摘要
多聚焦图像融合的关键问题是如何保持原始图像的细节。基于多小波分析具有多个分析基函数和产生更多分解子图像的特点,本文提出1种基于多小波分析的多聚焦图像融合方法。根据小波变换域分解信号的特点,分别对细节信号采用基于局部区域能量指导下的融合规则和对粗大信号采用局部均方误差加权均衡的融合规则,最后考虑了焦点区域的一致性调整。通过实验证明,在采用空间频率和清晰度作为客观评价条件之下,该方法比基于单小波分解、塔式分解和其它的融合方法都具有更好的融合效果。
The key problem in multi-focus image fusion is how to keep the detail in the original focus images. Based on the features of more decomposing sub-images generated by the multi-wavelet analysis function, a new image fusion method to the multi-focus images is proposed in this paper. The fusion rules are different to the details and the crude space in the wavelet transform. To the detail space, the local windows energy is selected as the fusion guide. And the least mean square error is selected as the guide to the crude space. After that the coherence caused by the focus are considering to adjust the fusion results. Taking the space frequency and the mean gradient as criterions, experiments with the popular multi-focus images show that this method is more effective compared with single wavelets and pyramid decompositions and the other fusion methods.
出处
《中国海洋大学学报（自然科学版）》
CAS
CSCD
北大核心
2007年第1期163-167,共5页
Periodical of Ocean University of China
基金
国家自然科学基金项目(60572064)资助
关键词
多聚焦图像
多小波分析
图像融合
multi-focus images
multi-wavelet analysis
image fusion
