A Quantitative Method to Control and Adjust the Accuracy of Adaptive Grid Depth Modeling

To address the limitations of current methods to control and adjust the accuracy of depth models and relatively low accuracy, a quantitative method to control and adjust the accuracy of adaptive grid depth modeling is proposed. First, uncertainties in source data and interpolated depths are estimated, and the representation uncertainty derived from finite and discrete points representing the continuous seafloor surface is analyzed. Second, mean vertical uncertainty in an arbitrary given area is calculated. Finally, interpolation of the depths at grid nodes from source data and the distribution framework of the grid nodes are optimized in each local area, and an adaptive grid depth model is created according to the expected accuracy. The experimental results demonstrate that (1) the proposed method can control and adjust the accuracy of the depth model in each local area such that the accuracy of the constructed model meets the requirements of the expected index as closely as possible and (2) the proposed method can improve the accuracy of the depth modeling by optimizing the interpolation and distribution of the grid nodes.