基于BEMD的多聚焦图像融合算法研究
|
摘要
多聚焦图像融合就是将源目标图像的清晰区域组合成一个各处都清晰的融合图像。论文首先利用bemd(bivariate emd)算法,将图像映射为一个复数信号,在分解中利用复数信号实部和虚部的互相关系,通过循环迭代将待融合图像分解成从高频到低频的有限个二维图像本征函数(bivariate intrinsic mode functions,BIMFs)。再根据每一个空间点的局部方差计算融合的权值系数,最终得到融合图像。这种融合图像算法不需要预先定义小波基函数或滤波器等先验信息,具有良好自适应性。实验结果表明该算法与传统的融合算法相比性能更加优越,可提高融合图像的质量。
Multi-focus image fusion is to extract clear areas of source images and combine them into a clear fusion image. In this paper,using the BEMD(bivariate EMD)algorithm,the image is mapped to a complex signal. In the decomposition,the correlation between the real part and the virtual part of the complex signal is used to decompose the finite two-dimensional image BIMFs(bivariate intrinsic mode functions,BIMFs)from the high frequency to the low frequency. According to the local variance of each spatial point,the weight coefficient of fusion is calculated and the fused image is finally obtained. This fusion algorithm does not need to pre define wavelet basis functions or filters and other prior information,so it has good adaptability. Experimental results show that the algorithm performs better than the traditional fusion algorithm and improves the quality of the fused image.
Multi-focus image fusion is to extract clear areas of source images and combine them into a clear fusion image. In this paper,using the BEMD(bivariate EMD)algorithm,the image is mapped to a complex signal. In the decomposition,the correlation between the real part and the virtual part of the complex signal is used to decompose the finite two-dimensional image BIMFs(bivariate intrinsic mode functions,BIMFs)from the high frequency to the low frequency. According to the local variance of each spatial point,the weight coefficient of fusion is calculated and the fused image is finally obtained. This fusion algorithm does not need to pre define wavelet basis functions or filters and other prior information,so it has good adaptability. Experimental results show that the algorithm performs better than the traditional fusion algorithm and improves the quality of the fused image.
引文
[1]Ardeshir G A,Nikolov S. Image fusion:Advances in the state of the art[J]. Information Fusion,2007,8(2):114-118.
[2]Mahmood A,Tudor P M,Oxford W,et al. Applied multidimensional fusion[J]. Computer Journal,2007,50(6):646-659.
[3]Rafael C G,Richard E W. Digital Image Processing[M].Beijing:Publishing House of Electronics Industry,2003.
[4]Smith M I,Heather J P. A review of image fusion technology in 2005[J]. In The International Society for Optical Engineering(SPIE),2005:29-45.
[5]Li H,Manjunath B S,Mitra S K. Multisensor Image Fusion Using the Wavelet Transform[J]. Graphical Models and Image Processing,1995,57(3):235-245.
[6]Amolins K,Zhang Y,Dare P. Wavelet based image fusion techniques--An introduction,review and comparison[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2007,62(4):249-263.
[7]Ardeshir G A,Nikolov S. Image fusion:Advances in the state of the art[J]. Information Fusion,2007,8(2):114-118.
[8]Huang N E,Shen Z,Long S R. The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinear and Non-stationary Time Series analysis[J]. Proceedings of the Royal Society of London,1998,454:903-995.
[9]宋平舰,张杰.二维经验模态分解在海洋遥感图像信息分离中的应用[J].高技术通讯,2001(9):62-67.SONG Pingjian,ZHANG Jie. The study on sea remote sensing information separation based on two dimensional empirical mode decomposition[J]. Chinese High Technology Letters. 2001(9):62-67.
[10]Nunes J C,Bouaoune Y,Delechelle E. Image analysis by bi-dimensional empirical mode decomposition[J]. Image and Vision Computing. 2003,21(12):1019-1026.
[11]Nunes J C,Guyot S,Delechelle E. Texture analysis based on local analysis of the bi-dimensional empirical mode decomposition[J]. Machine Vision and Applications. 2005,16(3):177-188.
[12]Yang Z H,Qi D X,Yang L H. Signal period analysis based on Hilbert-Huang transform and its application to texture analysis[J]. IEEE Proc.Third Int.Conf.on Image and Graphics. 2004:430-433.
[13]Xiong C Z,Xu J Y,Zou J C. Texture classification based on EMD and FFT[J]. Journal of Zhejiang University(Science A),2006,7(9):1516-1521.
[14]Cui Y L,Pan H,Hong B J. A novel surface interpolation approach for bi-dimensional empirical mode decomposition[C]//2011:335-338.
[15]Paul S,Sevcenco I S,Agathoklis P. Multi-Exposure and Multi-Focus Image Fusion in Gradient Domain[J].Journal of Circuits Systems and Computers,2016,25(10):1650123.
[2]Mahmood A,Tudor P M,Oxford W,et al. Applied multidimensional fusion[J]. Computer Journal,2007,50(6):646-659.
[3]Rafael C G,Richard E W. Digital Image Processing[M].Beijing:Publishing House of Electronics Industry,2003.
[4]Smith M I,Heather J P. A review of image fusion technology in 2005[J]. In The International Society for Optical Engineering(SPIE),2005:29-45.
[5]Li H,Manjunath B S,Mitra S K. Multisensor Image Fusion Using the Wavelet Transform[J]. Graphical Models and Image Processing,1995,57(3):235-245.
[6]Amolins K,Zhang Y,Dare P. Wavelet based image fusion techniques--An introduction,review and comparison[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2007,62(4):249-263.
[7]Ardeshir G A,Nikolov S. Image fusion:Advances in the state of the art[J]. Information Fusion,2007,8(2):114-118.
[8]Huang N E,Shen Z,Long S R. The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinear and Non-stationary Time Series analysis[J]. Proceedings of the Royal Society of London,1998,454:903-995.
[9]宋平舰,张杰.二维经验模态分解在海洋遥感图像信息分离中的应用[J].高技术通讯,2001(9):62-67.SONG Pingjian,ZHANG Jie. The study on sea remote sensing information separation based on two dimensional empirical mode decomposition[J]. Chinese High Technology Letters. 2001(9):62-67.
[10]Nunes J C,Bouaoune Y,Delechelle E. Image analysis by bi-dimensional empirical mode decomposition[J]. Image and Vision Computing. 2003,21(12):1019-1026.
[11]Nunes J C,Guyot S,Delechelle E. Texture analysis based on local analysis of the bi-dimensional empirical mode decomposition[J]. Machine Vision and Applications. 2005,16(3):177-188.
[12]Yang Z H,Qi D X,Yang L H. Signal period analysis based on Hilbert-Huang transform and its application to texture analysis[J]. IEEE Proc.Third Int.Conf.on Image and Graphics. 2004:430-433.
[13]Xiong C Z,Xu J Y,Zou J C. Texture classification based on EMD and FFT[J]. Journal of Zhejiang University(Science A),2006,7(9):1516-1521.
[14]Cui Y L,Pan H,Hong B J. A novel surface interpolation approach for bi-dimensional empirical mode decomposition[C]//2011:335-338.
[15]Paul S,Sevcenco I S,Agathoklis P. Multi-Exposure and Multi-Focus Image Fusion in Gradient Domain[J].Journal of Circuits Systems and Computers,2016,25(10):1650123.