14 December 2015 In-shore ship extraction from HR optical remote sensing image via salience structure and GIS information
Author Affiliations +
Proceedings Volume 9815, MIPPR 2015: Remote Sensing Image Processing, Geographic Information Systems, and Other Applications; 98150U (2015) https://doi.org/10.1117/12.2203662
Event: Ninth International Symposium on Multispectral Image Processing and Pattern Recognition (MIPPR2015), 2015, Enshi, China
Abstract
In order to solve the problem of in-shore ship extraction from remote sensing image, a novel method for in-shore ship extraction from high resolution (HR) optical remote sensing image is proposed via salience structure feature and GIS information. Firstly, the berth ROI is located in the image with the aid of the prior GIS auxiliary information. Secondly, the salient corner features at ship bow are extracted from the berth ROI precisely. Finally, a recursive algorithm concerning the symmetric geometry of the ship target is conducted to discriminate the multi docked in-shore targets into mono in-shore ships. The results of the experiments show that the method proposed in this paper can detect the majority of large and medium scale in-shore ships from the optical remote sensing image, including both the mono and the multi adjacent docked in-shore ship cases.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xiaoyuan Ren, Xiaoyuan Ren, Libing Jiang, Libing Jiang, Xiao-an Tang, Xiao-an Tang, } "In-shore ship extraction from HR optical remote sensing image via salience structure and GIS information", Proc. SPIE 9815, MIPPR 2015: Remote Sensing Image Processing, Geographic Information Systems, and Other Applications, 98150U (14 December 2015); doi: 10.1117/12.2203662; https://doi.org/10.1117/12.2203662
PROCEEDINGS
7 PAGES


SHARE
Back to Top