification decision-making method, which impacts the final classification accuracy. We added the focus model to the BiLSTM model, which could totally mine the favorable time series details, gave unique weights to many time dimension options inside the classification decision-making procedure, and strengthened the separability of rice and non-rice, so as to enhance the classification overall performance of the model. Inside the absence of a big amount of prior expertise of rice, there will inevitably be some misclassification inside the original classification results, so the original classification results must be post-processed. Lots of researchers employed post-processing strategies to optimize the classification benefits [36,613]. Therefore, we employed FROM-GLC10 for the post-processing of rice Sarizotan Autophagy extraction benefits, which reduced the false alarm to a specific extent. Regardless of whether compared with other procedures or with statistical data, our proposed system has accomplished fantastic results, which shows that our system has certain sensible value in the extraction of tropical and subtropical rice. Nevertheless, you’ll find nevertheless some deficiencies in the current research outcomes. In mountainous regions, the mountains and shadows in SAR photos trigger the omission of rice. Secondly, the riverside vegetation has comparable temporal qualities with rice, which leads to false alarm in rice extraction outcomes. Within the future, we’ll add some negative sample education to additional strengthen the performance with the process. five. Conclusions Based on the application needs of tropical and subtropical rice monitoring, this study proposed a set of rice extraction and mapping frameworks, like rice sample making process applying time qualities, rice classification technique primarily based on BiLSTMAttention model, and post-processing approach based on high-precision global land cover. Applying 66 scenes of Sentinel-1 data in 2019 and the proposed framework, rice mapping wasAgriculture 2021, 11,18 ofcarried out in Zhanjiang City, China. Experimental benefits show that the time series feature combination strategy of time series maximum, time series minimum, and typical can intuitively reflect the distribution of rice and increase the production efficiency of samples. The accuracy of rice area extraction by the proposed technique is 0.9351, that is superior than BiLSTM and RF techniques, as well as the extracted plots preserve very good integrity. In the coming years, we will carry out large-scale rice mapping research primarily based on multitemporal SAR information, further improve the classification accuracy, and market rice yield estimation based on yield estimation models, so as to provide.