期刊
NEURAL COMPUTING & APPLICATIONS
卷 29, 期 6, 页码 21-40出版社
SPRINGER LONDON LTD
DOI: 10.1007/s00521-016-2524-0
关键词
Shape matching; Flower pollination algorithm; Atomic potential matching; Numerical optimization
资金
- National Science Foundation of China [61463007, 6153008]
Visual shape matching has been a hot research topic. As a relatively new branch, atomic potential matching (APM) model is inspired by potential field attractions. Compared to the conventional edge potential function (EPF) model, APM has been verified to be less sensitive to intricate backgrounds in the test image and far more cost-effective in the computation process. The optimization process of shape matching can be regarded as a numerical optimization problem, which is disposed by flower pollination algorithm (FPA). This study comprehensively investigates the convergence performances of FPA and the other algorithms in shape matching problem based on APM model. Experimental results of three realistic examples show that FPA is able to provide very competitive results and to outperform the other algorithms.
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