In many imaging systems, the detector array is not sufficiently dense to adequately sample the scene with the desired
field of view. In order to enhance the existed image resolution, several approaches to solve this problem have been
investigated previously, such as maximum a posteriori probability (MAP), projection onto convex sets(POCS) etc. Those
algorithms enhance reconstruct high resolution with reduced aliasing, from a sequence of undersampled frames. But
whether POCS, or MAP estimator in space domain, image pixels are rearranged by using lexicographic ordering as a
large matrix in procession. These methods have to solve a large ill-condition equation group, which leads to a big burden
of computation and storage, complexity of algorithm. So they are rarely used in practical application.
In order to solve these problems, a novel reconstruction high resolution(HR) image algorithm based on the standard
displacements of low resolution(LR) images is proposed. Moreover, a set of recursive updating algorithm models is
presented. The results of simulating experiments show that the resolution, the details as well as the definition of the high
resolution image given by using our method are greatly enhanced. At the same time, the running speed of our method is
greatly faster than other super-resolution methods.