Genetic algorithm-based synthetic variable ratio image fusion

This study proposed to improve the performance of the conventional synthetic variable ratio (SVR) image fusion algorithm by means of the genetic algorithm (GA). The proposed GA-based SVR (GA-SVR) method utilizes the GA to estimate the optimum band weights used to generate the intensity component. Performance of the proposed method was investigated on singlesensor and multisensor images. The spectral and spatial qualities of the GA-SVR results were compared not only against those of conventional SVR results, but also against those of the results of various widely-used image fusion methods. The spectral quality metrics revealed that the GA-SVR method provided spectrally and spatially superior results, compared to the other methods used. It was also concluded that the GA-SVR method presented a good performance not only with singlesensor images, but also with multisensor images.