Super-resolution imaging techniques based on spatial coherence

A common problem in imaging is the need to resolve features of a composite light source below the spatial scale defined by the Rayleigh limit in a scenario when the emission process from the source cannot be manipulated. A great deal of work has been devoted to deblurring or deconvolution techniques which aim to process data collected with standard imaging systems. A recently proposed alternative is to perform measurement of light in the image plane by carefully demultiplexing spatial modes using, for example, the SPADE technique. The technique has been demonstrated to deliver reliable information on a separation between two point sources, with a diffraction-limited imaging system. However, it requires a priori information about the properties of the source, such as its centroid, for reliable estimation. Collecting this information also consumes resources that need to be taken into account in the overall protocol budget.

The objective of the proposed project is to explore systematically super-resolution techniques that make use of spatial coherence of light collected in the image plane. Such techniques build upon the information contained in the spatial phase of the collected light that escapes detection in conventional, direct imaging schemes. While spatial coherence has been exploited e.g. in SPADE, its full potential is yet to be understood and explored given a wide choice of spatially-resolved measurements and clever utilization of signal post-processing techniques which avoid the estimation bias during data collection.

During the project, we will consider and compare different approaches to detect spatial coherence based on spatial mode filtering, self-referencing interferometry, coherent detection using a local oscillator as a reference, as well as processing capabilities offered by spatially multimode quantum memories architectures. Attention will be paid to efficient retrieval of observables that carry information about image characteristics of interest. We will also investigate whether and to what degree superresolution imaging can be implemented at the stage of data postprocessing. Numerical tools will be developed with an aid of quantum metrology concepts and methods to achieve optimized performance. Additionally, the quantum metrologic toolbox will be used to identify the ultimate limits on the estimation of relevant image characteristics, to motivate measurement strategies and to implement and benchmark efficient estimation procedures. Finally, we will apply the developed theory and tools in an experimental demonstration of superresolution imaging based on spatial coherence.

Given successful completion, the project will pave the way for superresolution imaging techniques employing information contained in the spatial coherence of light collected in the image plane. Developed measurement and estimation strategies, as well as data processing procedures, will enhance the achievable estimation precision of relevant image characteristics. Created theoretical basis, as well as experimental proof of concept demonstrations realized during the project, will enable future construction of more robust and resource efficient imaging systems, especially aimed at, but not limited to, stellar and terrestrial applications. From a fundamental point of view, completion of the project will bring new insights into classical incoherent imaging by employing the tools of quantum metrology and exploring the potential of spatial coherence in the image plane.