Polyphase back-projection filtering for image resolution enhancement

The method for reconstruction and restoration of super-resolution images fr om sets of low-resolution images presented is an extension of the algorithm proposed by Irani and Peleg (1991). After estimating the projective transf ormation parameters between the image sequence frames, the observed data ar e transformed into a sequence with only quantised sub-pixel translations. T he super-resolution reconstruction is an iterative process, in which a high -resolution image is initialised and iteratively improved. The improvement is achieved by back-projecting the errors between the translated low-resolu tion images and the respective images obtained by simulating the imaging sy stem. The imaging system's point-spread function (PSF) and the back-project ion function are first estimated with a resolution higher than that of the superresolution image. The two functions are then decimated so that two ban ks of polyphase filters are obtained. The use of the polyphase filters allo ws exploitation of the input data without any smoothing and/or interpolatio n operations. The presented experimental results show that the resolution i mprovement is better than the results obtained with Irani and Peleg's algor ithm.