Super-resolution processing of millimeter wave imagery.

Abstract

The primary focus of the work being done in the area of super-resolution of imagery has been done using images taken in the optical portion of the electro-magnetic spectrum. The material presented here describes the development of a super-resolution algorithm for images in the millimeter wave portion of the spectrum. Due to the difference in the statistical properties of the photons in the optical and millimeter wave portions of the spectrum, different mathematical models are needed to describe the object and image in the two wavelength regions. Poisson statistics are commonly used to describe the object and image in the optical region. Development of statistical models for the object and image in the millimeter wave region are presented here. The statistical models investigated here are based on Rayleigh and exponential statistics. Rayleigh statistics are valid if the object of interest is linear in amplitude and the exponential statistics are valid if the object is linear in intensity.