Imaging through distributed-volume aberrations using single-shot digital holography

Abstract

This paper explores the use of single-shot digital holography data and a novel algorithm, referred to as multiplane iterative reconstruction (MIR), for imaging through distributed-volume aberrations. Such aberrations result in a linear, shift-varying or "anisoplanatic" physical process, where multiple-look angles give rise to different point spread functions within the field of view of the imaging system. The MIR algorithm jointly computes the maximum a posteriori estimates of the anisoplanatic phase errors and the speckle-free object reflectance from the single-shot digital holography data. Using both simulations and experiments, we show that the MIR algorithm outperforms the leading multiplane image-sharpening algorithm over a wide range of anisoplanatic conditions.

Digital holography experiments with degraded temporal coherence

“…Recent results show that digital holography (DH) is an enabling technology for tactical applications, such as deep-turbulence wavefront sensing [1][2][3] and long-range imaging. [4][5][6] By flood illuminating a distant object and interfering the scattered signal with a local reference, we can reconstruct the amplitude and phase of the complex-optical field. Furthermore, we can approach the the shot-noise limit, given a strong reference.…”

Coherent-Image Reconstruction Using Convolutional Neural Networks

“…For each reconstruction, we measured the image quality using the peak-signal-to-noise ratio (PSNR). Additionally, we used peak Strehl ratio, S p ∈ (0, 1), to measure the quality of our estimate for φ , where S p = 1 indicates a perfect estimate [2]. Figure 2 summarizes the results of our work and Fig.…”

“…In coherent imaging, we seek to reconstruct a scene's real-valued reflectance, r, from complex-valued measurements, y = A φ g + w, where A φ is the sensor's measurement matrix with unknown phase errors, φ , and w is additive white Gaussian noise with variance σ 2 w [1,2]. Here, g is the reflection coefficient, a zero-mean, circularlysymmetric complex normal random variable with variance E[|g s | 2 |r] = r s .…”

A STEM outreach tool for demonstrating the sensing and compensation of atmospheric turbulence

“…In most coherent imaging scenarios, we measure the field reflected off an object and model the measured data as y = A φ g + w, where A φ is the sensor's measurement matrix with unknown phase errors, φ, and w is additive white complex Gaussian noise with variance σ 2 w . 3,6 Here, g is the reflection coefficient, a zero-mean, circularly-symmetric complex normal random variable, with variance E[|g s | 2 |r] = r s , that produces a speckled image. The variance of g is produced by rough surface scattering.…”