Automatic Sounding Generalization Maintaining the Characteristics of Submarine Topography

Abstract

In recent years, demands for electronic nautical chart have increased significantly in various fields of applications, where sounding is one of the main features, and a large number of studies have been conducted toward sounding generalization. However, most automatic sounding generalization methodologies are focused on navigation safety aspects yet seldom consider the depiction function to submarine topography. To address this limitation, we propose a novel approach, termed as sounding generalization maintaining submarine topography (SGMST), that can maintain the submarine terrain as well as observe other generalization rules. Based on the topographic recognition, a new index (the importance values of soundings) was developed to measure the significance of each sounding and then indicate the generalization process. To verify the effectiveness of our method, we applied SGMST on a multibeam sounding experimental dataset and compared it with the influence circle algorithm (ICa) as a baseline. Experimental results showed advantages of our approach meeting the sounding generalization requirements, and the undersea surface modeled by generalized results raises by 0.27 m and 0.37 m through SGMST and ICa, respectively. The new method provides an alternative to future applications involving automatic sounding generalization.