基于遗传算法优化BP神经网络的玉米遥感估产方法

Method of remote sensing estimation of corn yield based on genetic algorithm optimized BP neural network

以内蒙古自治区开鲁县玉米作物为研究对象,将生育期内玉米遥感影像所提取的多种植被指数和实地采样点的测产数据作为训练值,利用BP(back propagation)神经网络和遗传算法优化BP(GA-BP)神经网络估产模型,得出网络预测的玉米产量数值。通过决定系数R 2和均方根误差RMSE,比较实测产量与预测产量之间的精度,BP神经网络模型R^2为0.8452,RMSE(%)为28.37;遗传算法优化BP神经网络模型R^2为0.9850,RMSE(%)为6.70,表明遗传算法优化BP神经网络估产模型具有一定可行性和可信度。

Taking the corn crops in Kailu County of Inner Mongolia Autonomous Region as the research object,using various vegetation indices extracted from the corn remote sensing images and the measured data of the field sampling points as training values,the BP(back-propagation)neural network and genetic algorithm were used to optimize the BP(GA-BP)neural network estimation model to obtain the network predicted corn yield.The accuracy between the measured output and the predicted output is compared by the coefficient R^2 and the root mean square error RMSE.R^2 from BP neural network model is 0.8452,RMSE(%)is 28.37,whereas R^2 from genetic algorithm optimized BP neural network model is 0.9850,RMSE(%)is 6.70.It shows that genetic algorithm optimized BP neural network estimation model has certain feasibility and credibility.