A Method for Interpolating Digital Depth Model Based on Semiparametric Regression and Uncertainty Weighting. Issue 3 (February 2019)

• Record Type: Journal Article
• Authors: Jia, Shuaidong; Chen, Qiu; Tang, Lulu; Dai, Zeyuan; Zhang, Lihua
• Is Part Of: IOP conference series. Volume 237:Issue 3(2019)
• Publication Date: 2019-02
• Subjects: Earth sciences -- Periodicals; Environmental sciences -- Congresses; Environmental sciences -- Periodicals; 550.5
• DOI: 10.1088/1755-1315/237/3/032001
• Languages: English
• ISSNs: 1755-1307

Abstract

Abstract: To address a limitation of DDM interpolation not considering the error distribution characteristics, a method for interpolating Digital Depth Model based on semiparametric regression and uncertainty weighting is proposed in this paper. Firstly, the observed error from the raw depths is considered to be random, and the weighting matrix is built for expressing this kind of random error. Secondly, the approximate error from the interpolation function is regarded to be systematic, and the semiparametric regression model is constructed for taking the part of systematic error into consideration. Finally, a new method for interpolating Digital Depth Model is proposed considering both the random and systematic error. The experimental results demonstrate that: (1) the proposed method has improved the quality of DDM; (2) the greater the uncertainties of the depths varies, the higher the proposed method has improved the quality of DDM; (3) the more complex the seafloor topography is, the higher the proposed method has improved the quality of DDM.