We report an imaging method, termed Fourier ptychographic microscopy (FPM), which iteratively stitches together a
number of variably illuminated, low-resolution intensity images in Fourier space to produce a wide-field, high-resolution
complex sample image. By adopting a wavefront correction strategy, the FPM method can also correct for aberrations and
digitally extend a microscope’s depth of focus beyond the physical limitations of its optics. As a demonstration, we built a
microscope prototype with a resolution of 0.78 µm, a field of view of ∼120 mm^2 and a resolution-invariant depth of focus
of 0.3 mm (characterized at 632 nm). Gigapixel colour images of histology slides verify successful FPM operation. The
reported imaging procedure transforms the general challenge of high-throughput, high-resolution microscopy from one
that is coupled to the physical limitations of the system’s optics to one that is solvable through computation.

Wide-field, high-resolution Fourier ptychographic microscopy

Quantitative phase imaging, i.e., measuring the map of pathlength shifts due to the specimen of interest, has been developing rapidly over the past decade. The main methods and exciting applications to biomedicine will be reviewed.

Quantitative phase imaging in biomedicine

Transport of intensity equation: a tutorial

Holography is one of the most attractive approaches for reconstructing optical images, due to its capability of recording both the amplitude and phase information on light scattered from objects. Recently, optical metasurfaces for manipulating the wavefront of light with well-controlled amplitude, phase, and polarization have been utilized to reproduce computer-generated holograms. However, the currently available metasurface holograms have only been designed to achieve limited colors and record either amplitude or phase information. This fact significantly limits the performance of metasurface holograms to reconstruct full-color images with low noise and high quality. Here, we report the design and realization of ultrathin plasmonic metasurface holograms made of subwavelength nanoslits for reconstructing both two- and three-dimensional full-color holographic images. The wavelength-multiplexed metasurface holograms with both amplitude and phase modulations at subwavelength scale can faithfully produce not...

Full-Color Plasmonic Metasurface Holograms

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.

/pdf/phase-imaging-with-an-untrained-neural-network-225tm5sdo3.pdf

Phase imaging with an untrained neural network.

Phase unwrapping is an important but challenging issue in phase measurement. Even with the research efforts of a few decades, unfortunately, the problem remains not well solved, especially when heavy noise and aliasing (undersampling) are present. We propose a database generation method for phase-type objects and a one-step deep learning phase unwrapping method. With a trained deep neural network, the unseen phase fields of living mouse osteoblasts and dynamic candle flame are successfully unwrapped, demonstrating that the complicated nonlinear phase unwrapping task can be directly fulfilled in one step by a single deep neural network. Excellent anti-noise and anti-aliasing performances outperforming classical methods are highlighted in this paper.

One-step robust deep learning phase unwrapping.

Theoretical analysis shows that, to improve the resolution and the range of the field of view of the reconstructed image in digital lensless Fourier transform holography, an effective solution is to increase the area and the pixel number of the recorded digital hologram. A new approach based on the synthetic aperture technique and use of linear CCD scanning is presented to obtain digital holographic images with high resolution and a wide field of view. By using a synthetic aperture technique and linear CCD scanning, we obtained digital lensless Fourier transform holograms with a large area of 3.5 cm×3.5 cm (5000×5000 pixels). The numerical reconstruction of a 4 mm object at a distance of 14 cm by use of a Rayleigh-Sommerfeld integral shows that a theoretically minimum resolvable distance of 2.57 μm can be achieved at a wavelength of 632.8 nm. The experimental results are consistent with the theoretical analysis.

High resolution digital holographic microscopy with a wide field of view based on a synthetic aperture technique and use of linear CCD scanning

The numerical recording and reconstruction of a color holographic image are achieved by using digital lensless Fourier transform holography. Firstly, for a color object, three monochromatic digital holograms with different wavelengths (red, green, blue) are recorded by a black-white CCD, respectively. Then the reconstructed monochromatic holographic images (red, green, blue) are adjusted to be same in size through padding digital holograms with zeros, and the corresponding digital color holographic image is acquired by accurately syncretizing the resized reconstructed monochromatic images. One of the advantages using lensless Fourier transform holography is that it can well assure the precise superposition of the reconstructed images. By applying median filtering technique and superposing the speckle fields with different distributions, the speckle noises are well suppressed and the quality of the digital color holographic image is greatly improved. This digital color holography with high quality of reconstruction effect would have potential applications on digital holographic display of color objects.

Recording and reconstruction of a color holographic image by using digital lensless Fourier transform holography

In this Letter, for the first time, to the best of our knowledge, we propose a digital holographic reconstruction method with a one-to-two deep learning framework (Y-Net). Perfectly fitting the holographic reconstruction process, the Y-Net can simultaneously reconstruct intensity and phase information from a single digital hologram. As a result, this compact network with reduced parameters brings higher performance than typical network variants. The experimental results of the mouse phagocytes demonstrate the advantages of the proposed Y-Net.

Jianglei Di

Papers

One-step robust deep learning phase unwrapping.

High resolution digital holographic microscopy with a wide field of view based on a synthetic aperture technique and use of linear CCD scanning

Recording and reconstruction of a color holographic image by using digital lensless Fourier transform holography

Y-Net: a one-to-two deep learning framework for digital holographic reconstruction.

Phase aberration compensation of digital holographic microscopy based on least squares surface fitting