Deriving inherent optical properties through training samples with known concentration of water constituents and reflectance spectra
Deriving inherent optical properties through training samples with known concentration of water constituents and reflectance spectra
复制标题
DOI：
10.18307/2009.0208
复制
发表时间：
2009
期刊：
影响因子：
--
通讯作者：
Yang Wei;Matsushita Bunkei;Chen Jin
中科院分区：
文献类型：
--
作者：
Yang Wei;Matsushita Bunkei;Chen Jin
吸收系数和反向散射系数是水的固有光学特性 (IOP)，在水成分浓度的检测中发挥着极其重要的作用。然而，测量水体的眼压非常复杂且耗时。在这项研究中，提出了一种新的算法，通过测量成分浓度和反射光谱的训练样本来推导水成分的 IOP。基于生物光学模型的模拟数据用于评估所提出的算法。结果表明，当样本光谱满足解析模型且训练样本相互独立时，算法的性能可以接受，说明本文提出的算法在理论上是合理的。
Coefficients of absorption and backscattering are inherent optical properties (IOPs) of water, which play extraordinarily important roles in the detection of water constituent concentrations. However, it is very complicate and time-consuming to measure the IOPs of waters. In this study, a novel algorithm was proposed to derive the IOPs of water constituents through training samples with measured constituents’ concentrations and reflectance spectra. Simulated data based on bio-optical model was used to assess the proposed algorithm. Results demonstrated that when the spectra of samples satisfied the analytical model and the training samples were independent each other, the performance of the algorithm can be accepted, which indicates that the algorithm proposed in this paper is theoretically reasonable.