Early rice,as one of the key crops for ensuring stable food production,holds significant importance.Accurate monitoring of the planting area and cultivation information of early rice is of crucial scientific significance for government departments to understand the early rice production sit-uation and formulate relevant work plans.This study utilized a phenology algorithm based on Senti-nel-2 satellite data,combined with auxiliary features and a random forest classifier,to accurately i-dentify and extract the planting area of early rice in Gao'an city in 2023.The results showed that the overall accuracy obtained through a confusion matrix accuracy assessment was 92.50%,confir-ming an outstanding classification performance of the method.Furthermore,compared to the data from the previous year,the validation accuracy of the early rice planting area in 2023 reached a high level of 93.44% .These two validation methods verify the robustness of the phenology algorithm based on Sentinel-2 satellite data.The research results provide reference for government depart-ments to gain in-depth insights into early rice production and formulate work plans.
维普
万方数据