A PDE-Driven Laplacian Multi-Exposure Fusion for Structure Preserving Image

Multi-Exposure Image Fusion (MEF) aims to fuse a single visually symmetricalimage, which combines the luminosity and preserves the structural data of a scene atdifferent exposure levels. Traditionally, Laplacian Pyramid (LP) fusion methods rely onGaussian smoothing, which diffuses intensity isotropically, resulting in blurred edges, haloartifacts, and loss of fine detail. To overcome these limitations, the paper proposes amethod comprising Mean Curvature and Laplacian Pyramid that combines curvature-based edge preservation with multiscale structural decomposition. Mean CurvatureFiltering (MCF), formulated on curvature flow principles, acts as an anisotropicsmoothing operator that preserves edges and geometric discontinuities while suppressinglow-frequency noise. The MCF is integrated into the pyramid construction to ensurestructure-aware decomposition, producing curvature-sensitive base and detail layers. Acovariance-based adaptive weighting strategy is employed for detail layer fusion,enhancing locally salient textures, while the base layer is averaged to maintain consistentillumination. The fused image is reconstructed through recursive MCF-based expansion toachieve seamless integration of global brightness and multiscale structural details.Experimental evaluation indicates that the proposed framework achieves enhancedsharpness, natural contrast, and reduced artifacts by effectively balancing noisesuppression with edge fidelity. The approach thus provides a mathematically consistent andperceptually coherent formulation for multi-exposure fusion using curvature-drivenmultiscale representation.