Affiliation of Author(s):电子信息工程学院

Journal:PECCS - Proc. Int. Jt. Conf. Pervasive Embed. Comput. Commun. Syst.

Abstract:With the development of computer-assisted algorithms, to improve the accuracy of segment synthetic aperture radar (SAR) sea ice imagery, a novel MRF energy function segmentation approach is proposed and applied. Different initial label field and feature field of MRF energy function have important impacts on the final segmentation results. The initial label field of energy function is semi-supervised by sea ice S-KPFD recognition result. And the feature field is improved by K distribution model in GLCM feature. The optimal energy function of the MRF model is obtained by EM method. The weighting parameter of the feature field is taken as a function of the annealing temperature. The influence of energy field on the classification result will auto-adjust to the annealing temperature. Finally, the four classes result of sea ice is accurately and boundaries distinct. Experiments demonstrate that the proposed algorithm is able to successfully segment various SAR sea ice images and achieve improvement over existing published methods including the SA MRF, GMM, and K-means clustering. © 2017 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.

Translation or Not:no

Date of Publication:2017-01-01

Co-author:Leung, Henry,Xing, Shiyu,Lu, Da

Correspondence Author:孔莹莹,Leung, Henry,Xing, Shiyu,Katherine Kong