Denoising is significant in many fields, especially for computational imaging. Coherent diffraction imaging and speckle correlation imaging are regarded as the most promising computational imaging techniques. The above two imaging techniques can be classified as phase-retrieval-based imaging due to the phase-retrieval is a vital procedure for object reconstruction. However, the acquisition process would generate unavoidable noise and participate in the iteration process of phase-retrieval. Hence, it is necessary to denoising after obtained the original reconstruction image. Here, a denoising method that based on connected domain is proposed for phase-retrieval method. We experimentally demonstrate the denoising results and quantitatively analyze the effect. Comparison of the classical median filter, wiener filter and bilateral filter, our method shows a satisfactory denoising effect. Our results prove that connected domain denoising is useful and promising, which provides a new post-processing denoising method for phase-retrieval-based imaging.
Imaging objects hidden behind opaque layers is significant in many fields, with applications ranging from biomedical imaging to defense security. Techniques based on memory-effect scattering imaging have been developed in the past decade. The existing memory-effect-based scattering imaging techniques can be divided into two categories based on the working principle of light sources. In these methods, phase-retrieval algorithm is used to reconstruct object from the power spectrum diffraction patterns as the last step. Although both of them achieve single-shot scattering imaging, the experimental set-up is quite different. It is noted that the coherent diffraction imaging is introduced to the scattering imaging field using the visible coherent light. The principle and setup of the aforementioned two methods are analyzed and summarized respectively. We experimentally demonstrate the reconstruction and evaluate the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Image Measurement (SSIM). As these technologies are limited to short range and memory effect range, the potential to imaging with wide field of view and long distance requires further exploration.
The mutual interference method is introduced to detect vortex beams. For the coaxial interference between l1-order and l2-order vortex beams, the intensity distribution is spirally fan-shaped. The partition number N=|l2-l1|, and the spiral direction can distinguish the sign of topological charge with larger absolute value between |l1| and |l2|. Fork-shaped fringes appear in the center for the small-angle interference between vortex beams with incident angle β1 and β2. The forking number between two fringes agrees with N=|l2-l1|, and the forks face upward when l1<l2 & β1<β2, or l1<l2 & β1<β2, and face downward when l1<l2 & β1<β2, or l1<l2 & β1<β2. Especially, when one of the beams is a Gaussian beam, such as l1=0, the value and sign of the topological charge l2 of the other vortex beam can be simply detected. The mutual interference method can conveniently detect the value and sign of vortex beams without borrowing redundant devices.
We analyzed the light spectrum after passing through the encoding template element. Comparing the influence of the ideal coding template elements and coding template elements with rounded errors on the far-field spectrum. The far-field spectrum has a red offset. With the changing of the aperture error value r, the deviation of the far-field spectrum relative to the light source spectrum is slightly different, and the closer the aperture is to the circle, the smaller the deviation of the spectrum, that is, the closer the light source spectrum. Five different constituent units from the two-dimensional random coding template is found, and these five constituent units are randomly distributed. The far-field spectral distribution when the beam passes through two units at the same time is numerically calculated, and the analysis shows that the spectral deviation after passing through different types of coding units are obviously different. The analysis result will provide prior information for the target recovery of compressed-sampling hyperspectral imaging, and make the target recovery more accurate.
