W变换和NSCT相结合的多聚焦图像融合方法
作者:
作者单位:

北京林业大学理学院北京100083

中图分类号:

TP391

基金项目:

国家自然科学基金(61571046,61372190,61370193)资助项目


Multifocus image fusion algorithm based on Wtransform and NSCT
Author:
Affiliation:

College of Sciences, Beijing Forestry University, Beijing 100083, China

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [22]
  • |
  • 相似文献 [20]
  • | | |
  • 文章评论
    摘要:

    图像的多尺度分解技术和融合规则是决定多聚焦图像融合效果的关键因素。k次W系是一类以k次多项式和k次分段多项式为基函数的新的混合正交函数系统,对应的W变换是一种具有正交性和精确重构性的有效多分辨分析工具。结合W变换的多尺度特点和非子采样方向滤波器组变换的多方向性,提出了一种新的基于W变换和非子采样方向滤波器组(NSDFB)的多尺度多方向变换。该变换利用W变换对图像进行多尺度分解,利用二维NSDFB对W分解的高频子带系数进行方向分解,得到不同尺度不同方向的子带图像。在此基础上,提出了一种新的多聚焦图像融合算法。该算法针对多聚焦图像高频系数的特点,改进了常用简化脉冲耦合神经网络算法,并将其用于高频系数的融合规则中。实验结果表明,提出的融合方法能够有效地选择源图像中的聚焦良好区域,抑制伪影信息,产生视觉效果更好的融合图像,且在标准差、信息熵、平均梯度和空间频率等客观评价指标上都优于传统的基于Contourlet变换、非下采样Contourlet变换、离散小波变换的融合方法。

    Abstract:

    Multiscale decomposition method and fusion rule are two key factors for multifocus image fusion method. The Wsystem of degree k is an orthogonal hybrid function system consisting of polynomials of degree k and piecewise polynomials of degree k. The corresponding Wtransform is an effective multiresolution analysis tool with orthogonality and reproducibility. By combining the multiscale characteristics of Wtransform and multidirection property of the nonsubsampled directional filter bank (NSDFB) transform, a new multiscale and multidirectional transform is proposed in this paper, in which multiscale decomposition is performed by the Wtransform and the multidirectional decomposition is obtained from NSDFB. On this basis, a new multifocus image fusion algorithm is proposed. According to the characteristics of the high frequency coefficients, the often used simplified PCNN is modified, and applied on the fusion rules of the high frequency coefficients. The experimental results show that the proposed fusion method can select pixels from focus areas of the source images and effectively avoid generation of artifacts. The fused image has better visual effect. The objective evaluation index including standard deviation, entropy, average gradient, and spatial frequency of the fused image demonstrate that the proposed method outperforms the fusion methods based on Contourlet transform, NSCT transform and discrete wavelet.

