摘要
针对复杂背景中/长波红外图像的特点,提出了一种基于小波变换的中/长波红外图像融合检测方法。首先,对原始中/长波红外图像分别进行小波变换,提取出两幅图像的低频和高频信息;然后,分别对图像的低频和高频信息采用不同的融合准则进行融合,并用小波逆变换重建成融合图像,自适应选取阈值,获得最终的目标检测结果。在实验中,采用相同的方法分别用融合图像、单波段红外图像进行目标检测,实验结果证明,用文中提出的融合检测算法优于仅用长波或中波红外图像的目标检测。
Aiming at the features of mid/long-wave infrared image in complex condition, a fusion detection method based on wavelet transform is presented. Low-frequency and high-frequency information are extracted from original mid/long-wave image using wavelet transform. Then fused images are reconstructed according to different fusion rules by wavelet inverse transform. And the background image abstracted from fused image are segmented by adaptive threshold. The final target detection result is obtained. The experimental results show that the fusion detection method superior to the method used midwave or long -wave detecting targets alone,and this method is efficient.
出处
《红外与激光工程》
EI
CSCD
北大核心
2007年第2期240-243,共4页
Infrared and Laser Engineering
基金
国防预研基金资助项目(51483020104JW0516
51476040304JW0508)
关键词
红外图像
小波变换
信息融合
目标检测
Infrared image
Wavelet transform
Information fusion
Target detection
