Focus-pixel estimation and optimization for multi-focus image fusion

Abstract To integrate the effective information and improve quality of multi-source images, many spatial or transform domain-based image fusion methods have been proposed in field fusion. The key purpose multi-focus is to focused pixels remove redundant each source image. Theoretically, if complementary different images are detected completely, with best can be obtained. For this goal, we propose a focus-pixel estimation optimization based framework paper. Because an same depth (DOF), multi-scale focus-measure algorithm for matting region firstly. Then, boundaries defocused regions obtained accurately by optimizing strategy. And also fused reduce influence insufficient boundary precision. experimental results demonstrate that method outperforms some previous typical both objective evaluations visual perception.