计算机科学 ›› 2016, Vol. 43 ›› Issue (Z6): 122-124.doi: 10.11896/j.issn.1002-137X.2016.6A.029

• 模式识别与图像处理 • 上一篇    下一篇

基于拉普拉斯金字塔与PCNN-SML的图像融合算法

王佺,聂仁灿,金鑫,周冬明,贺康建,余介夫   

  1. 云南大学信息学院 昆明650500,云南大学信息学院 昆明650500,云南大学信息学院 昆明650500,云南大学信息学院 昆明650500,云南大学信息学院 昆明650500,云南大学信息学院 昆明650500
  • 出版日期:2018-12-01 发布日期:2018-12-01
  • 基金资助:
    本文受国家自然科学基金(61365001,61463052),云南省应用基础研究计划项目(2012FD003),云南省科技创新强省计划(2014AB016)资助

Image Fusion Algorithm Using LP Transformation and PCNN-SML

WANG Quan, NIE Ren-can, JIN Xin, ZHOU Dong-ming, HE Kang-jian and YU Jie-fu   

  • Online:2018-12-01 Published:2018-12-01

摘要: 基于拉普拉斯金字塔(LP)与脉冲耦合神经网络(PCNN)变换,提出了一种有效的多聚焦图像融合算法。首先,利用拉普拉斯金字塔对图像进行对多尺度分解,并利用PCNN对每一尺度的分解图像进行处理,以获取描述特征聚类的神经元点火频率图;然后,利用点火频率图的局部拉普拉斯分量绝对和(SML),实现了图像每一尺度LP分解的融合;最后,通过LP分解的重构实现了对多聚焦图像的融合。实验结果表明,所提方法在各项客观评价指标上均优于传统融合算法,体现出了良好的性能。

关键词: 多聚焦图像融合,拉普拉斯金字塔变换,脉冲耦合神经网络,局部SML

Abstract: Using Laplace pyramid algorithm (LP) and pulse coupled neural network (PCNN),this paper proposed an effective fusion algorithm of the multi-focus image.First,the paper used Laplace pyramid to do multi-scale decomposition of the image,and the decomposition images were processed by PCNN,thus the corresponding neuron ignition frequency map was obtained.Then the paper calculated local entropy for every pixel’s ignition frequency map,and took local sum of modified Laplacian(SML) as a measure of the quality of the pixels for the source image fusion.Finally,the paper used the Laplace pyramid reconstruction algorithm to generate the fused image.Experimental results indicate that proposed method is effective and better than other traditional fusion algorithms.

Key words: Multi-focus image fusion,Laplace pyramid transform,Pulse coupled neural network,Local sum of modified laplacian

[1] Vladimir P,Vladimir D.Focused pooling for image fusion evalua-tion[J].Information Fusion,2015,2(3):119-126
[2] Ma Xian-xi,Peng Li,Xu Hong.PCA-based Laplacian pyramid in image fusion[J].Computer Engineering and Applications,2012,48(8):211-213
[3] 黄小丹.基于拉普拉斯金字塔变换的小波域图像融合[J].电子科技,2014,27(6):170-173
[4] Ganasala P,Kumar V.Feature-Motivated Simplified AdaptivePCNN-Based Medical Image Fusion Algorithm in NSST Domain[J].Journal of Digital Imaging,2015,29(1):73-85
[5] 薛寺中,周爱平,梁久祯.基于小波变换的PCNN多传感器图像融合[J].计算机工程与应用,2011,47(35):210-213
[6] Jiang P,Zhang Q,Jing L I,et al.Fusion algorithm for infrared and visible image based on NSST and adaptive PCNN[J].Laser & Infrared,2014,44(1):108-116
[7] 杨艳春,王晓明,党建武,等.基于非下采样Contourlet变换的医学图像融合方法[J].计算机科学,2013,40(3):310-312
[8] 金鑫,聂仁灿,周冬明.一种基于PCNN的改进型虹膜识别算法[J].计算机科学,2014,41(S2):110-115
[9] Li H,Jin X,Yang N,et al.The recognition of landed aircrafts based on PCNN model and affine moment invariants ☆[J].Pattern Recognition Letters,2015,51(C):23-29
[10] Johnson JL,Padgett ML.PCNN models and applications[J].IEEE Transactions on Neural Networks,1999,10(3):480-498
[11] 李智勇,邢应春,刘同旭.基于拉普拉斯金字塔变换的电力图像融合[J].计算机系统应用,2014,23(9):89-93
[12] 左雷,王炜,张鑫.基于像素邻域SML的多焦距图像融合方法[J].微计算机信息,2010,26(35):221-222
[13] Lang J,Hao Z.Novel image fusion method based on adaptive pulse coupled neural network and discrete multi-parameter fractional random transform[J].Optics & Lasers in Engineering,2014,52(1):91-98

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!