Adaptive Multi exposure Image Fusion Algorithm Based on Structural Patch Decomposition

Abstract

A new multi exposure image fusion algorithm based on structural patch decomposition theory is proposed to address the issues of color and detail information loss in traditional multi exposure image fusion. Firstly, the image is decomposed into three independent components by using structural block decomposition, and different fusion rules are designed according to the characteristics of each component. In order to make the algorithm applicable to dynamic scenes, the structural consistency constraint is added, and the guided filter is used to smooth the mean intensity component, which effectively overcomes the block effect defect, retains the image edge information and reduces artifacts. Finally, reconstruct the fused three components to obtain the final fused image. The experimental results show that this algorithm can effectively preserve the color and detailed texture information of the input image.