Fusion of Ikonos Imagery Based on Maximum Likelihood Estimation

In order to improve the fusion quality of IKONOS multispectral (MS) and panchromatic (Pan) images, this paper proposes a fusion method using maximum likelihood (ML) estimation. The proposed method firstly uses the sensor characteristics to model the observation process of both MS and Pan images. Then, the cost function with respect to the estimated high-resolution MS images is constructed based on the ML estimation. Finally, the fused images are obtained using a steepest descent optimization algorithm. The experimental results demonstrate that the proposed method can have better spectral result compared with the WT fusion method and perform as well as the maximum a posteriori (MAP) fusion method with a lower computational cost.