A Detection Framework for Ship in Sea-Sky Background Based on Constrained Feature Points

An automated scheme for sea-sky background ship segmentation is proposed. Feature points based on Harris operator are first calculated to extract the important information of the image, and then a constrained feature points methodology based on fractal theory is utilized in order to get the precise location of the ship. Finally region growing algorithm is used in the vicinity of the reserved feature points to segment the ship target from the sea-sky background. The proposed scheme is validated on a large number of sea-sky background images. It is shown to perform well to get the outline of the ship with little time.