An Algorithm for Identification of Inland River Shorelines based on Phase Correlation Algorithm

Accurate detection of water surface boundary is an important condition for autonomous navigation of unmanned surface vehicle(USV), but the existence of water reflection becomes one of those difficulties during the detection. In order to solve this problem, this paper proposes to use the phase correlation algorithm to determine the location of shorelines in inland river, the precision of phase registration is improved by feature extraction under constraint conditions, and the inclination angle of shoreline is determined by affine transform angle of feature. The algorithm could overcomes the influence of water reflection during the detection of shorelines.