CAJ | 학술논문

提出了一种基于局部特征分析的多聚焦图像融合方法。首先采用二维经验模式分解（ BEMD ）对不同聚焦图像进行分解，得到多个内蕴模函数（ IMF ）分量，然后提取出第一个 IMF 分量的统计信息作为图像融合的依据，对多张不同聚焦的图像进行融合，得到最终的融合图像。实验结果表明，本文提出的均值 IMF 方案与其他方案相比，具有较高的融合质量和较低的计算复杂度。
제출료일충기우국부특정분석적다취초도상융합방법。수선채용이유경험모식분해（ BEMD ）대불동취초도상진행분해，득도다개내온모함수（ IMF ）분량，연후제취출제일개 IMF 분량적통계신식작위도상융합적의거，대다장불동취초적도상진행융합，득도최종적융합도상。실험결과표명，본문제출적균치 IMF 방안여기타방안상비，구유교고적융합질량화교저적계산복잡도。
In this paper, a multi-focus image fusion method is proposed based on local characteristics analyses. Firstly, multiple intrinsic mode function (IMF) components are produced by applying Bidimensional Empirical Mode Decomposition (BEMD) to the source images. Then, the statistical information extracted from the first IMF component is used as the basis for image fusion. Finally, the fused image can be obtained by comparing the statistical information of different multi-focus images. Experimental results show that, the proposed designed IMF-Mean scheme can achieve better fusion quality and lower computational complexity, compared with other schemes.