摘要
针对原有红外偏振融合算法中单一差异特征对不确定和随机变化的图像特征信息不能有效描述而产生不利于融合的问题,本文在分析源图像差异特征形成机理基础上,提出了一种基于NSCT(non-subsampled contourlet transform)的红外偏振与光强图像的特征差异驱动融合算法。通过实验仿真表明,相比SVT(support value transform)、WPT(wavelet packet transform)和NSCTLELV(NSCT local energy and local variance),该算法能更有效地融合源图像的互补和细节信息,且在实际的目标识别中具有一定的应用价值。
Given that the single difference features cannot effectively describe uncertain and random image feature information in the original infrared polarization fusion algorithms,and result in problems against the image fusion,we propose a feature differ-ences driven fusion algorithm of infrared polarization and intensity image based on NSCT in the analysis of the source image for-mation mechanism of the difference features.The simulation results indicate that the proposed algorithm is more effective than SVT algorithm,WPT and NSCTLELV,and is able to integrate the complementary and detail information of the source image, which makes it have certain value in the practical target recognition.
出处
《中国科技论文》
CAS
北大核心
2014年第1期96-102,共7页
China Sciencepaper
基金
高等学校博士学科点专项科研基金资助项目(20121420110004)
山西省回国留学人员科研资助项目(20120706ZX)
中北大学科学基金资助项目(20110604)
