Sea Ice Drift Tracking with a MCC Method of Automatically Acquiring Features

Features of sea ice is often not obvious, and to delineate feature areas manually would limit the effect of MCC method for tracking sea ice drift. This study focuses on automatic selection of feature regions. First, an object-oriented segmentation method, which is based on the level iterative merging algorithm, is used for the sea ice areas. Then the corner points are obtained as the candidate feature points, and some obscure candidate feature points are eliminated by a proposed removing rule. Finally, these feature points are used to monitor the sea ice drift based on the MCC method of automatically acquiring features. Based on sea ice data set in the Bohai Sea from Geostationary Ocean Color Image (GOCI) satellite, the results show that this proposed method can improve the efficiency of the MCC method, and the experimental results are also verified with a back-matching method.