期刊
NEURAL PROCESSING LETTERS
卷 45, 期 1, 页码 75-94出版社
SPRINGER
DOI: 10.1007/s11063-016-9513-2
关键词
Image fusion; Multi-focus image; PCNN; Guided filter; Selection rule
资金
- China Postdoctoral Science Foundation [2013M532097]
- Fundamental Research Funds for the Central Universities [lzujbky-2014-52]
- National Science Foundation of China [61201421, 61175012]
- Science Foundation of Gansu Province of China [1208RJYA058]
This paper proposes a novel multi-focus image fusion method based on pulse coupled neural networks (PCNN) and guided filter. PCNN matches human visual perception very well. And guided filter is an edge-preserving filter which is proposed in recent years. In our method, the fusion process consists of the following steps: firstly, the source images are preliminarily fused with the guided filter. Then the intermediate fused image is employed to motivate the improved PCNN to generate a fusion map. Finally, the source images are fused according to the fusion map. Six contrast methods are employed to evaluate the performance of the proposed approach in six groups of experiments. The experimental results show that the proposed method outperforms the most existing methods in both subjective visual effect and objective evaluation criteria.
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