基于帧差检测技术与区域特征的红外与可见光图像融合算法
|
摘要
为了提高红外图像的融合质量并降低复杂度,提出一种基于帧差检测技术与区域特征的红外与可见光图像融合算法。首先,设计帧差法对红外图像中的目标完成检测,从而进行目标分簇与图像分割;并借助帧之间的信息完成目标的准确定位;再根据目标区域的特征设计不同的融合规则,充分利用可见光和红外图像的有效信息进行互补,完成图像融合,并对融合算法的复杂度进行理论分析。同时,在可见光和红外图像中目标不可动可观察及目标可运动可观察两种条件下进行融合实验,实验结果表明,与当前图像融合技术相比,所提技术具有更高的融合质量,其融合图像能够准确反映目标和背景。
An infrared and visible image fusion algorithm based on frame difference detection technique and region feature is proposed to improve the fusion quality of the infrared image and reduce its fusion complexity. The frame difference method is designed to detect the target in infrared image for target clustering and image segmentation. The target is accurately located by means of the information among frames. Some different fusion rules are designed according to the characteristics of target region,and the image fusion is realized by means of the effective information complementation of infrared and visible images. The theoretical analysis of the complexity of the fusion algorithm is carried out. The fusion experiments are performed for the unmoving observable target,and moving observable target in infrared and visible image. The experimental results show that,in comparison with the available image fusion techniques,the proposed technique has higher fusion quality,and its fusion image can reflect the target and background more accurately.
An infrared and visible image fusion algorithm based on frame difference detection technique and region feature is proposed to improve the fusion quality of the infrared image and reduce its fusion complexity. The frame difference method is designed to detect the target in infrared image for target clustering and image segmentation. The target is accurately located by means of the information among frames. Some different fusion rules are designed according to the characteristics of target region,and the image fusion is realized by means of the effective information complementation of infrared and visible images. The theoretical analysis of the complexity of the fusion algorithm is carried out. The fusion experiments are performed for the unmoving observable target,and moving observable target in infrared and visible image. The experimental results show that,in comparison with the available image fusion techniques,the proposed technique has higher fusion quality,and its fusion image can reflect the target and background more accurately.
引文
[1] LIU Zhanwen,FENG Yan,ZHANG Yifan. A fusion algorithm for infrared and visible images based on RDU-PCNN and ICAbases in NSST domain[J]. Infrared physics and technology,2016,79(8):183-190.
[2] YIN Ming,DUAN Puhong,LIU Wei. A novel infrared and visible image fusion algorithm based on shift-invariant dual-tree complex shearlet transform and sparse representation[J]. Neurocomputing,2016,32(6):11-17.
[3] LIU C H,QI Y,DING W R. Infrared and visible image fusion method based on saliency detection in sparse domain[J]. Infrared physics and technology,2017,83(4):94-102.
[4]王珺,彭进业,何贵青,等.基于非下采样Contourlet变换和稀疏表示的红外与可见光图像融合方法[J].兵工学报,2014,34(7):815-820.WANG Jun,PENG Jinye,HE Guiqing,et al. Fusion method for visible and infrared image based on non-subsampled Contourlet transform and sparse representation[J]. Acta armamentarii,2014,34(7):815-820.
[5]钱小燕,韩磊,王帮峰.红外与可见光图像快速融合算法[J].计算机辅助设计与图形学学报,2012,23(7):1211-1216.QIAN Xiaoyan,HAN Lei,WANG Bangfeng. A fast fusion algorithm of visible and infrared images[J]. Journal of computeraided design and computer graphics,2012,23(7):1211-1216.
[6]付炜,裴欢,廖晓玉,等.多源遥感图像融合的数据同化算法[J].自动化学报,2013,37(3):309-315.FU Wei,PEI Huan,LIAO Xiaoyu,et al. Data assimilation algorithm of multi-fountain remote sensing image fusion[J]. Acta automatica Sinica,2013,37(3):309-315.
[7] SUN Zhenfeng,LIU Jun,CHANG Qimin. Fusion of infrared and visible images based on focus measure operators in the curvelet domain[J]. Applied optics,2012,51(12):1910-1921.
[8]李光鑫,吴伟平,胡君.红外和彩色可见光图像亮度-对比度传递融合算法[J].中国光学,2011,4(2):161-166.LI Guangxin,WU Weiping,HU Jun. Luminance-contrast transfer base fusion algorithm for infrared and color visible images[J].Chinese journal of optics,2011,4(2):161-166.
[9]邢雅琼,王晓丹,刘健,等.基于NSST域的红外和彩色可见光图像融合[J].系统工程理论与实践,2016,35(6):536-544.XING Yaqiong,WANG Xiaodan,LIU Jian,et al. Fusion technique for infrared and color visible image in non-subsample shearlet transform domain[J]. Systems engineering:theory and practice,2016,35(6):536-544.
[10]杨风暴,董安冉,张雷.DWT、NSCT和改进PCA协同组合红外偏振图像融合[J].红外技术,2017,39(3):201-208.YANG Fengbao,DONG Anran,ZHANG Lei. Infrared polarization image fusion using the synergistic combination of DWT,NSCT and Improved PCA[J]. Infrared technology,2017,39(3):201-208.
