A Multispectral and Panchromatic Images Fusion Method Based on Weighted Mean Curvature Filter Decomposition

Since the hardware limitations of satellite sensors, spatial resolution multispectral (MS) images is still not consistent with panchromatic (PAN) images. It especially important to obtain MS high in field remote sensing image fusion. In order and spectral resolutions, a novel PAN fusion method based on weighted mean curvature filter (WMCF) decomposition proposed this paper. Firstly, local frequency-based (WLSF) utilized fuse all bands generate an intensity component IC. accordance matting model, IC taken as original α channel for estimation foreground background Secondly, decomposed into small-scale (SS), large-scale (LS) basic by Gaussian (GF). The multi-scale morphological detail measure (MSMDM) value used inputs Parameters Automatic Calculation Pulse Coupled Neural Network (PAC-PCNN) model. With MSMDM-guided PAC-PCNN are effectively fused. fused well LS SS linearly combined together construct last channel. Finally, image, reconstructed acquire final image. experimental results four pairs show that achieves superior terms subjective objective evaluations. particular, can different resolutions higher operational efficiency, which effective means