Application of sentinel-2a image in shallow water depth inversion of Lingshui Sanya coast

The flexibility of traditional bathymetric measurement technology is poor, and the update cycle of bathymetric data is long and the timeliness is poor. Remote sensing technology is used to carry out bathymetric inversion to make up for the deficiency of traditional oceanographic measurement to some extent. Based on the in-situ measured water depth data and Sentinel-2A multispectral image, the water depth inversion of Lingshui-Sanya coast is studied. Through correlation analysis of sentinel-2A images and measured data, ratio model, logarithm model and ratio logarithm model were constructed by selecting blue, green and red wavelengths with high correlation, and accuracy verification was carried out for these three models. The results show that the ratio model has the highest correlation coefficient with high accuracy, but the ratio logarithm model has a better fitting effect in the inversion of water depth over three meters. In this study, the ratio model is selected to perform water depth inversion based on the water depth situation in the experimental area, and good inversion results are obtained.