Joint motion estimation and frame selection algorithm for multi-frame super-resolution reconstruction

Registration of consecutive frames and selection of frame is quite essential in multi-frame image super-resolution. An adaptive frame selection principle is proposed. A method which joints the optical flow algorithm for motion estimation registration and the super-resolution is designed. First, using the optical flow algorithm to calculate the inter-frame motion estimation, designing an adaptive frame selection method to discard some of the larger inter-frame motion frames, and then through sub-pixel image registration to calculate the accurate motion estimation parameters, and finally combine the MAP method for image super-resolution which take into account two iterations of the difference between the resulting image vectors of the next iteration algorithm. Experimental results show that this method not only achieve sub-pixel accurate registration, but also achieve better results in the visual effects and the peak signal to noise ratio.