Matched Filtering of Small-Sized Objects in Images Containing Constant Spatial Interference

Abstract—The influence of spatial inhomogeneity of the photoelectric characteristics of a matrix photodetector (PDD) on the quality of detection of small-sized objects in the image formed by it is considered. It is shown that in the conditions of spatially inhomogeneous natural noise of the PDD, the traditionally applied two-point correction is not optimal for subsequent matched image filtering performed in order to increase the reliability of object detection. A two-stage algorithm for constructing a linear spatially invariant matched filter is proposed, which provides a significant increase in the reliability of detecting weak small-sized objects against the background of uncorrelated noise in images containing interference due to the spatial heterogeneity of the PDD characteristics.