Due to the poor temporal sampling by low orbiting satellites that carry microwave radiometers, data gaps exist in satellite derived time series of precipitation. This poses a challenge for assimilating rainfall data into forecast models. To yield a continuous time series, the classic image processing technique of digital image morphing has been used. However, the digital morphing technique was applied manually and that is time consuming. In order to avoid human intervention in the process, an automatic procedure for image morphing is needed for real-time operations. For this purpose, Genetic Algorithm Based Image Registration Automatic Morphing (GRAM) model was developed and tested in this paper. Specifically, automatic morphing technique was integrated with Genetic Algorithm and Feature Based Image Metamorphosis technique to fill in the data gaps between satellite coverage. This technique was tested using NOWRAD rainfall radar reflectivity data which are generated from the network of NEXRAD radars. Time series of NOWRAD data from Hurricane Floyd that occurred along the US east coast on September 16, 1999 for 0000, 0100, 0200,0300, and 0400 UTC were used. The GRAM technique was applied to data collected at 0000 and 0400 UTC. These images were also manually morphed. Images at 0100, 0200 and 0300 UTC were interpolated from the GRAM and manual morphing and compared with the original NOWRAD radar reflectivities. The results show that the GRAM technique outperformed manual morphing. The correlation coefficients between the images generated using manual morphing are 0.905, 0.900, and 0.905 for the images at 0100, 0200,and 0300 UTC while the corresponding correlation coefficients are 0.946, 0.911, and 0.913, respectively, based on the GRAM technique.