A Novel Pulse Coupled Neural Network Based Method for Multi-focus Image Fusion

Multi-focus image fusion means to fuse multiple source images with different focus settings into one image, so that the resulting image appears sharper. In order to extract the focused regions of the fused image efficiently, a novel pulse coupled neural network (PCNN) method for multi-focus image fusion is proposed. The registered source images are decomposed into principal components and sparse components by robust principal component analysis (RPCA) decomposition, and the important features of the sparse components are used to motivate the PCNN neurons, whose outputs detect the focused regions of the source images and integrate them to construct the final fused image. Experimental results show that the proposed scheme works better in extracting the focused regions and improving the fusion quality compared to the other existing fusion methods in terms of mutual information (MI) and

[1]  屈小波 Xiaobo Qu,et al.  Image Fusion Algorithm Based on Spatial Frequency-Motivated Pulse Coupled Neural Networks in Nonsubsampled Contourlet Transform Domain , 2008 .

[2]  Reinhard Eckhorn,et al.  Feature Linking via Synchronization among Distributed Assemblies: Simulations of Results from Cat Visual Cortex , 1990, Neural Computation.

[3]  Yingjie Zhang,et al.  Efficient fusion scheme for multi-focus images by using blurring measure , 2009, Digit. Signal Process..

[4]  Qiguang Miao,et al.  A novel adaptive multi-focus image fusion algorithm based on PCNN and sharpness , 2005, SPIE Defense + Commercial Sensing.

[5]  Yaonan Wang,et al.  Combination of images with diverse focuses using the spatial frequency , 2001, Inf. Fusion.

[6]  Mark E. Oxley,et al.  Physiologically motivated image fusion for object detection using a pulse coupled neural network , 1999, IEEE Trans. Neural Networks.

[7]  Zengchang Qin,et al.  Multifocus image fusion based on robust principal component analysis , 2013, Pattern Recognit. Lett..

[8]  Zhi Liu,et al.  Segmentation Driven Low-rank Matrix Recovery for Saliency Detection , 2013, BMVC.

[9]  B. S. Manjunath,et al.  Multisensor Image Fusion Using the Wavelet Transform , 1995, CVGIP Graph. Model. Image Process..

[10]  Harishwaran Hariharan,et al.  Extending Depth of Field via Multifocus Fusion , 2011 .

[11]  G. Sapiro,et al.  A collaborative framework for 3D alignment and classification of heterogeneous subvolumes in cryo-electron tomography. , 2013, Journal of structural biology.

[12]  Yi Ma,et al.  The Augmented Lagrange Multiplier Method for Exact Recovery of Corrupted Low-Rank Matrices , 2010, Journal of structural biology.

[13]  Yi Ma,et al.  Robust principal component analysis? , 2009, JACM.

[14]  John L. Johnson.  Pulse-coupled neural networks , 1994, Defense + Commercial Sensing.

[15]  Shutao Li,et al.  Image matting for fusion of multi-focus images in dynamic scenes , 2013, Inf. Fusion.

[16]  David J. C. MacKay,et al.  Information Theory, Inference, and Learning Algorithms , 2004, IEEE Transactions on Information Theory.

[17]  Vladimir Petrovic,et al.  Objective image fusion performance measure , 2000 .

[18]  Omid Omidvar,et al.  Neural Networks and Pattern Recognition , 1997 .

[19]  Yide Ma,et al.  Review of pulse-coupled neural networks , 2010, Image Vis. Comput..

[20]  Zhongliang Jing,et al.  Evaluation of focus measures in multi-focus image fusion , 2007, Pattern Recognit. Lett..

[21]  Zhongliang Jing,et al.  Multi-focus image fusion using pulse coupled neural network , 2007, Pattern Recognit. Lett..