Explore the vast range of titles available.

Most of the neural networks proposed so far for computational imaging (CI) in optics employ a supervised training strategy, and thus need a large training set to optimize their weights and biases. Setting aside the requirements of environmental and system stability during many hours of data acquisition, in many practical applications, it is unlikely to be possible to obtain sufficient numbers of ground-truth images for training. Here, we propose to overcome this limitation by incorporating into a conventional deep neural network a complete physical model that represents the process of image formation. The most significant advantage of the resulting physics-enhanced deep neural network (PhysenNet) is that it can be used without training beforehand, thus eliminating the need for tens of thousands of labeled data. We take single-beam phase imaging as an example for demonstration. We experimentally show that one needs only to feed PhysenNet a single diffraction pattern of a phase object, and it can automatically optimize the network and eventually produce the object phase through the interplay between the neural network and the physical model. This opens up a new paradigm of neural network design, in which the concept of incorporating a physical model into a neural network can be generalized to solve many other CI problems.

Scattering media, such as diffused glass and biological tissue, are usually treated as obstacles in imaging. To cope with the random phase introduced by a turbid medium, most existing imaging techniques recourse to either phase compensation by optical means or phase recovery using iterative algorithms, and their applications are often limited to two-dimensional imaging. In contrast, we utilize the scattering medium as an unconventional imaging lens and exploit its lens-like properties for lensless three-dimensional (3D) imaging with diffraction-limited resolution. Our spatially incoherent lensless imaging technique is simple and capable of variable focusing with adjustable depths of focus that enables depth sensing of 3D objects that are concealed by the diffusing medium. Wide-field imaging with diffraction-limited resolution is verified experimentally by a single-shot recording of the 1951 USAF resolution test chart, and 3D imaging and depth sensing are demonstrated by shifting focus over axially separated objects. Scattering media: lensless imaging By exploiting the lens-like properties of a scattering medium, scientists have imaged objects concealed by a diffuse material. Alok Kumar Singh and co-workers from the University of Stuttgart in Germany performed imaging based on speckle intensity correlation but with the important addition of a reference point source near the object to be imaged. The object is illuminated with spatially incoherent light, which is scattered by a suitable medium, such as a diffuser or piece of chicken breast. An image sensor then detects the resulting two-dimensional speckle intensity patterns. Experiments with a USAF standard test target demonstrated that the approach can deliver diffraction-limited spatial resolution. Three-dimensional imaging is possible by capturing a series of speckle patterns with the image sensor at different axial positions.

Microlens array-based light-field imaging has been one of the most commonly used and effective technologies to record high-dimensional optical signals for developing various potential high-performance applications in many fields. However, the use of a microlens array generally suffers from an intrinsic trade-off between the spatial and angular resolutions. In this paper, we concentrate on exploiting a diffuser to explore a novel modality for light-field imaging. We demonstrate that the diffuser can efficiently angularly couple incident light rays into a detected image without needing any lens. To characterize and analyse this phenomenon, we establish a diffuser-encoding light-field transmission model, in which four-dimensional light fields are mapped into two-dimensional images via a transmission matrix describing the light propagation through the diffuser. Correspondingly, a calibration strategy is designed to flexibly determine the transmission matrix, so that light rays can be computationally decoupled from a detected image with adjustable spatio-angular resolutions, which are unshackled from the resolution limitation of the sensor. The proof-of-concept approach indicates the possibility of using scattering media for lensless four-dimensional light-field recording and processing, not just for two- or three-dimensional imaging.Imaging: Lensless light-field imaging lays foundations for new applicationsBy replacing lenses with a diffuser in light-field imaging, scientists could expand its use in applications ranging from light-field microscopy and synthetic aperture imaging to visual odometry. Microlens array-based light-field imaging is one of the most commonly used technologies to record high-dimensional optical images. However, they suffer from a trade-off between the spatial and angular resolutions. Xiaoli Liu and colleagues from Shenzhen University in China, in collaboration with researchers from the University of Stuttgart in Germany, have now developed a novel technique that uses a diffuser to angularly couple incident light rays into a detected image without the need for a lens. The diffuser allows each sub-beam emitted by a point source to form a distinguishable sub-image covering a region on the sensor and could lead to lensless four-dimensional light-field imaging.

