期刊
EXPERT SYSTEMS WITH APPLICATIONS
卷 38, 期 7, 页码 8515-8523出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.eswa.2011.01.052
关键词
Self-Generating Neural Network; Fuzzy logic; Optimization; Pruning; Fuzzy fusion
类别
资金
- Science and Technology on Electro-optic Control Laboratory under Aeronautics Foundation [20095151022]
A new fusion algorithm for multi-sensor images based on Self-Generating Neural Network (SGNN) and fuzzy logic is proposed in this paper. This study is an extension of the work described in Qin and Bao (2005). First, the order and frequency modifications for the current McKusick and Langley (M-L) optimization are proposed; next, by combining optimization and pruning together, the Pruning-And-One-Optimization-Composite (PAOOC) processing method is raised; and finally, a modified fuzzy fusion scheme using improved SGNN is put forward. Experimental results demonstrate that the posed fuzzy fusion scheme outperforms region-based fusion using wavelet multi-resolution (MR) segmentation, and region-based fusion using tree-structure wavelet MR segmentation, both in visual effect and objective evaluation criteria. In the meantime, simulations also show the effectiveness of our modifications for the current optimization and pruning methods, visually and objectively. (C) 2011 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据