Studies on Remote Sensing Dynamic Detection Model of Cropland Based on the Classification of Artificial Neural Network
基于ANN分类的农田遥感动态监测模型研究

To protect basic cropland is the precondition of the agricultural sustainable development in China,dynamically detecting the change of cropland on a spatio-temporal scale can help to make out the agricultural development plan and managing the agricultural economic development.This article discussed the classification method of the remote sensing image using the BP artificial neural network, and on the condition of the precision of the classification, it explored a remote sensing dynamic detection model comprised of single component detection and multicomponent detection,and described the algorithm about it in detail. Finally, an application of this model has been shown in an experimental area,and the result of the application has been analysed with the real data. It is shown that this model has excellently evaluated the amount and potential of the basic cropland in this experimental area, and opened out the change regular of the basic cropland in the experimental area on a spatio-temporal scale in a qualitative,quantitative and orientative way.