Study on crop classification in Sanjiang Plain based on random forest
In order to monitor and manage farmland effectively in Sanjiang Plain,an efficient method for crop classification was proposed in this paper.Sentinel-2 satellite images in Google Earth Engine were used,and the ground objects in Sanjiang Plain were classified based on random forest classification model with different classification features.On this basis,the accuracy of the classification model was evaluated,and the model with the best classification features was selected to identify and extract the rice,maize and soybean planting areas in Sanjiang Plain.It was found that the total planting areas of rice and soybean extracted based on this method were highly consistent with the data of the statistical yearbook,and the maize planting area extracted based on this method was 0.83 times of the actual sown area.The results show that the proposed method can effectively extract the planting areas of different crops from Sentinel-2 remote sensing images.
NETL
NSTL
万方数据
