摘要
分析了红外中波段的两个细分波段(3.4μm^4.1μm和4.5μm^5.3μm)的成像特点,提出了基于二维小波包变换和局部能量的两个红外中波细分波段图像的融合算法.实验结果表明,融合图像同细分图像和未细分图像相比,标准偏差分别增加了6.47%和1.50%,多向粗糙度分别增加了4.82%和4.32%,而太阳影响参数分别下降了4.96%和9.03%,融合了两幅细分图像中比较清晰的信息,减少了太阳饱和区,证明了融合算法的有效性,说明了通过两个细分波段融合的方法可以获得比原中波波段成像效果更好的图像.
Image characterizations in subdivision band 3.4 -4. μm and 4.5 - 5.3μm of mid-wave infrared（MWIR） were studied. A fusion algorithm of images of two MWIR subdivision band was presented based on both two dimension wavelet packet transform and local energy. Comparing the subdivision image with the no-subdivision image, the standard deviation was increased 6.47% and 1.50% , respectively; the roughness concentration of multi direction increased 4.82% and 4. 32% , moreover the sun effect parameter decreased 4.96% and 9.03% , respectively. The fusion result contains the relatively legible information of double sub-band MWIR image and the sun saturation section is reduced. Hence, the validity of algorithm is proved. The experimental results show that the image of fusion of two subdivision band images is better than that of original MWIR.
出处
《红外与毫米波学报》
SCIE
EI
CAS
CSCD
北大核心
2008年第4期275-279,共5页
Journal of Infrared and Millimeter Waves
基金
973(51325010402)资助项目
关键词
图像融合
中波红外
细分波段
小波包变换
image fusion
mid-wave infrared （MWIR）
subdivision band
wavelet packet transform
