Multi-focus image fusion algorithm based on region detection and ACS network

To address the problems of artifacts and information residuals in existing multi-focus image fusion algorithms, this paper proposes a multi-focus image algorithm based on region detection and the ACS network. First, region detection is used to identify the focus and boundary regions, allowing for the creation of an initial fusion image. Second, the ACS network learns the fusion rules for multi-focus images, which results in a network fusion image. Finally, using the boundary region information, the initial and network fusion images are weighted and combined to create the final fused image. The experimental results show that the algorithm outperforms other comparative algorithms in both the focus region and the boundary region, and the evaluation indexes are improved by more than 4.8% and 1.5%, respectively; at the same time, the subjective effect is more in line with the HVS, and there are no visual artifacts in various regions.