自动化学报2012，Vol.38Issue(11)：1735-1750,16.DOI:10.3724/SP.J.1004.2012.01735
基于图割和模糊连接度的交互式舰船红外图像分割方法
Interactive Ship Infrared Image Segmentation Method Based on Graph Cut and Fuzzy Connectedness
摘要
Abstract
In order to solve the problem of low contrast, fuzzy edges, and inhomogeneous target gray level in the ship infrared image segmentation, an interactive segmentation method based on graph cut and fuzzy connectedness is proposed. The interaction process uses a rectangular pen brush for selecting the seed points of target and background. The segmentation method is based on graph cut method, in which fuzzy connectedness is introduced to compute the likelihood energy of graph cut, and the automatic computation of its weights is also given, furthermore, a histogram decomposition method is proposed for estimating the number of components of a mixture and the parameters of that model. The simulation results show that the proposed method can segment ship infrared image target effectively under various complicated environments.关键词
交互式图像分割/图割/模糊连接度/高斯混合模型Key words
Interactive image segmentation/ graph cut/ fuzzy connectedness/ Gaussian mixture model (GMM)引用本文复制引用
刘松涛,王慧丽,殷福亮..基于图割和模糊连接度的交互式舰船红外图像分割方法[J].自动化学报,2012,38(11):1735-1750,16.基金项目
中国博士后科学基金(20100471451),水下测控技术国家级重点实验室基金(9140C2603051003)资助 （20100471451）
