用TM影像进行湖泊水色反演研究的人工神经网络模型

利用人工神经网络技术进行了湖泊水色遥感的反演研究,在同步实验的基础上了构造了包含一个隐含层的BP神经网络模型,利用TM卫星影像反演悬浮物、COD_Mn、溶解氧、总磷、总氮和叶绿素浓度反演精度较高,相对误差基本在25%以下,同时分析了该人工神经网络反演模型的误差来源,改进措施以及应用前景.研究表明,在进行小规模的同步监测的基础上,此模型可用于湖泊水质调查、分析和评价.

An Artificial Neural Network Model for Lake Color Inversion Using TM Imagery

The technology of artificial neural network was used for inversing water quality parameters from TM imagery data in the paper in order to study water quality and eutrophic status of lake. On the basis of satellite synchronous monitoring experiment, a BP neural network model was constructed, in which concentrations of SS, COD_Mn, DO, T-N, T-P and chlo a were inversed from Landsat TM data and the accuracy of which was good, the relative error of which could be controlled below 25%. Moreover, the reasons of simulating error, ways of improving model and applications of the model were also analyzed in detail. The results of this research told that based on a small scale of satellite synchronous experiment, the model could be applied successfully in investigation, analysis and estimation of lake water quality.