In Computational Spectral imaging, two-dimensional coded apertures and dispersive elements realize the mixed modulation of spatial information and spectral information of the target respectively, and then reconstruct the threedimensional data cube. Therefore, coded aperture plays a vital role. In the imaging process, by moving the coded aperture to increase the number of measurements, the aperture moved one code element at each step to simulate the actual push-broom process. Three types of coded apertures were considered, which are Gauss random coded aperture, Hadamard coded aperture and Harmonic coded aperture, and the reconstruction effect of the three coded apertures were analyzed. The Least Square (LS) algorithm was considered to reconstruct three-dimensional data cube. Compared with the classical Two-step Iterative Shrinkage/Thresholding (TwIST) algorithm, the reconstructed Structural Similarity Index Measurement (SSIM) and Peak Signal to Noise Ratio (PSNR) by LS algorithm were better than TwIST algorithm. It was indicated that the SSIM and PSNR increased with the increasing number of measurements. When the number of measurements was similar with the number of spectral segments, the SSIM of the three coded apertures reached more than 0.9 by LS algorithm. However, the SSIM and PSNR of the Gauss random coded aperture were the largest Obviously, which are 0.995 and 52.560, respectively. And the PSNR of Gauss random coded aperture was 13 dB more than that of Hadamard and Harmonic coded apertures. When the number of measurements was constant, the SSIM and PSNR decrease gradually with the increasing number of spectral segments. The simulation results showed that the LS algorithm was superior to the TwIST algorithm in the reconstruction process, and the Gauss random coded aperture had the best performance.
Super-resolution hyperspectral imaging is a key technology for many applications, especially in the fields of remote sensing, military, agriculture, and geological exploration. Recovering a high resolution image needs enormous data, which puts forward very high requirements on image system hardware. Compressed sampling spectral imaging technology could well solve this problem and achieve high-resolution objects with low-resolution compressed data. In this paper, the method of a compressed sampling spectral imaging based on push-broom coded aperture and dispersion prism is proposed. A spectral aliasing image is formed when the object passing through the dispersive prism. According to the prism dispersion condition and the CCD pixel size, the visible spectrum can be divided into N spectral bands, and the measurement matrix of the coded aperture is respectively calibrated for the center wavelength of each spectral band. By controlling a stepper to implement the push broom of the coded aperture to change the measurement matrix, multiple spectral aliasing images can be obtained. The pixel size of the coded aperture becomes half of the CCD by a relay lens, which means the pixel of CCD is low-resolution for the coded aperture. The super-resolution hyperspectral image of the object is obtained by the improved LS reconstruction algorithm. Simulation results show that, the recovered hyperspectral image has twice resolution compared with the low-resolution CCD image, and the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) increase with the increasing compressed sampling hyperspectral images. For N=31, the average PSNR and SSIM recovered from six aliasing images is 22.019 and 0.235, respectively. The average PSNR and SSIM of the recovered 31 bands are also increasing with increasing aliasing images. While the aliasing imaging is 156, The average PSNR and SSIM exceeds 38 and 0.9. This method proves that super-resolution hyperspectral imaging can be achieved by capturing less low-resolution object images.
The encoding aperture errors with different types and different degrees occurred during the process of encoding aperture by micro-Nano technology. The encoding aperture is a key component of the CSSI, and the analysis of errors in encoding aperture processing provides an important evidence for the CSSI. In this paper, based on the error occurring in the process of encoding aperture, the simulation is established by commercial software FDTD by which the optical field modulation of incident light in the CSSI system is analyzed by comparing the ideal encoding aperture and the error encoding aperture. The simulation results show that there is a significant difference in the optical intensity distribution of incident light modulated by a single error aperture and a single ideal aperture, the optical intensity distribution modulated by the ideal aperture has two distinct peaks at the aperture surface, and the optical intensity distribution modulated by the error aperture is approximately twice as large as by the ideal aperture; the optical intensity distribution modulated by the two type aperture has obvious peaks while leaving the aperture surface, and the optical intensity distribution modulated by the ideal aperture is approximately twice as large as by the error aperture; changing the number of pixels of the encoding aperture, the ideal encoding aperture and the error encoding aperture have little difference in modulation of the incident light; comparing the ideal aperture, as the increasing of the rounded radius of error aperture, the influence of the optical field distribution modulation becomes more obvious.
