Object-based image fusion for minimization of color distortion in high-resolution satellite imagery

Color or spectral distortion is a serious issue in the field of Image Fusion. The problem is more prevalent in high-resolution satellite imagery due to spectral mismatching of panchromatic and MS bands. Current pixel-based techniques are not well-equipped to handle the problem. Moreover, these techniques do not consider the spatial, spectral or radiometric characteristics of various objects in the image and all the pixels are fused with the same criterion. To bring the application closer to the real-world scene, a more realistic, self-adjustable, attribute-based object-oriented paradigm towards fusion has been proposed in the paper. Various objects attributes have been considered at different levels to embed greater spatial details into the low resolution MSS image with minimum spectral distortion. Quantitative assessment indices have also been calculated to prove that our method is superior in terms of minimization of color distortion and maximization of spatial details as compared to other methods.