Estimation of Shallow Water Bathymetry from Remote Sensing Data in Consideration of Variation of Water

从遥感数据中准确地估计浅海的水深是很重要的，因为可以以较低的成本获得大范围的水深数据。在现有的浅水测深研究中，假定浅海海底反射率和水体衰减系数是均匀的，但它们会因底质和悬浮物的不同而不同。在这项研究中，提出了一种考虑水体衰减系数和海底反射率变化的水深估算新方法。为了获取浅海测深的有效信息，提出了一种新的去噪算法，该算法包含一个辐射对数值的四阶偏微分方程组。通过测量浅水砂层的光谱反射率，利用基准线上的水深数据估算水的衰减系数。估算的海底光谱反射率和水体衰减系数都不是均匀的，而是在谷区略有变化。利用SPOT/XS影像数据和这些参数对浅水4m以下的水深进行了评价。在距基准线两侧600m范围内估算的水深，其标准偏差小于0.4m。

It is important to accurately estimate a water depth in shallow water from remote sensing data, since data on water depth in a wide region can be obtained with a cheaper cost. In existing researches on estimation of shallow water bathymetry, bottom reflectance and water attenuation coefficients are assumed to be uniform in shallow water, nevertheless they are varied by bottom materials and suspended particulates. In this investigation, a new method of estimating water depth have been developed in consideration of variation of water attenuation coefficients and bottom reflectance. In order to retrieve an effective information on shallow water bathymetry, a new noise reduction algorithm which includes an 4th-order partial differential equation for an absolute radiance logarithmic value has been developed. Spectral reflectances of sand bottom in shallow water is measured to evaluate water attenuation coefficients with the data of water depth along a reference line. Both the spectral reflectance of the bottom and water attenuation coefficients estimated are not uniform but slightely vary in a trough region. The water depthes shallower than 4m have been evaluated with SPOT/XS image data and these parameters. The estimated water depthes in the range of 600m distant bothside from a reference line have an error with the standard deviation of less than 0.4m.