Andrew Robert Harvey

Human situational awareness is powered by the ultra-wide field of view of the human visual system and enhanced perception provided by stereopsis. Current thermal imagers, with their narrow, monocular field of view, cannot match this capability and functionality. Increasing the field of view requires consideration of fundamental optical scaling laws for lenses, which dictate that increasing the field of view of fast optics requires increased optical complexity, which can critically impact radiometric sensitivity, size, weight and cost. We report recent progress in the development of our previously reported multicamera architecture, whose 360°×56° field of view provides an immersive situational awareness for day and night panoramic imaging. We employ small, low-cost consumer grade microbolometer arrays, which enable this architecture to be conformally integrated into diverse vehicle platforms or body-worn systems. Human vision has not evolved to enable such a wide field of view to be interpreted, so we employ artificial intelligence algorithms to provide automated tracking of moving objects and people. We furthermore use the parallax provided by a person-worn camera array undergoing natural motion to enhance the recognition of partially occluded objects throughout the entire field of view. Additionally, we present a method for synthesising an occlusion-removed view of partially-occluded objects for enhanced imaging and detection.

The 3D location of light emitters provides essential information in many biological, chemical and physical systems. 4Pi single-molecule localization microscopy (SMLM) being one of the state-of-the-art techniques provides exceptional precision in the axial direction; however, the lateral precision degrades significantly when the emitter is displaced several wavelengths away from the mutual focal plane. Here we propose a hybrid system employing type-B 4Pi microscope and engineered point-spread function (PSF), the resultant interferometric PSF is optimized against the Cramer-Rao metric, yielding the richest information regarding the emitter’s 3D locations. It outperforms conventional 4Pi SMLM with a larger axial range and compares advantageously to the non-interferometric PSFs with doubled photon efficiency and better precision in all directions.

The maximum number of pixels that can be recorded in a retinal imager is limited by ocular aberrations to about 5Mpixels. We report a new multi-scale concept able to yield diffraction-limited images with 50MPixels space-bandwidth product.

We report how illumination of the retina through the sclera enables the recording of high-contrast reflectance and fluorescence images of the retina, free of the cornea and lens glare and autofluoresence that degrades images recorded using traditional illumination through the pupil.

The biological world involves intracellular and intercellular interactions that occur at high speed, at multiple scales and in three dimensions. Acquiring 3D images, however, typically requires a compromise in either spatial or temporal resolution compared to 2D imaging. Conventional 2D fluorescence imaging provides high spatial resolution but requires plane-by-plane imaging of volumes. Conversely, snapshot methods such as light-field microscopy allow video-rate imaging, but at the cost of spatial resolution. Here we introduce 3D engineered point-spread function microscopy (3D-EPM), enabling snapshot imaging of real-world 3D extended biological structures while retaining the native resolution of the microscope in space and time. Our new computational recovery strategy is the key to volumetrically reconstructing arbitrary 3D structures from the information encapsulated in 2D raw EPM images. We validate our technique on both point-like and extended samples, and demonstrate its power by imaging the intracellular motion of chloroplasts undergoing cyclosis in a sample of Egeria densa. Our technique represents a generalised computational methodology for 3D image recovery which is readily adapted to a diverse range of existing microscopy platforms and engineered point-spread functions. We therefore expect it to find broad applicability in the study of rapid biological dynamics in 3D.

The ability of a microscope to rapidly acquire wide-field, high-resolution images is limited by both the optical performance of the microscope objective and the bandwidth of the detector. The use of multiple detectors can increase electronic-acquisition bandwidth, but the use of multiple parallel objectives is problematic since phase coherence is required across the multiple apertures. We report a new synthetic-aperture microscopy technique based on Fourier ptychography, where both the illumination and image-space numerical apertures are synthesized, using a spherical array of low-power microscope objectives that focus images onto mutually incoherent detectors. Phase coherence across apertures is achieved by capturing diffracted fields during angular illumination and using ptychographic reconstruction to synthesize wide-field, high-resolution, amplitude and phase images. Compared to conventional Fourier ptychography, the use of multiple objectives reduces image acquisition times by increasing the area for sampling the diffracted field. We demonstrate the proposed scaleable architecture with a nine-objective microscope that generates an 89-megapixel, 1.1 µm resolution image nine-times faster than can be achieved with a single-objective Fourier-ptychographic microscope. New calibration procedures and reconstruction algorithms enable the use of low-cost 3D-printed components for longitudinal biological sample imaging. Our technique offers a route to high-speed, gigapixel microscopy, for example, imaging the dynamics of large numbers of cells at scales ranging from sub-micron to centimetre, with an enhanced possibility to capture rare phenomena.

Pixelation occurs in many imaging systems and limits the spatial resolution of the acquired images. This effect is notably present in quantum imaging experiments with correlated photons in which the number of pixels used to detect coincidences is often limited by the sensor technology or the acquisition speed. Here, we introduce a pixel super-resolution technique based on measuring the full spatially-resolved joint probability distribution (JPD) of spatially-entangled photons. Without shifting optical elements or using prior information, our technique increases the pixel resolution of the imaging system by a factor two and enables retrieval of spatial information lost due to undersampling. We demonstrate its use in various quantum imaging protocols using photon pairs, including quantum illumination, entanglement-enabled quantum holography, and in a full-field version of N00N-state quantum holography. The JPD pixel super-resolution technique can benefit any full-field imaging system limited by the sensor spatial resolution, including all already established and future photon-correlation-based quantum imaging schemes, bringing these techniques closer to real-world applications.

In Fourier ptychography, multiple low resolution images are captured and subsequently combined computationally into a high-resolution, large-field of view micrograph. A theoretical image-formation model based on the assumption of plane-wave illumination from various directions is commonly used, to stitch together the captured information into a high synthetic aperture. The underlying far-field (Fraunhofer) diffraction assumption connects the source, sample, and pupil planes by Fourier transforms. While computationally simple, this assumption neglects phase-curvature due to non-planar illumination from point sources as well as phase-curvature from finite-conjugate microscopes (e.g., using a single-lens for image-formation). We describe a simple, efficient, and accurate extension of Fourier ptychography by embedding the effect of phase-curvature into the underlying forward model. With the improved forward model proposed here, quantitative phase reconstruction is possible even for wide fields-of-views and without the need of image segmentation. Lastly, the proposed method is computationally efficient, requiring only two multiplications: prior and following the reconstruction.

Multi-scale imaging promises wide-field imaging of the retina with a space-bandwidth product that approaches the Fresnel limit. We describe an implementation that achieves 10× improvement in the space-bandwidth product compared to current retinal imaging systems.

High clutter and low signal mean that existing retinal-imaging techniques are not well suited to measuring retinal metabolism. We describe spectral/lifetime phasor measurement of fluorescent biomarkers as a route to routine assessment of retinal health.