[11]刘卫,殷明,栾静,等.基于平移不变剪切波变换域图像融合算法[J].光子学报,2013,42(4):496-503.LIU Wei,YIN Ming,LUAN Jing,et al. Image fusion algorithm based on shift-invariant shearlet transform[J]. Acta photonica Sinica,2013,42(4):496-503.
[12]赵春晖,刘春红,王克成.基于第二代小波的超谱遥感图像融合算法研究[J].光学学报,2015,25(7):891-896.ZHAO Chunhui,LIU Chunhong,WANG Kecheng. Research on fusion of hyperspectral remote sensing image based on second generation wavelet[J]. Acta optica Sinica,2015,25(7):891-896.
[13] ZHOU Zehua,TAN Min. Infrared image and visible image fusion based on wavelet transform[J]. Advanced materials research,2014,35(6):1011-1017.
[14] JING Z L,PAN H,LI Y K,et al. Evaluation of focus measures in multi-focus image fusion[J]. Pattern recognition letters,2007,28(4):493-500.
[2] YIN Ming,DUAN Puhong,LIU Wei. A novel infrared and visible image fusion algorithm based on shift-invariant dual-tree complex shearlet transform and sparse representation[J]. Neurocomputing,2016,32(6):11-17.
[3] LIU C H,QI Y,DING W R. Infrared and visible image fusion method based on saliency detection in sparse domain[J]. Infrared physics and technology,2017,83(4):94-102.
[4]王珺,彭进业,何贵青,等.基于非下采样Contourlet变换和稀疏表示的红外与可见光图像融合方法[J].兵工学报,2014,34(7):815-820.WANG Jun,PENG Jinye,HE Guiqing,et al. Fusion method for visible and infrared image based on non-subsampled Contourlet transform and sparse representation[J]. Acta armamentarii,2014,34(7):815-820.
[5]钱小燕,韩磊,王帮峰.红外与可见光图像快速融合算法[J].计算机辅助设计与图形学学报,2012,23(7):1211-1216.QIAN Xiaoyan,HAN Lei,WANG Bangfeng. A fast fusion algorithm of visible and infrared images[J]. Journal of computeraided design and computer graphics,2012,23(7):1211-1216.
[6]付炜,裴欢,廖晓玉,等.多源遥感图像融合的数据同化算法[J].自动化学报,2013,37(3):309-315.FU Wei,PEI Huan,LIAO Xiaoyu,et al. Data assimilation algorithm of multi-fountain remote sensing image fusion[J]. Acta automatica Sinica,2013,37(3):309-315.
[7] SUN Zhenfeng,LIU Jun,CHANG Qimin. Fusion of infrared and visible images based on focus measure operators in the curvelet domain[J]. Applied optics,2012,51(12):1910-1921.
[8]李光鑫,吴伟平,胡君.红外和彩色可见光图像亮度-对比度传递融合算法[J].中国光学,2011,4(2):161-166.LI Guangxin,WU Weiping,HU Jun. Luminance-contrast transfer base fusion algorithm for infrared and color visible images[J].Chinese journal of optics,2011,4(2):161-166.
[9]邢雅琼,王晓丹,刘健,等.基于NSST域的红外和彩色可见光图像融合[J].系统工程理论与实践,2016,35(6):536-544.XING Yaqiong,WANG Xiaodan,LIU Jian,et al. Fusion technique for infrared and color visible image in non-subsample shearlet transform domain[J]. Systems engineering:theory and practice,2016,35(6):536-544.
[10]杨风暴,董安冉,张雷.DWT、NSCT和改进PCA协同组合红外偏振图像融合[J].红外技术,2017,39(3):201-208.YANG Fengbao,DONG Anran,ZHANG Lei. Infrared polarization image fusion using the synergistic combination of DWT,NSCT and Improved PCA[J]. Infrared technology,2017,39(3):201-208.
[11]刘卫,殷明,栾静,等.基于平移不变剪切波变换域图像融合算法[J].光子学报,2013,42(4):496-503.LIU Wei,YIN Ming,LUAN Jing,et al. Image fusion algorithm based on shift-invariant shearlet transform[J]. Acta photonica Sinica,2013,42(4):496-503.
[12]赵春晖,刘春红,王克成.基于第二代小波的超谱遥感图像融合算法研究[J].光学学报,2015,25(7):891-896.ZHAO Chunhui,LIU Chunhong,WANG Kecheng. Research on fusion of hyperspectral remote sensing image based on second generation wavelet[J]. Acta optica Sinica,2015,25(7):891-896.
[13] ZHOU Zehua,TAN Min. Infrared image and visible image fusion based on wavelet transform[J]. Advanced materials research,2014,35(6):1011-1017.
[14] JING Z L,PAN H,LI Y K,et al. Evaluation of focus measures in multi-focus image fusion[J]. Pattern recognition letters,2007,28(4):493-500.