基于小波变换的双波段红外图像融合方法  

Fusion algorithm of dual waveband infrared images based on wavelet transformation

作　　者：朱祥玲[1,2] 吴钦章[3] 陈洪[2] 

机构地区：[1]中国科学院大学,北京100049 [2]中国西昌卫星发射中心技术部,四川西昌615000 [3]中国科学院光电技术研究所,四川成都610209

出　　处：《激光与红外》2014年第5期572-576,共5页Laser & Infrared

摘　　要：针对红外探测器长波图像和中波图像的特征,采用基于小波变换的方法实现双波段红外图像的融合。在对不同波段红外图像小波分解的基础上,对低频子带图像采用加权平均,对高频子带图像采用区域能量最大的融合规则得到融合图像的多分辨率结构,再利用小波逆变换重构融合图像。仿真结果表明,和普通基于空域加权平均的方法相比无论从主观视觉还是通过客观评价指标的计算结果来看,基于小波变换的双波段红外图像融合方法均具有更为理想的图像融合效果。According to the radiation characteristics of long wave infrared image and medium wave image,a fusion al-gorithm of dual waveband infrared images based on wavelet transformation is presented.Firstly,the low frequency sub-band images are fused with corresponding weighting coefficients,and then the high frequency sub-band images are fused according to the rule of the max region power.Finally,the fused image is reconstituted through the inverse wave-let transformation.Computer simulation shows that the fused dual band infrared image based on wavelet transformation has more preferable effect in subjective visual and objective evaluation compared with the fusion algorithm of spatial weighed average.

关 键 词：红外图像 小波变换 图像融合 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]