CAJ | 학술논문

针对同一场景两幅严格配准的多聚焦图像融合问题,设计一种基于区域一致性的多聚集图像融合算法.该算法利用比率金字塔变换得到图像的多分辨序列,采用基于区域一致性的融合方法,在图像的相应层次序列的各级金字塔图像上进行融合以获取最终的融合图像.从目视效果来看,采用文中算法进行图像融合,基本上能够将两幅左右聚焦的图像融合成一幅清晰的单一聚焦图像.实验结果表明:该算法与同类其他算法相比较,相关系数提高了0.08%,偏差度降低了30.67%~31.05%,互信息提高了7.92%~7.97%.
침대동일장경량폭엄격배준적다취초도상융합문제,설계일충기우구역일치성적다취집도상융합산법.해산법이용비솔금자탑변환득도도상적다분변서렬,채용기우구역일치성적융합방법,재도상적상응층차서렬적각급금자탑도상상진행융합이획취최종적융합도상.종목시효과래간,채용문중산법진행도상융합,기본상능구장량폭좌우취초적도상융합성일폭청석적단일취초도상.실험결과표명:해산법여동류기타산법상비교,상관계수제고료0.08%,편차도강저료30.67%~31.05%,호신식제고료7.92%~7.97%.
Focusing on the fusion problem of the multi-focus images from same scene, the multi-focus image fusion algorithm based on regional consistency is designed. By using multi-resolution image sequences obtained from the ratio pyramid transform, the source images were fused on each corresponding levels with the regional consistency method, then the fused image was achieved through inverse ratio pyramid transform. From the visual effect, the designed algorithm can fuse into a clear single focus image from left-focus image and right-focus image. Experimental results and comparisons with three traditional algorithms demonstrate that the designed algorithm is superior with 0.08% improvement of correlation coefficient, 30.67% to 31.05% reduction in the deviation degree and 7.92% to 7.97% enhancement in mutual information.