Conference on Lasers and Electro-Optics

Optimal strategies in image processing : a quantum study

Nicolas Treps, V. Delaubert, Agnès Maître, Claude Fabre

Published Jan 1, 2004• 0 citations

Abstract

We consider the problem of extracting information from an optical image. The modes responsible for the quantum noise affecting the extraction are identified, which leads to a strategy optimizing the information read out.

DOI: 10.1364/iqec.2004.itum6

Related papers

Passive optical superresolution at the quantum limit

A. I. Lvovsky, Michael R. Grace, Saikat Guha +3 more

For more than a century, the diffraction limit has defined the resolution achievable by passive optical imaging systems. Although some resolution improvement can be gained through classical data processing of the image, it is limited by the noise arising from quantum nature of light. Minimizing the effect of this noise requires quantum treatment of optical imaging. By reformulating imaging as a problem of quantum measurement and estimation, it becomes possible to identify optimal detection strategies that recover spatial information previously thought inaccessible. This review summarizes the theoretical framework that underpins this development, from the formulation of quantum Cramer-Rao bounds and Chernoff bounds to the construction of receivers that attain them, such as those based on spatial-mode demultiplexing. We show how these methods can beat conventional imaging in the classification, localization, and imaging of sub-Rayleigh incoherent sources. We then discuss extensions to multiparameter and partially coherent scenarios, and highlight the unifying connections between estimation and discrimination tasks. Finally, we survey recent experimental demonstrations that approach quantum-limited resolution and outline emerging applications in microscopy, astronomy, and optical sensing.

quant-pharXiv:2605.10767OpticaMay 12, 20260 citations

What is the optimum way to extract information from optical images at the standard quantum limit and beyond ?

Claude Fabre, Nicolas Treps, V. Delaubert +2 more

The problem of extracting a given piece of information from an image with the maximum sensitivity is solved in the general case. Implementations using image intensity processing techniques and homodyne measurements are given.

Conference on Lasers and Electro-OpticsJan 1, 20060 citations

Quantum limits in image processing

Claude Fabre, V. Delaubert, Nicolas Treps +2 more

We determine the general limit to the maximum achievable sensitivity in the estimation of a parameter from the information contained in an optical image in the presence of quantum noise, either coherent or squeezed.

International Conference on Quantum InformationJan 1, 20071 citation

Interference-based optical image encryption using three-dimensional phase retrieval

Wen Chen, Xudong Chen

In recent years, optical image encryption has attracted more and more attention in information security due to its unique advantages, such as parallel processing and multiple-parameter characteristics. In this paper, we propose a new method using three-dimensional (3D) processing strategy for interference-based optical image encryption. The plaintext is considered as a series of particles distributed in 3D space, and any one sectional extraction cannot render information about the plaintext during image decryption. In addition, the silhouette problem in the conventional interference-based optical encryption method is effectively suppressed, and the proposed optical cryptosystem can achieve higher security compared with the previous work. A numerical experiment is conducted to demonstrate the feasibility and effectiveness of the proposed method.

Applied OpticsAug 24, 201238 citations

Optimal strategies in image processing - a quantum study

Treps, V. Delaubert, Maitre +1 more

Conference on Lasers and Electro-OpticsJan 1, 20040 citations

Quantum-Inspired Detection Strategies for Optical Superresolution Imaging

Amit Kumar Jha

Quantum information theory has opened a new paradigm in the field of optical imaging, establishing the narrative that pre-processing the information bearing light following a quantum optimal detection strategy can achieve the fundamental performance limit. In this dissertation, we apply this narrative to study two important imaging areas : (1) passive optical superresolution imaging and (2) beam-displacement active optical sensing. Over the past decade, passive optical superresolution imaging has been extensively studied using the canonical problem of resolving two incoherent point sources. Direct detection suffers from Rayleigh’s curse when the separation goes below the Rayleigh criterion, whereas performing Spatial Mode Demultiplexing (SPADE), a quantum optimal detection strategy, gets rid of this curse, refuting Rayleigh’s minimum resolution formula. SPADE is built with multiple stratified phase masks in a multi-plane light-conversion (MPLC) setup. In the first part of this dissertation, we study how placing a single-phase mask at the pupil plane of an imaging system can affect its resolution under direct detection. We use the Classical Fisher Information (CFI) as the information metric to quantify the performance of separation estimation. Also known as PSF Engineering, we find that while this method improves the small-separation Fisher Information scaling for Gaussian PSFs (a zero-less PSF), it is not able to do the same for PSFs with zeros (Sinc PSF/hard aperture). We also study this behavior from the perspective of hypothesis testing between one versus two-point sources and derive the Bhattacharyya exponent. After this, we explore sparse aperture configurations for the hypothesis test between one versus two-point source discrimination problem. We also analyze the separation estimation problem under atmospheric turbulence and low-order wavefront correction to study the robustness of direct detection and BSPADE (Binary SPADE) performance. The atmospheric turbulence is realized using phase masks. We find that under low to moderate turbulence, BSPADE offers an advantage over direct detection measurement. We also present a Photomagnetically actuated deformable mirror (PMADM) design which can be useful for low-order wavefront correction. In the second part of the thesis, we study coherent beam displacement sensing, an active optical sensing problem. We study a free-space propagation channel with soft Gaussian apertures that supports Hermite–Gaussian (HG) modes, with the relative displacement between the apertures representing the change to be detected. Classical relative entropy (CRE) between the pre-change and post-change measurement distributions is used as the performance metric as a larger relative entropy corresponds to a shorter asymptotic change detection delay. We show that coherent probe and mode engineering strategies can increase this relative entropy and thereby reduce detection latency. In particular, we compare an optimized classical coherent probe with a spray-painted probe in which a small quantum resource (squeezing) is distributed across an optimized coherent spatial mode. We evaluate the performance using the CRE between the probability distributions of the pre-change and post-change states produced by phase matched homodyne detection. Together, these results support a unified view of pre-processing the information bearing light and using it for optical superresolution and optical sensing purposes. The dissertation explores methods such as PSF, aperture, and modal engineering, which can find applications ranging from astronomy and microscopy to optomechanical sensing.

Jan 1, 20260 citations