With a piece of far-field diffraction image, the purpose of reconstruction an object can be achieved by the Coherent Diffraction Imaging (CDI) method under some certain conditions. Practically, the far-field diffraction images captured by the optical system are not always matched well with the phase retrieval algorithms, which frequently leads to lower resolution of the reconstructed object image. However, we find that the gray distortion of the power spectrum has a great impact on the object reconstruction, and even a good phase retrieve algorithm can not reconstruct the object. Based on experiment and simulation results, we find that the spatial power spectrum pattern gray-level distortion has much influence on the CDI reconstruction, and the acquired pattern distortion rate should be less than 0.1. When the gray-level distortion is less than 0.1, clear object can be reconstructed in fewer iterations. The reconstruction algorithm is fault-tolerant to the distortion of power spectrum. The convergence speed of the algorithm can be accelerated through giving an upper bound of gray-level distortion. This result provides a reference for other researches in CDI to avoid the convergence stagnation caused by the distortion of spatial power spectrum collected by experiments.
By iteratively stitching together the series of low-resolution (LR) images captured by either various small-aperture illuminations or angle-varied illuminations, the Fourier ptychography (FP) can recover large space-bandwidth- product (SBP) and high-resolution (HR) object images. The FP has been considered to be promising in various computational imaging fields. However, the illumination-based FP is limited by strict requirements of the objects which must be thin and satisfy the one-to-one mapping relationship in the Fourier plane, and the aperture-scanning Fourier ptychography is also limited by the long-time scanning and stable scanning mechanical structures requirements even though it can achieve super-resolution macroscopic imaging. Furthermore, the position and shape of the scanning aperture must be accurately modeled for the reconstruction, otherwise false object images may be output. Herein, based on the 4-f optical correlator structure, we proposed a novel method, termed variable-aperture Fourier ptychography, for reconstructing HR images from series of LR images. The numerical simulations illustrated that the variable-aperture Fourier ptychography can use a small number of LR images to reconstruct the object images, The experiments demonstrated that a high-quality object image with better resolution and contrast than other schemes, include direct imaging based on 4-f system and aperture scanning FP, can be obtained by our method. Two additional experiments proved that it is almost unaffected by the position and shape of the apertures.
This article depicts a experiment of utilizing multi-spectral image(MSI) system, which can benefit from compressed sensing to reduce data acquisition demands, with the employment of a dispersal prism and push-broom compressive sampling system, to realize image super-resolution both in spatial and in spectral.
Under coherent light illumination, several approaches need either angle scanning or diffuser rotating to reconstruct the image through opaque scattering media. We propose a linear model to restore the hidden object through the actual power spectrum with disturbance of the scattering layer. The experimental results confirm that, the algorithm quickly converge to the only correct reconstruction solution with the accuracy power spectrum pattern of Fourier transform, and the method can reconstruct the high accuracy image of the object hidden by the scattering media with one-shot power spectrum.
We study nonlinear far-field propagation of Kerr spatial solitons along a graphene monolayer embedded planar dielectric waveguide. The volumetric permittivity approach model of graphene is introduced to incorporate this carbon atoms layer into our optical simulations, which mathematically approximate graphene by a very thin layer with a finite thickness. A remarkably large third-order nonlinear optical susceptibility of graphene measured in previous experiment is considered in the numerical simulations. We demonstrate numerically that the TE-polarized beam forms Kerr spatial solitons at high beam intensity, due to the nonlinearity of graphene compensates diffraction losses. It’s very interesting that the Kerr optical solitons can adjust the beam width when propagating to become narrower. We suppose that it’s the selfregulation of the solitons after separating a portion of energy during the transmission process to become more compact. Our simulation results also reveal that the optical field distribution of Kerr optical solitons exhibits obvious periodic oscillation along the propagating path. This is a novel phenomenon that the dynamic regulation of the light field causes spatial oscillation of the solitons and a periodic change in the effective refractive index of graphene monolayer, forming a Kerr-induced index grating in the waveguide. We emphasize that the spatial oscillation of the solitons is due to the dynamic regulation of the light field, with a process of alternating self-focusing and defocusing. We predicate that the transmittance will be improved due to the nonlinear phase modulation by the Kerr-induced index grating through the waveguide.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.