作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2008, Vol. 34 ›› Issue (16): 230-231. doi: 10.3969/j.issn.1000-3428.2008.16.079

• 人工智能及识别技术 • 上一篇    下一篇

多聚焦图像融合新算法

赵 晶,杨志豪,林鸿飞   

  1. (大连理工大学计算机系,大连 116023)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-08-20 发布日期:2008-08-20

Novel Multifocus Image Fusion Algorithm

ZHAO Jing, YANG Zhi-hao, LIN Hong-fei   

  1. (Department of Computer, Dalian University of Technology, Dalian 116023)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-08-20 Published:2008-08-20

摘要: 提出一种基于限邻域经验模式分解(NLEMD)的多聚焦图像融合算法。该算法利用NLEMD的自适应特性及高频细节信息的较强获取能力,通过对不同图像对应的内蕴模式函数分量IMF中的像素按照局部最大能量,即最优的原则进行选取,然后将融合后的内蕴模式函数分量和剩余量反向重构获取融合图像。实验结果表明,该算法具有更强的细节获取能力,融合效果优于传统的基于小波分解的融合算法。

关键词: 限邻域经验模式分解, 局域波, 图像融合, 多聚焦图像

Abstract: This paper proposes one new multifocus image fusion algorithm based on Neighborhood Limited Empirical Mode Decomposition (NLEMD). The images are decomposed by NLEMD, and this algorithm makes full use of adaptive features of NLEMD and the powerful achieving capability of high frequency data to fuse images. In different Intrinsic Mode Functions(IMF) the pixel which has the local max energy is selected as the fusion pixel, and the image is reconstructed using IMFs and remnants. Experiments prove that the new algorithm has more powerful achieving capability of high frequency data and better effect than fusion algorithm based on wavelet analysis.

Key words: Neighborhood Limited Empirical Mode Decomposition(NLEMD), local wave, image fusion, multifocus image fusion

中图分类号: