基于方差分析特征检测的多聚焦图像融合方法(英文)

ANOVA Based Feature Detection for Multifocus Image Fusion

由于光学CCD传感器深度特性的限制,不可能得到一幅场景的全聚焦图像.为了克服这一缺陷,很多学者在这方面作出了研究,到目前为止,一种较好的方法是将多幅多聚焦图像通过某种融合策略汇总成为一幅图像,该幅图像充分表现场景各个位置的全聚焦信息.提出一种基于方差分析方法的多聚焦图像融合算法,对每一幅聚焦图像,求取每一个像素点的方差分析对比度函数.融合过程是根据这一对比度系数确定融合策略,得出融合结果图像.实验结果表明,该算法显示了融合性能的有效性.

Due to the limited depth-of-field of optical lenses, it is difficult to get an image with all objects in focus. One way to overcome this problem is to take several images with different focus points and combine them into a single composite which contains all the regions fully focused. An ANOVA （ANalysis of VAriance） based multitocus image fusion algorithm is described. For each pixel in the source multifocus images, the ANOVA-based contrast function is calculated. The fusion procedure is performed by a selection mode according to the magnitude of the contrast function. Experiments show that the proposed algorithm works well in multifocus image fusion.