A ship target automatic detection method for high-resolution remote sensing

With the increasement of spatial resolution of remote sensing, the ship detection methods for low-resolution images are no longer suitable. In this study, a ship target automatic detection method for high-resolution remote sensing is proposed, which mainly contains steps of Otsu binary segmentation, morphological operation, calculation of target features and target judgment. The results show that almost all of the offshore ships can be detected, and the total detection rates are 94% and 91% with the experimental Google Earth data and GF-1 data respectively. The ship target automatic detection method proposed in this study is more suitable for detecting ship targets offshore rather than anchored along the dock.