An efficient approach for multi-focus image fusion

To extend the depth-of-field of multi-focus images into one single sharp composite,a new efficient fusion scheme is proposed. one promising method is based on the wavelet transform. as shift invariance of the wavelet transform is important in ensuring robust subband fusion,the proposed method is based on dual-tree complex wavelet transform. we use two different focus measures to select the coefficients of the flised image. a modified energy of laplacian,which can reflect features of the edges,is proposed when choosing the low frequency subband coefficients. for the high frequency subband coefficients,a new sum of laplacian is presented and adopted as local neighborhood feature. the algorithm has been used to merge several sets of multi-focus images; both visual and quantitative analysis show that,comparing with conventional image fusion algorithms,the proposed approach can provide a more satisfactory fusion outcome.