基于方向性对比度金字塔的图像融合方法

摘要/Abstract

摘要： 针对基于对比度塔形分解的图像融合方法具有良好的物理意义，却没有强调方向性的不足，提出了一种具有方向性的对比度金字塔图像融合方法。对多聚焦图像进行对比度塔形分解后，利用方向滤波器组对高频加方向，得到不同方向的高频子分量。针对不同频率域特点，采用低频分量系数取加权平均，高频分量系数绝对值取大的融合规则，对分解后的子图像进行融合。对比基于对比度塔形分解(CP)和基于离散小波变换(DWT)的融合方法，提出的方法既能保持对比度的含义，又可提供个方向信息。实验结果表明，提出的方法其融合结果图像有较高的清晰度和空间分辨率。

Abstract: The fusion method based on contrast pyramid decomposition has good physical meaning, but there is no emphasis on direction. In order to solve this problem, an image fusion method based on directional contrast pyramid decomposition is proposed. The contrast pyramid decomposition of multi-focus images is performed. And then multiple directional sub-images are obtained for each high frequency image by using directional filter banks. According to the characteristics of the different frequency domain, the fusion rule is that the weighted average value is selected for low frequency subband coefficients; while for high frequency subband coefficients, the coefficients with larger absolute value are employed. Compared to the fusion methods based on the contrast pyramid (CP) decomposition and discrete wavelet transform (DWT), the proposed method can not only keep the meaning of contrast, but also provide direction information. The experiment results show that the fused images with the proposed method have high definition and spatial resolution.

Key words: directional filter banks

中图分类号:

TP391.4

刘斌，董迪，陈俊霖. 基于方向性对比度金字塔的图像融合方法[J]. J4, 2017, 34(4): 405-413.

参考文献

[1] 汪晓波, 刘斌. 基于多分辨奇异值分解的多聚焦图像融合[J]. 量子电子学报, 2014, 31(3): 257-263.
Wang Xiaobo, Liu Bin. Multi-focus image fusion based on multi-resolution singular value decomposition[J]. Chinese Journal of Quantum Electronics (量子电子学报), 2014, 31(3): 257-263 (in Chinese).
[2] 侯庆岑, 潘晨, 杨勇. 基于视觉显著度的多聚焦图像融合方法[J]. 计算机工程, 2015, 41(1): 211-217.
Hou Qingcen, Pan chen, Yang yong. Multi-focus image fusion method based on visual saliency degree[J]. Computer Engineering (计算机工程), 2015, 41(1): 211-217 (in Chinese).
[3] 刘贵喜, 杨万海. 基于多尺度对比度塔的图像融合方法及性能评价[J].光学学报, 2001, 21(11): 1336-1342.
Liu Guixi, Yang Wanhai. A multi-scale contrast-pyramid-based image fusion scheme and its performance evaluation[J]. Acta Optica Sinica (光学学报), 2001, 21(11): 1336-1342 (in Chinese).
[4] 金海燕, 刘芳, 焦李成. 基于多尺度对比度塔和方向滤波器组的图像融合[J]. 电子学报, 2007, 35(7): 1295-1300.
Jin Haiyan, Liu Fang, Jiao Licheng. A method of image fusion based on multi-scale contrast pyramid and directional filter banks[J]. Acta Electronica Sinica (电子学报), 2007, 35(7): 1295-1300 (in Chinese).
[5] 邹北骥, 胡艺龄, 辛国江. 基于小波分解和PCNN的图像融合方法[J]. 计算机工程与科学, 2011, 33(2): 102-107.
Zou Beiji, Hu Yiling, and Xin Guojiang. A new image fusion algorithm based on wavelet transform and PCNN[J]. Computer Engineering and Science (计算机工程与科学), 2011, 33(2): 102-107 (in Chinese).
[6] 刘斌, 刘维杰, 马嘉利. 基于三通道不可分对称小波的多聚焦图像融合[J]. 仪器仪表学报, 2012, 33(5): 1110-1116.
Liu Bin, Liu Weijie, Ma Jiali. Multi-focus image fusion based on three channel non-separable symmetrical wavelets[J]. Chinese Journal of Scientific Instrument (仪器仪表学报), 2012, 33(5): 1110-1116 (in Chinese).
[7] 刘斌, 乔双梁, 魏艳萍. 基于采样三通道不可分小波的多光谱图像融合[J]. 仪器仪表学报, 2015, 36(3): 645-653.
Liu Bin, Qiao Shuangliang, Wei Yanping. Multi-spectral image fusion method based on sub-sampled three channel non-separable wavelets[J]. Chinese Journal of Scientific Instrument (仪器仪表学报), 2015, 36(3): 645-653 (in Chinese).
[8] 徐小军, 王友仁, 陈帅. 基于下采样分数阶小波变换的图像融合新方法[J]. 仪器仪表学报, 2014, 35(9): 2061-2069.
Xu Xiaojun, Wang Youren, Chen Shuai. Novel image fusion method based on down-sampling fractional wavelet transform[J]. Chinese Journal of Scientific Instrument (仪器仪表学报), 2014, 35(9): 2061-2069 (in Chinese).
[9] Bamberger R H, Smith M J T. A filter bank for the directional decomposition of image: theory and design [J]. IEEE Transaction on Signal Processing, 1992, 40(4): 882-893.
[10] Do M N, Vetterli M. The contourlet transform: an efficient directional multi-resolution image representation[J]. IEEE Transaction on Image Processing, 2005, 14(12): 2091-2106.
[11] Li Shutao, Yang Bin, Hu Jianwen. Performance comparison of different multi-resolution transforms for image fusion[J]. Information Fusion, 2011, 12(2): 74-84.
[12] Nguyen T T , Oraintara S. Multi-resolution direction filter banks: theory, design and applications[J]. IEEE Transaction on Signal Processing, 2005, 53(10): 3895-3905.
[13] Nguyen T T, Oraintara S. A class of multi-resolution directional filter banks[J]. IEEE Transaction on Signal Processing, 2007, 55(3):949-961.
[14] 刘坤, 郭雷, 常威威. 基于Contourlet变换的区域特征自适应图像融合算法[J]. 光学学报, 2008, 28(4): 681-686.
Liu Kun, Guo Lei, and Chang Weiwei. Regional feature self-adaptive image fusion algorithm based on contourlet transform[J]. Acta Optica Sinica (光学学报), 2008, 28(4): 681-686 (in Chinese).
[15] 涂超平, 肖进胜, 杜康华等. 基于各向异性热扩散方程的多聚焦图像融合算法[J]. 电子学报, 2015, 43(6): 1192-1199.
Tu Chaoping, Xiao Jinsheng, Du Kanghua, et al. Multi-focus image fusion algorithm based on the anisotropic thermal diffusion equation [J]. Acta Electronica Sinica (电子学报), 2015, 43(6): 1192-1199 (in Chinese).
[16] Chai Y, Li H F, Guo M Y. Multifocus image fusion scheme based on features of multiscale products and PCNN in lifting stationary wavelet domain[J]. Optics and Communications, 2011, 284(5):1146-1158.