一种改进的半分析模型的高光谱遥感水深反演方法

A hyperspectral water depth inversion method based on an improved semi-analytical model

利用高光谱遥感的浅海水深反演由于具有成本低、精度高等特点,近年来已经逐步发展成为一种探测水柱参数的可靠方法。半分析模型是一种应用广泛的高光谱遥感浅海水深反演模型,MILEBI (maximum likelihood estimation including environmental noise and bottom intra-class variability)方法为基于半分析模型的一种反演方法。鉴于水深增加时MILEBI会出现水深被高估的问题,本文提出一种改进的半分析模型的水深反演方法。本方法通过增设关于深度的先验分布,然后在MILEBI方法的损失函数中加入正则化项构成新的目标函数,使深度值对水面下反射率影响力增大,从而提高水深反演精度。本文运用了真实数据集和模拟数据集进行实验,实验结果表明:在15—20 m的深度范围内,MILEBI方法的平均绝对误差约为4.27 m,而改进后方法的平均绝对误差约为2.27 m,提高了反演精度。

The estimation of shallow water depth based on hyperspectral sensed data has been gradually matured as a reliable approach for detecting water column parameters with advantages of low cost and high accuracy.Semi-analytical model is widely used for many hyperspectral shallow water depth inversion models, among which maximum likelihood estimation including environmental noise and bottom intra-class variability(MILEBI) method is an important water depth inversion method.One of the serious problems within the MILEBI method is the overestimation when the water depth is increased.Hense, an improved water depth inversion method is proposed in this paper based on a semi-analytical model.The method considers a prior distribution of depth by including the influence of depth value on the subsurface reflectance, which may lead to an improvement of the accuracy for water depth inversion.The detailed solution is found by adding a regularization term to the loss function of MILEBI method, and a new form of objective function is applied thereby.In this paper, a real dataset and a simulated dataset are used to evaluate the proposed method bench marked by the traditional approaches.The results show that in the scenario of water depth range between 15—20 m, the mean absolute error of MILEBI is about 4.27 m, whereas that of the proposed method is about 2.27 m, significantly better than the classic methods.