Towards Reducing Severe Defocus Spread Effects for Multi-Focus Image Fusion via an Optimization Based Strategy

TL;DR

This paper introduces an optimization-based method to improve multi-focus image fusion by reducing defocus spread effects, enhancing image sharpness near focus boundaries using a new assessment metric.

Contribution

It proposes a novel assessment metric combining structure similarity and focus maps, and formulates MFF as an optimization problem solved by gradient ascent.

Findings

Outperforms existing methods on real-world datasets

Effectively reduces defocus spread near focus boundaries

Demonstrates improved image sharpness and clarity

Abstract

Multi-focus image fusion (MFF) is a popular technique to generate an all-in-focus image, where all objects in the scene are sharp. However, existing methods pay little attention to defocus spread effects of the real-world multi-focus images. Consequently, most of the methods perform badly in the areas near focus map boundaries. According to the idea that each local region in the fused image should be similar to the sharpest one among source images, this paper presents an optimization-based approach to reduce defocus spread effects. Firstly, a new MFF assessmentmetric is presented by combining the principle of structure similarity and detected focus maps. Then, MFF problem is cast into maximizing this metric. The optimization is solved by gradient ascent. Experiments conducted on the real-world dataset verify superiority of the proposed model. The codes are available at https://github.com/xsxjtu/MFF-SSIM.