底质信息在水深反演中的应用研究

Application of substrate composition information in water depth inversion

海底底质和水深密切相关,经研究发现,光谱微分技术可以很好的区分出底质类型。以不同底质的高光谱数据为基础,提出底质分类算法,通过研究不同底质类型的光谱反射率与水深之间的关系,分析并得到了不同底质反射率与水深的相关系数,在分析检验混合底质下各种水深反演算法的适用性的同时,计算出各个算法反演出水深的误差;比较底质分类算法与其它水深反演模型的优劣,底质分类算法的水深反演结果的误差小于其他算法,得到结论,底质分类算法适用于混合底质的情况,误差小于其他水深反演算法,而且能较准确的区分出底质类型,具有一定的实用性和研究价值。

Seabed sediments and water depth are closely related. Our study found that spectral differential technology had a good effort on distinguishing substrate types. This paper, based on hyperspeetral data of different sediment, researched the relationship between spectral reflectance of different type substrate composition and water depth, examined the correlation coefficient between different sediment reflectance and water depth, analyzed and tested the applicability in the mixed depth of sediments of various algorithms, and proved the superiority of sediment classification algorithm. We also studied the difference of sediment types of differential spectral reflectance characteristics. Anti-depth error was less than the other algorithms, and more accurate to distinguish sediment type, had a certain relevance and research value.