海口湾可见光遥感测深方法研究

Study on the water depth extraction method using visible remote sensing in the Haikou Bay

以海口湾为研究区,利用Landsat资料进行可见光遥感测深方法研究。采用NDWI、MNDWI、EWI、NWI指数的阈值分割法进行水陆分离,结果表明NDWI指数更适合海口湾。选取与水深相关性比较好的反射率因子TM2、TM3、TM1/TM2、TM1/TM3,采用目前常见的可见光遥感测深方法,建立了线性、对数、幂指数、指数测深模型,与实测水深的比较结果表明TM1/TM3的指数测深模型在0-10 m水深的平均绝对误差为1.67 m,在实际应用中有参考价值。此外,建立了TM2、TM3、TM1/TM2、TM1/TM3的二次多项式测深模型以及对数二次多项式测深模型,建立了它们的二元和四元线性、对数测深模型,还建立了它们的分段函数测深模型,与实测水深的比较结果表明,二次多项式测深模型优于线性测深模型,对数二次多项式测深模型优于对数测深模型,分段函数测深模型优于非分段函数测深模型。

The study of water depth extraction methods using visible Landsat remote sensing is made in the Haikou Bay.Four types of water indices, namely NDWI, MNDWI, EWI and NWI, are used to extract the waterline from Landsat sensing image, in which NDWI is proved to be the best. TM2, TM3, TM1/TM2 and TM1/TM3 are checked to have better correlation with the water depth, so they are selected for the water depth extraction models. Based on the familiar methods, the linear model, the logarithmic model, the power exponent model and the exponential model for water depth extraction are made,among which the exponential model using TM1/TM3 is the best with the average absolute error of 1.67 m at the water depth of0-10 m. Furthermore, the quadratic polynomial model and the logarithmic quadratic polynomial model are made for the water depth extraction using TM2, TM3, TM1/TM2 and TM1/TM3 respectively, which are also used to make the binary（ or quaternary） linear（or logarithmic） models and the segmented function models. It is shown that the quadratic polynomial model is better than the linear model, and the logarithmic quadratic polynomial model is better than the logarithmic model and the segment function model is better than the non-segmented function model.