Large depth of field (DOF) is a longstanding goal in optical imaging field. In this paper we presented a simple but efficient method to extend the DOF of a diffraction-limited imaging system using a thin scattering diffuser. The DOF characteristic of the imaging system with random phase modulation was analyzed based on the analytical model of ambiguity function as a polar display of the optical transfer function (OTF). The results of numerical simulation showed that more high-frequency components existed in the defocused OTF curve when the exit pupil of the imaging system exhibited a random phase modulation. It proved the important role of the scattering diffuser in extending the DOF of imaging systems. For the reconstruction, a stack of point spread functions (PSFs) corresponding to different axial locations within a measurement range were superimposed to construct the stacked PSF. Then the large DOF image was recovered from a speckle pattern by deconvolution. In this proof-of-concept, we experimentally demonstrated the single-shot imaging with larger DOF using a thin glass scattering diffuser in both a single-lens imaging system and a microscopic imaging system.

Imaging through scattering media is one of the main challenges in optics while the deep learning (DL) technique is well known as one of the promising ways to handle it. However, most of the existing DL approaches for imaging through scattering media adopt the end-to-end strategy, which significantly limits its generalization capability for various or dynamic scattering media. In this work, we propose an alternative DL-based method to achieve the goal of imaging through different scattering media under the framework of off-axis digital holography. As a result, the severe ill-posed inverse problem in scattering imaging is simplified as a relatively easy denoising issue for a deteriorated hologram. The experimental results of the proposed method show good generalization for not only different scattering media but also different types of objects.

In recent years, optical interferometry-based techniques have been widely used to perform noncontact measurement of dynamic deformation in different industrial areas. In these applications, various physical quantities need to be measured in any instant and the Nyquist sampling theorem has to be satisfied along the time axis on each measurement point. Two types of techniques were developed for such measurements: one is based on high-speed cameras and the other uses a single photodetector. The limitation of the measurement range along the time axis in camera-based technology is mainly due to the low capturing rate, while the photodetector-based technology can only do the measurement on a single point. In this paper, several aspects of these two technologies are discussed. For the camera-based interferometry, the discussion includes the introduction of the carrier, the processing of the recorded images, the phase extraction algorithms in various domains, and how to increase the temporal measurement range by using multiwavelength techniques. For the detector-based interferometry, the discussion mainly focuses on the single-point and multipoint laser Doppler vibrometers and their applications for measurement under extreme conditions. The results show the effort done by researchers for the improvement of the measurement capabilities using interferometry-based techniques to cover the requirements needed for the industrial applications.

Scattering media have always been looked upon as an obstacle in imaging. Various methods, ranging from holography to phase compensation as well as to correlation techniques, have been proposed to cope with this obstacle. We, on the other hand, have a different understanding about the role of the diffusing media. In this paper we propose and demonstrate a ‘scatter-plate microscope’ that utilizes the diffusing property of the random medium for imaging micro structures with diffraction-limited resolution. The ubiquitous property of the speckle patterns permits to exploit the scattering medium as an ultra-thin lensless microscope objective with a variable focal length and a large working distance. The method provides a light, flexible and cost effective imaging device as an alternative to conventional microscope objectives. In principle, the technique is also applicable to lensless imaging in UV and X-ray microscopy. Experiments were performed with visible light to demonstrate the microscopic imaging of USAF resolution test target and a biological sample with varying numerical aperture (NA) and magnifications.

Quantitative phase imaging can reveal morphological features without having to stain the biological sample. This property has important implications for intraoperative applications since the time spent during histopathology can be reduced from a few minutes to a few seconds. However, most common quantitative phase imaging techniques are based on the interferometric principle, which makes them more prone to disturbing environmental influences, such as temperature drift and air turbulence. In the last decade, with the advance of computing power, many different iterative quantitative phase imaging techniques, which only require the recording of the diffracted wavefield, and therefore offer increased robustness towards environmental disturbances, have been proposed. These are particularly well-suited for the application outside the well-controlled lab environment such as an operating theatre. The optical performance of our developed iterative phase retrieval method based on variable wavefront curvature will be evaluated by reference to off-axis digital holography and applied for intraoperative discrimination of tissue.

In this paper, we have applied a recently developed complex-domain hyperspectral denoiser for the object recognition task, which is performed by the correlation analysis of investigated objects’ spectra with the fingerprint spectra from the same object. Extensive experiments carried out on noisy data from digital hyperspectral holography demonstrate a significant enhancement of the recognition accuracy of signals masked by noise, when the advanced noise suppression is applied.