海洋高光谱正交变换及悬浮粒子浓度预测模式确立

Orthogonal Transform to Oceanic Hyper-Spectra Reflectance Data and Algorithm Model for Calculation Suspended Solids Concentration (SSC) in the Sea

如何从海洋光谱数据中获取海洋水体中的悬浮粒子浓度(SSC)一直是人们关注和研究的实际问题之一。本文就以往此类问题研究中的不定,提出了新的方法。它未采用直接从海洋光谱向量的分量中选取因子的作法,而是首先对海洋光谱向量实施正交变换,然后从变换产生的新向量的分量中选取少量几个分量构建SSC预测方程的新方法。这个方法可作到充分利用海洋光谱中所含的几乎全部SSC信息,克服了以往作法存在的不足。作为实例,文中运用此方法,对渤海海区测得的海洋高光谱数据进了处理计算,在误差限≤0.00017的条件下,选取了6个分量构建了SSC的预测式,并给出了式中的7个系数值及预测精度2.25％。

The oceanic hyper-spetra reflectance measurement has been conducted. How to acquire oceanic information from oceanic hyper-spectra data is followed with interest and studied up on. To solve this problem,in this paper,an orthogonal transform method is applied to 256-channel spectra data and an algorithm model is established for calculating SSC. There are only six orthogonal variates in the model.