We propose a plug-and-play (PnP) method that uses deep-learning-based denoisers as regularization priors for spectral snapshot compressive imaging (SCI). Our method is efficient in terms of reconstruction quality and speed trade-off, and flexible enough to be ready to use for different compressive coding mechanisms. We demonstrate the efficiency and flexibility in both simulations and five different spectral SCI systems and show that the proposed deep PnP prior could achieve state-of-the-art results with a simple plug-in based on the optimization framework. This paves the way for capturing and recovering multi- or hyperspectral information in one snapshot, which might inspire intriguing applications in remote sensing, biomedical science, and material science. Our code is available at: https://github.com/zsm1211/PnP-CASSI.

The applications of present nanoscopy techniques for live cell imaging are limited by the long sampling time and low emitter density. Here we developed a new single frame wide-field nanoscopy based on ghost imaging via sparsity constraints (GISC Nanoscopy), in which a spatial random phase modulator is applied in a wide-field microscopy to achieve random measurement for fluorescence signals.This new method can effectively utilize the sparsity of fluorescence emitters to dramatically enhance the imaging resolution to 80 nm by compressive sensing (CS) reconstruction for one raw image. The ultrahigh emitter density of 143 μm -2 has been achieved while the precision of single-molecule localization below 25 nm has been maintained. Thereby working with high-density of photo-switchable fluorophores GISC nanoscopy can reduce orders of magnitude sampling frames compared with previous singlemolecule localization based super-resolution imaging methods.

Ghost imaging LiDAR via sparsity constraints using push-broom scanning is proposed. It can image the stationary target scene continuously along the scanning direction by taking advantage of the relative movement between the platform and the target scene. Compared to conventional ghost imaging LiDAR that requires multiple speckle patterns staring the target, ghost imaging LiDAR via sparsity constraints using push-broom scanning not only simplifies the imaging system, but also reduces the sampling number. Numerical simulations and experiments have demonstrated its efficiency.

Coherent diffraction imaging (CDI), as a lensless imaging technique, can achieve a high-resolution image with intensity and phase information from a diffraction pattern. To capture high-speed and high-spatial-resolution scenes, we propose a temporal compressive CDI system. A two-step algorithm using physics-driven deep-learning networks is developed for multi-frame spectra reconstruction and phase retrieval. Experimental results demonstrate that our system can reconstruct up to eight frames from a snapshot measurement. Our results offer the potential to visualize the dynamic process of molecules with large fields of view and high spatial and temporal resolutions.

Ghost imaging (GI) can nonlocally image objects by exploiting the fluctuation characteristics of light fields, where the spatial resolution is determined by the normalized second-order correlation function g ( 2 ) . However, the spatial shift-invariant property of g ( 2 ) is distorted when the number of samples is limited, which hinders the deconvolution methods from improving the spatial resolution of GI. In this paper, based on prior imaging systems, we propose a preconditioned deconvolution method to improve the imaging resolution of GI by refining the mutual coherence of a sampling matrix in GI. Our theoretical analysis shows that the preconditioned deconvolution method actually extends the deconvolution technique to GI and regresses into the classical deconvolution technique for the conventional imaging system. The imaging resolution of GI after preconditioning is restricted to the detection noise. Both simulation and experimental results show that the spatial resolution of the reconstructed image is obviously enhanced by using the preconditioned deconvolution method. In the experiment, 1.4-fold resolution enhancement over Rayleigh criterion is achieved via the preconditioned deconvolution. Our results extend the deconvolution technique that is only applicable to spatial shift-invariant imaging systems to all linear imaging systems, and will promote their applications in biological imaging and remote sensing for high-resolution imaging demands.

No abstract

A spectral camera based on ghost imaging via sparsity constraints (GISC) acquires a three-dimensional (3D) spatial-spectral data cube of the target through a two-dimensional (2D) detector in a single snapshot. However, the spectral and spatial resolution are interrelated because both of them are modulated by the same spatial random phase modulator. In this paper, we theoretically and experimentally demonstrate a system by equipping the GISC spectral camera with a flat-field grating to disperse the light fields before the spatial random phase modulator, hence consequently decoupling the spatial and spectral resolution. By theoretical derivation of the imaging process we obtain the spectral resolution 1nm and spatial resolution 50μm about the new system which are verified by the experiment. The new system can not only modulate the spatial and spectral resolution separately, but also provide a possibility of optimizing the light field fluctuations of different wavelengths according to the imaging scene.

Spectral camera based on ghost imaging via sparsity constraints (GISC spectral camera) obtains three-dimensional (3D) hyperspectral information with two-dimensional (2D) compressive measurements in a single shot, which has attracted much attention in recent years. However, its imaging quality and real-time performance of reconstruction still need to be further improved. Recently, deep learning has shown great potential in improving the reconstruction quality and reconstruction speed for computational imaging. When applying deep learning into GISC spectral camera, there are several challenges need to be solved: 1) how to deal with the large amount of 3D hyperspectral data, 2) how to reduce the influence caused by the uncertainty of the random reference measurements, 3) how to improve the reconstructed image quality as far as possible. In this paper, we present an end-to-end V-DUnet for the reconstruction of 3D hyperspectral data in GISC spectral camera. To reduce the influence caused by the uncertainty of the measurement matrix and enhance the reconstructed image quality, both differential ghost imaging results and the detected measurements are sent into the network's inputs. Compared with compressive sensing algorithm, such as PICHCS and TwIST, it not only significantly improves the imaging quality with high noise immunity, but also speeds up the reconstruction time by more than two orders of magnitude.