    参考文献
    [1]NEJATI M,SAMAVI S,SHIRANI S.Multifocus image fusion using dictionarybased sparse representation [J]. Information Fusion,2015,25(9):7284.
    [2]张寅玥,付炜. 多源遥感图像融合算法研究[J]. 电子测量技术,2010,33(6):3438. ZHANG Y, FU W, Research on the algorithms of multisource remote sensing image fusion [J]. Electronic Measurement Technology,2010,33(6):3438.
    [3]李萍,张波,尚怡君. 基于红外图像和特征融合的飞机目标识别方法[J]. 电光与控制,2016(8):9296. LI P, ZHANG B, SHANG Y J. Aircraft target identification based on infrared image and feature fusion [J]. Electronics Optics & Control,2016(8):9296.
    [4]许良凤,林辉,胡敏. 基于差分进化算法的多模态医学图像融合[J]. 电子测量与仪器学报,2013,27(2):110114. XU L F, LIN H, HU M. Multimodality medicine image fusion based on differential evolution algorithm [J]. Journal of Electronic Measurement and Instrument, 2013,27(2):110114.
    [5]BURKE C J, BENCARDINO J, ADLER R. The potential Use of ultrasoundmagnetic resonance imaging fusion applications in musculoskeletal intervention [J]. Journal of Ultrasound in Medicine, 2017, 36(1): 217224.
    [6]TANG L, LI L D, QIAN J S H, et al. NSCTbased multimodal medical image fusion with sparse representation and pulse coupled neural network[J]. Journal of Information Hiding and Multimedia Signal Processing, 2016, 7(6): 13061316.
    [7]BAVIRISETTI D P, DHULI R. Fusion of infrared and visible sensor images based on anisotropic diffusion and karhunenloeve transform [J]. IEEE Sensors Journal,2016,16(1): 203209.
    [8]JURADO J,FISHER K,VETH M. Inertial and imaging sensor fusion for imageaided navigation with affine distortion prediction [C]. Proceedings of IEEE/ION Plans,2012:518526.
    [9]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 B, QIAO S L, WEI Y P. Multi spectral image fusion method based on subsampled three channel nonseparable wavelets [J]. Chinese Journal of Scientific Instrument ,2015,36(3):645653.
    [10]曹风云,赵凯,王筱薇倩,等. 自适应水下彩色图像增强算法[J]. 电子测量与仪器学报,2016,30(5):772778. CAO F Y, ZHAO K, WANG X W Q, et al. An adaptive underwater image enhancement algorithm [J]. Journal of Electronic Measurement and Instrument, 2016, 30(5):772778.
    [11]ZHAO H J, SHANG Z W, TANG Y Y, et al. Multifocus image fusion based on the neighbor distance [J]. Pattern Recognition, 2013, 46 (3):10021011.
    [12]BI X J, PAN T W, DIAO P F. Multifocus image fusion based on local clarity of SCM [J]. Journal of Information Hiding and Multimedia Signal Processing, 2017, 8(1):6875.
    [13]闫光豹,李开宇,杨盛亚. 基于FPGA的红外和CCD图像融合研究及实现[J]. 电子测量技术,2015,38(6):4347. YAN G B, LI K Y, YANG S Y. Infrared and CCD image fusion research and implementation based on FPGA [J]. Electronic Measurement Technology,2015,38 (6):4347.
    [14]郭雷,程塨,赵天云. 基于小波变换和邻域特征的多聚焦图像融合算法[J].西北工业大学学报,2011,29(3):454459. GUO L, CHENG G, ZHAO T Y. A new and effective multifocus image fusion algorithm based on wavelet transform and neighborhood features [J].Journal of Northwestern Polytechnical university,2011,29(3):454459.
    [15]肖进胜,饶天宇,贾茜,等.基于图切割的拉普拉斯金字塔图像融合算法[J]. 光电子·激光,2014(7): 14161424. XIAO J S,RAO T Y,JIA X,et al. An image fusion algorithm of laplacian pyramid based on graph cutting [J].Journal of Optoelectronics Laser,2014(7):14161424.
    [16]DO M N,VETTERLI M. The contourlet transform: An efficient directional multiresolution image representation [J]. IEEE Transactions on Image Processing, 2005, 14(12):20912106.
    [17]KONG W,LEI Y,LEI Y,et al. Fusion technique for greyscale visible light and infrared images based on nonsubsampled contourlet transform and intensityhuesaturation transform [J]. IET Signal Processing,2011,5(1):7580.
    [18]BHATNAGAR G, JONATHAN W Q M, LIU ZH. Directive contrast based multimodal medical image fusion in NSCT domain [J]. IEEE Transactions on Multimedia, 2013,15(5): 10141024.
    [19]WANG X CH,WANG Y,SUN H L,et al. Application of a hybrid orthogonal function system on trademark image retrieval [J]. Journal of Advanced Mechanical Design, Systems, and Manufacturing,2014,8(7):113.
    [20]PADMA G, VINOD K. CT and MR image fusion scheme in nonsubsampled contourlet transform domain [J]. Journal of Digital Imaging,2014,27(3):407418.
    [21]YANG Y,TONG S,HUANG SH Y,et al. LogGabor energy based multimodal medical image fusion in nsct domain [J]. Computational and Mathematical Methods in Medicine,2014,835481:112.
    [22]杨勇,郑文娟,黄淑英,等.人眼视觉感知特性的非下采样Contourlet变换域多聚焦图像融合[J]. 中国图象图形学报,2014,19(3):447455. YANG Y, ZHENG W J, HUANG SH Y, et al. Multifocus image fusion based on human visual perception characteristic in nonsubsampled Contourlet transform domain [J]. Journal of Image and Graphics, 2014, 19(3): 447455.
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

王璐,王小春. W变换和NSCT相结合的多聚焦图像融合方法[J].电子测量与仪器学报,2017,31(5):756-765

复制
分享
文章指标
  • 点击次数:3585
  • 下载次数: 18043
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 在线发布日期: 2017-07-27
文章二维码