Localized Random Sampling for Compressive-sensing Image Processing

We formulate a new sampling scheme motivated by physiological receptive field structure, localized random sampling, which yields significantly improved CS image reconstructions. For each set of localized image measurements, our sampling method first randomly selects an image pixel and then measures its nearby pixels with probability depending on their distance from the initially selected pixel. We compare the uniformly-random and localized random sampling methods and show that, higher quality image reconstructions can be consistently obtained by using localized random sampling.