Spectral ghost imaging camera with super-Rayleigh modulator

To solve the problem of poor quality in ghost imaging via sparsity constraints (GISC) multispectral image reconstruction with correlation operations and compressed sensing algorithms under low sampling rate detection conditions, we propose an end-to-end deep-learning-based method. Based on the U-Net, Res2Net-SE-Conv is employed instead of convolutional blocks to extract local and global image features at a more fine-grained level while adaptively adjusting the channel feature response. The two-dimensional contextual transformer is constructed to fully use contextual correlation information to enhance the effectiveness of feature representations. We employ the two-dimensional contextual transformer in the decoder part, dubbed CoT-Unet, to reconstruct the desired 3D cube. The results show that compared with U-Net, TSA-Net based on spatial-spectral self-attention, the PSNR of reconstructed images by the CoT-Unet is improved by 5 dB and 3 dB, respectively, SSIM is improved by 0.23 and 0.07, and SAM is decreased by 0.06 and 0.58. Compared with conventional algorithms such as DGI and CS, our method significantly improves the quality of reconstructed images. Furthermore, the comparison results at 10%, 20%, and 30% sampling rates show that our approach has the best quality in reconstructing GISC multispectral images at low sampling rates.

The image information acquisition ability of a conventional camera is usually much lower than the Shannon Limit since it does not make use of the correlation between pixels of image data. Applying a random phase modulator to code the spectral images and combining with compressive sensing (CS) theory, a spectral camera based on true thermal light ghost imaging via sparsity constraints (GISC spectral camera) is proposed and demonstrated experimentally. GISC spectral camera can acquire the information at a rate significantly below the Nyquist rate, and the resolution of the cells in the three-dimensional (3D) spectral images data-cube can be achieved with a two-dimensional (2D) detector in a single exposure. For the first time, GISC spectral camera opens the way of approaching the Shannon Limit determined by Information Theory in optical imaging instruments.

For lower sampling rate and better imaging quality, a compressive adaptive ghost imaging is proposed by adopting the sharing mechanism and fellow relationship in the wavelet tree. The sharing mechanisms, including intrascale and interscale sharing mechanisms, and fellow relationship are excavated from the wavelet tree and utilized for sampling. The shared coefficients, which are part of the approximation subband, are localized according to the parent coefficients and sampled based on the interscale sharing mechanism and fellow relationship. The sampling rate can be reduced owing to the fact that some shared coefficients can be calculated by adopting the parent coefficients and the sampled sum of shared coefficients. According to the shared coefficients and parent coefficients, the proposed method predicts the positions of significant coefficients and samples them based on the intrascale sharing mechanism. The ghost image, reconstructed by the significant coefficients and the coarse image at the given largest scale, achieves better quality because the significant coefficients contain more detailed information. The simulations demonstrate that the proposed method improves the imaging quality at the same sampling rate and also achieves a lower sampling rate for the same imaging quality for different types of target object images in noise-free and noisy environments.

In response to existing compressed sensing ghost imaging (CSGI) schemes, an innovative Bayesian compressed sensing ghost imaging with better anti-noise performance is proposed, by using the sparse representation of K-Singular Value Decomposition (KSVD) and a 3Level (3L)-hierarchical variational message passing (VMP) algorithm. Simulation and experimental results confirm that, this innovative method overcomes the limitations of presetting specific parameters (sparsity, noise level, etc.), and also demonstrates superior performance in terms of reconstruction accuracy and imaging quality, especially for highly complex objects, where it effectively achieves accurate imaging under varying levels of noise at a low sampling rate (below 12.2%). In addition, compared to existing Bayesian compressive sensing ghost imaging (BCSGI), our algorithm moderately reduces time consumption while ensuring high precision. Our results may provide potential applications of CSGI in the field of biomedical imaging.

An efficient and reliable MRF-based methodology for designing low-power VLSI circuits

Ghost imaging is widely used in underwater active optical imaging because of its simple structure, long distance, and non-local imaging. However, the complexity of the underwater environment will greatly reduce the imaging quality of ghost imaging. To solve this problem, an underwater ghost imaging method based on the generative adversarial networks is proposed in the study. The generator of the proposed network adopts U-Net with the double skip connections and the attention module to improve the reconstruction quality. In the network training process, the total loss function is the sum of the weighted adversarial loss, perceptual loss, and pixel loss. The experiment and simulation results show that the proposed method effectively improves the target reconstruction performance of underwater ghost imaging. The proposed method promotes the further development of active optical imaging of underwater targets based on ghost imaging technology.

Experimental investigation of the quality of lensless super-resolution ghost imaging via sparsity constraints

The existing adaptive single-pixel imaging methods suffer from a waste of sampling resources. The sampling resources are not used adequately for superior localization of significant coefficients and reconstruction. In this paper, an adaptive single-pixel imaging method via the guided coefficients in the Haar wavelet tree is proposed. The goal is to achieve high quality imaging with less sampling resources. The guided coefficients are selected from the unsampled coefficients by a proposed same-scale prediction method based on the sampled coefficients. These guided coefficients are used to localize the significant coefficients with higher resolution belonging to the sampled coefficients and the significant coefficients belonging to the guided coefficients by a proposed guided prediction method. The significant guided coefficients are then used in the composite reconstruction method to reconstruct the image. Performance analysis shows that the proposed method reduces waste of the sampling resources and localizes more significant coefficients. Simulation results demonstrate that the proposed method improves the imaging quality in terms of peak signal-to-noise ratio up to 29.7 dB for the images containing regular and chaotic textures in the noise-free environment. The sampling rate for the same imaging quality can be reduced up to 56%. Under the noisy condition, the proposed method also achieves better imaging quality at a lower sampling rate.

