Improving mid-frequency contrast in sparse aperture optical imaging systems based upon the Golay-9 array

Abstract

Sparse aperture imaging systems are capable of producing high resolution images while maintaining an overall light collection area that is small compared to a fully filled aperture yielding the same resolution. This is advantageous for applications where size, volume, weight and/or cost are important considerations. However, conventional sparse aperture systems pay the penalty of reduced contrast at midband spatial frequencies. This paper will focus on increasing the midband contrast of sparse aperture imaging systems based on the Golay-9 array. This is one of a family of two-dimensional arrays we have previously examined due to their compact, non-redundant autocorrelations. The modulation transfer function, or normalized autocorrelation, provides a quantitative measure of both the resolution and contrast of an optical imaging system and, along with an average relative midband contrast metric, will be used to compare perturbations to the standard Golay-9 array. Numerical calculations have been performed to investigate the behavior of a Golay-9 array into which autocorrelation redundancy has been introduced and our results have been experimentally verified. In particular we have demonstrated that by proper choice of sub-aperture diameters the average relative midband contrast can be improved by over 55%.

High‐Efficiency Optical Sparse Aperture Metalens Based on GaN Nanobrick Array

“…[23,39] The characteristic of the MTF can be determined directly by the pupil function of the optical system. [23,40,41] The 3D normalized MTF of the all-dielectric UOSA metalens are shown in Figure 2a. As is apparent in Figure 2a, several second peaks around the main peak effectively extend the non-zero MTF area.…”

Active Sparse-Aperture Millimeter-Wave Imaging Using Digital Correlators

“…Golay arrived at these configurations using algorithms based on random guessing and by limiting those solutions to various grid patterns [14]. These configurations can be applied to arrays that have different fill factors and have been applied to imaging systems having individual apertures of varying sizes [16]. Cornwell developed algorithms with non-redundant autocorrelations, but his arrays had the limitation that the detectors need to lay on a circle.…”

An end-to-end deep convolutional neural network for image restoration of sparse aperture imaging system in geostationary orbit

“…Since 1971, researchers have designed some classical aperture patterns with large fill factors and simpler structures [2,3], such as tri-arms, annular, and Golay configurations, they considered the problem of frequency compensation of the system, and subsequently used the traditional image processing algorithms such as the Wiener filter algorithm, Richard-Lucy algorithm, and least squares algorithm to recover blurred images. Many other scholars have designed systems using subapertures with different sizes or implementing multiple imaging by rotating the optical pupil to compensate for the declining or missing frequency information, which combined with image recovery algorithms can obtain a recovery effect that matches the restoration of the human eye, these systems have the light weight and low requirements for sub-aperture design, which is more in line with the application requirements of ground monitoring and environmental detection [4][5][6][7].…”