Coast dock extraction method based on waterline and perceptual organization

Abstract

Navigation landmark features such as docks are often used in ships for localization and searching for shore targets during the voyage, which are of great economic and military significance. Continuous and complete dock data could hardly be extracted by existing coast dock extraction method, because the spatial relationships and other characteristics of dock are often ignored which only considers grayscale characteristics of remote sensing image .A dock extraction method based on waterline and perceptual organization is proposed in this paper. To make full use of characteristics spatial relationships, perceptual organization is introduced into dock extraction. Firstly, waterline of coast zone remote sensing image is extracted. Then, perceptual organization and evidence theory are introduced, after which doing perceptual organization based on grayscale characteristics and structural features following the rules of perceptual organization. Finally, evidence is verified to accurately extract dock contours. Experimental results show that the method is able to accurately extract continuous and complete dock contours, providing a new approach for coast dock accurate extraction.