Infusion of images the process involves gathering important information from two or more images which gives better consolidated results for further analysis in any domain. In practice before the fusion of images applied the images have to be registered, registration helps in aligning the images which is necessary before fusion of images is performed. In this paper, pre-registered images using SIFT and RANSAC are utilized to use in our proposed method of fusion that is an optimized selective hybrid fusion of registered images. The technique is proposed and investigated for obtaining better results. The proposed method utilizes the DWT transform with different rules to perform fusion of DWT bands such as average, maximum, edge and energy based rules which are combined in a novel way to achieve higher results in comparison. Entropy is the evaluation metric used for performance comparison. As the experiments are performed the proposed method more effective and gives improved results for fusion of images.