多光谱遥感在长江口水深探测中的应用

Application of multi-spectral imagery to water depth extraction in the Changjiang River Estuary

利用L andsat-7 ETM+遥感图像反射率和实测水深值之间的相关性,建立了单波段模型、双波段模型、比值模型和多波段模型等4种线性回归模型,以及动量BP人工神经网络水深反演模型,对长江口南港航道水深进行了反演,对比分析了不同方法在长江口水深反演计算中的优劣性,试验表明,神经网络反演模型标准误差最小,精度最高。

The single, double, ratio and multi-band bathymetry models and momentum BP neural network model （MBPNNM） are constructed to retrieve the water depth information for the South Channel of the Changjiang River Estuary using the relationship between reflectance derived from Landsat ETM ＋ satellite data and water depth information, and the evaluation of the different arithmetic is given. The results indicated that the standard error of the MBPNNM was the least and the accuracy was the best.