论文部分内容阅读
采用遥感技术和BP神经网络技术,结合野外实测的盐渍土光谱特征和实验室化验的土壤含盐数据,对盐渍土盐分的遥感反演进行了模型的设计与编程实现。BP神经网络模型的预测精度在62.5%,明显高于传统统计模型的预测精度,表明BP神经网络能较好地模拟土壤含盐量与光谱数据之间的关系,可用于建立土壤盐分遥感反演模型。
Using remote sensing technology and BP neural network technology, combined with the spectral characteristics of saline soil measured in the field and laboratory soil salt data, the model of remote sensing retrieval of saline soil was designed and programmed. BP neural network prediction accuracy of 62.5%, significantly higher than the traditional statistical model prediction accuracy, indicating that BP neural network can better simulate the relationship between soil salinity and spectral data can be used to establish soil salinity remote sensing inversion model.