Computational Imaging Group at Xiamen University 


Matlab code for Image Fusion Algorithm based on Bandelet Transform 

please click on the name of criteria. Just try it... Bandelet_ImageFusion_Toolbox.zip 

Please run bandelet_fusion_256.m Matlab Image Fusion Toolbox for image fusion algorithm based on bandelet transform. You can download this toolbox at http://www.quxiaobo.org/project/Bandelet/Bandelet_ImageFusion_Toolbox.zip or http://dspace.xmu.edu.cn/dspace/bitstream/2288/19179/1/Bandelet_ImageFusion_Toolbox.zip This toolbox contains Matlab files that implement the image fusion algorithms described in paper Xiaobo Qu, Jingwen Yan, Guofu Xie, et al. A Novel Image Fusion Algorithm Based on Bandelet Transform. Chinese Optics Letters,Vol.5,No.10,pp:569572, 2007 English Version of this paper can be downloaded at http://dspace.xmu.edu.cn/dspace/bitstream/2288/15239/1/A+novel+image+fusion+algorithm+based+on+bandelet+transform.pdf %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % References: Xiaobo Qu, Jingwen Yan, Guofu Xie, et al. A Novel Image Fusion Algorithm Based on Bandelet Transform. Chinese Optics Letters,Vol.5,No.10,pp:569572, 2007 (Check http://www.quxiaobo.org,http://quxiaobo.go.8866.org, http://quxiaobo.blog.edu.cn or http://naec.stu.edu.cn for these and other related papers.) QU XiaoBO, %  1. FusionEvalution toolbox at http://dspace.xmu.edu.cn/dspace/bitstream/2288/16979/1/FusionEvaluation.rar 2. Bandelet Toolbox at http://www.cmap.polytechnique.fr/~peyre/bandelets/ or http://www.mathworks.com/matlabcentral/fileexchange/7914 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Acknowledgement 1. This code is copyright and is supplied free of charge for research 2. Thanks Dr.Gabriel Peyre for sharing his bandelet code for our image fusion algorithm. The copyright of this bandelet toolbox belongs to him.


Comment on the selection of geometry of images for image fusion In the original paper Xiaobo Qu, Jingwen Yan, Guofu Xie, et al. A Novel Image Fusion Algorithm Based on Bandelet Transform. Chinese Optics Letters,Vol.5,No.10,pp:569572, 2007 the geometry of two source images are selected by maximum rule. I think it only works well for multifocus image fusion since the geometry may looks like each other. In fact, comparion on which geometry is greater is meaningless since it denotes the direction. So, you'd better chang the rule to get the geometry of fused image from the source images. Especially for the source images whose structure are very different to each other. If you have any comment, please send email to me: <quxiaobo [at] xmu.edu.cn> or <qxb_xmu [at] yahoo.com.cn> 

