CAJ | 학술논문

基于多尺度多分辨率的图像融合是医学图像融合的重要方法，二维经验模式分解(BEMD)方法是一种新的多尺度多分辨率图像分解方法。本文提出了一种基于 BEMD 的医学图像融合方法。首先将待融合的两幅图像进行BEMD分解，获得多个BIMF分量和一个剩余分量；然后针对BIMF分量和剩余分量采用不同的融合规则进行图像融合；最后对融合后的各分量进行BEMD逆变换，得到最终的融合结果。实验结果表明，本文方法可得到较好的融合效果，融合图像清晰，含有的更多信息。
기우다척도다분변솔적도상융합시의학도상융합적중요방법，이유경험모식분해(BEMD)방법시일충신적다척도다분변솔도상분해방법。본문제출료일충기우 BEMD 적의학도상융합방법。수선장대융합적량폭도상진행BEMD분해，획득다개BIMF분량화일개잉여분량；연후침대BIMF분량화잉여분량채용불동적융합규칙진행도상융합；최후대융합후적각분량진행BEMD역변환，득도최종적융합결과。실험결과표명，본문방법가득도교호적융합효과，융합도상청석，함유적경다신식。
Multi-scale and multi-resolution based image fusion is an important fusion method for medical images. The bidimensional empirical mode decomposition (BEMD) method is a new image multi-resolution decomposition approach. A medical image fusion algorithm based on BEMD is presented. Firstly, the two images to be fusined are decomposed into multiple BIMFs components and a residual component by BEMD decomposition. Then, according to the BIMFs and residual components, use different fusion rules for image fusion. Finally, inverse transform each fusioned component of BEMD, the final fusion result is attained. The experimental results show that better fusion result can be attained by using the method, the fusioned image is clearer and more information is contained.