Orthogonal sparse dictionary based on Chirp echo for ultrasound imaging

To get a better understanding of people's mobility, cellular network signalling data including location information, is a promising large-scale data source. In order to estimate travel demand and infrastructure usage from the data, it is necessary to identify the trips users make. We present two trip extraction methods and compare their performance using a small dataset collected in Sweden. The trips extracted are compared with GPS tracks collected on the same mobiles. Despite the much lower location sampling rate in the cellular network signalling data, we are able to detect most of the trips found from GPS data. This is promising, given the relative simplicity of the algorithms. However, further investigation is necessary using a larger dataset and more types of algorithms. By applying the same methods to a second dataset for Senegal with much lower sampling rate than the Sweden dataset, we show that the choice of the trip extraction method tends to be even more important when the sampling rate is low.

.SignificanceOptical properties (absorption coefficient and scattering coefficient) of tissue are the most critical parameters for disease diagnosis-based optical method. In recent years, researchers proposed spatial frequency domain imaging (SFDI) to quantitatively map tissue optical properties in a broad field of contactless imaging. To solve the limitations in wavebands unsuitable for silicon-based sensor technology, a compressed sensing (CS) algorithm is used to reproduce the original signal by a single-pixel detectors. Currently, the existing single-pixel SFDI method mainly uses a random sampling policy to extract and recover signals in the acquisition stage. However, these methods are memory-hungry and time-consuming, and they cannot generate discernible results under low sampling rate. Explorations on high performance and efficiency single-pixel SFDI are of great significance for clinical application.AimFourier single-pixel imaging can reconstruct signals with less time and space costs and has fewer reconstruction errors. We focus on an SFDI algorithm based on Fourier single-pixel imaging and propose our Fourier single-pixel image-based spatial frequency domain imaging method (FSI-SFDI).ApproachFirst, we use Fourier single-pixel imaging algorithm to collect and compress signals and SFDI algorithm to generate optical parameters. Given the basis that the main energy of general image signals is concentrated in the range of low frequency of Fourier frequency domain, our FSI-SFDI uses a circular-sampling scheme to sample data points in the low-frequency region. Then, we reconstruct the image details from these points by optimization-based inverse-FFT method.ResultsOur algorithm is tested on simulated data. Results show that the root mean square error (RMSE) of optical parameters is lower than 5% when the data reduction is 92%, and it can generate discernible optical parameter image with low sampling rate. We can observe that our FSI-SFDI primarily recovers the optical properties while keeping the RMSE under the upper bound of 4.5% when we use an image with resolution as the example for calculation and analysis. Not only that but also our algorithm consumes less space and time for an image with resolution, the signal reconstruction takes only 1.65 ms, and requires less RAM memory. Compared to CS-SFDI method, our FSI-SFDI can reduce the required number of measurements through optimizing algorithm.ConclusionsMoreover, FSI-SFDI is capable of recovering high-quality resolvable images with lower sampling rate, higher-resolution images with less memory and time consumed than previous CS-SFDI method, which is very promising for clinical data collection and medical analysis.

Aiming at the problems of large storage space, blurred the reconstructed image and poor reconstructed effect under low sampling rate while reconstructing an image with the orthogonal matching pursuit algorithm. A variable sampling compressed sensing reconstruction algorithm based on texture information is proposed. The algorithm divides the image into blocks first, distributes the sampling frequency adaptively according to the texture information and sparsity of the image blocks, and smoothes the reconstructed image using median filtering algorithm to reduce the effect of the reconstructed image blocks. The simulation results show that the subjective and objective reconstruction effects of the algorithm in this paper are better than other reconstructed algorithms under the same conditions, and the PSNR value is 2~4 dB higher than other reconstructed algorithms at a low sampling rate, and the higher reconstruction quality is achieved at the low sampling rate.

Microwave-induced thermal acoustic tomography (MITAT) is an innovative technique to image biomedical tissues based on their electric properties. It has the advantages of both high contrast and high spatial resolution. Image reconstruction method in MITAT is always a critical issue. In this paper, a CS-MITAT (CS: compressive sensing) imaging method is proposed. Compressive sensing (CS) is a recently developed sparse signal representation and analysis framework which handles medical imaging measurements using low sampling rate or increasing imaging quality. The CS-MITAT imaging method applies CS theory to the MITAT for breast tumor imaging. In this method, an over-complete dictionary is established to make sparse measurements in the spatial domain. This treatment greatly saves measurement time. Simulations and experiments with real breast tumor tissues demonstrate the feasibility and effectiveness of the method. Compared with conventional time reversal mirror method which has been used in MITAT research, CS-MITAT provides the same peak signal-to-noise ratio imaging quality by using significantly fewer acoustic sensor positions or scanning times.

In this paper, self-modulated ghost imaging (SMGI) in a surrounded scattering medium is proposed. Different from traditional ghost imaging, SMGI can take advantage of the dynamic scattering medium that originally affects the imaging quality and generate pseudo-thermal light through the dynamic scattering of free particles’ Brownian motion in the scattering environment for imaging. Theoretical analysis and simulation were used to establish the relationship between imaging quality and particle concentration. An experimental setup was also built to verify the feasibility of the SMGI. Compared with the reconstructed image quality and evaluation indexes of traditional ghost imaging, SMGI has better image quality, which demonstrates a promising future in dynamic high-scattering media such as dense fog and turbid water.

Contingency Estimation of States for Unmanned Aerial Vehicle using a Spherical Simplex Unscented Filter