Current Reconstruction Algorithms Research Articles

Bubble boundary R-CNN: A multitask model for segmenting and reconstructing overlapping bubbles

When measuring the dynamic characteristics of bubbles using image-based methods, the images including complex overlapping of bubbles with irregular shape pose a huge challenge to current bubble reconstruction algorithms. Based on the Mask R-CNN architecture, a multitask model “Bubble Boundary R-CNN” is proposed for detection, segmentation and shape reconstruction of overlapping bubbles with high deformation at high gas velocities. By leveraging shared deep features from detection and segmentation, model achieves flexible bubble reconstruction and accurate size distribution extraction from high gas velocity bubble flows. To overcome the limitations of Mask R-CNN and address the challenges posed by high-deformation and overlapping bubbles, a semantic segmentation auxiliary head and the Boundary-preserving Mask Head (BMask Head) are introduced to enhance edge information and extract bubble boundary features. Additionally, an Overlapping Boundary Prediction Network (OBPN) is proposed to reconstruct boundary features that are lost due to bubble occlusion and predict the complete bubble shapes. The optimized model shows its ability to flexibly and accurately reconstruct overlapped and deformed bubbles by varying occlusion rates and circularity. The new model has also proven its reliability when applied to bubble images from a bubble column at gas velocities of up to 0.067 m3/s. In future work, more efforts are needed to further enhance the detection capability of the model and reduce missed and multiple detections at high gas velocity.

PLIC-Net: A machine learning approach for 3D interface reconstruction in volume of fluid methods

The accurate reconstruction of immiscible fluid–fluid interfaces from the volume fraction field is a critical component of geometric Volume of Fluid methods. A common strategy is the Piecewise Linear Interface Calculation (PLIC), which fits a plane in each mixed-phase computational cell. However, recent work goes beyond PLIC by using two planes or even a paraboloid. To select such planes or paraboloids, complex optimization algorithms as well as carefully crafted heuristics are necessary. Yet, the potential exists for a well-trained machine learning model to efficiently provide broadly applicable solutions to the interface reconstruction problem at lower costs. In this work, the viability of a machine learning approach is demonstrated in the context of a single plane reconstruction. A feed-forward deep neural network is used to predict the normal vector of a PLIC plane given volume fraction and phasic barycenter data in a 3×3×3 stencil. The PLIC plane is then translated in its cell to ensure exact volume conservation. Our proposed neural network PLIC reconstruction (PLIC-Net) is equivariant to reflections about the Cartesian planes. Training data is analytically generated with O(106) randomized paraboloid surfaces, which allows for the sampling a broad range of interface shapes. PLIC-Net is tested in multiphase flow simulations where it is compared to standard LVIRA and ELVIRA reconstruction algorithms, and the impact of training data statistics on PLIC-Net’s performance is also explored. It is found that PLIC-Net greatly limits the formation of spurious planes and generates cleaner numerical break-up of the interface. Additionally, the computational cost of PLIC-Net is lower than that of LVIRA and ELVIRA. These results establish that machine learning is a viable approach to Volume of Fluid interface reconstruction and is superior to current reconstruction algorithms for some cases.

A Super-Resolution Reconstruction Model for Remote Sensing Image Based on Generative Adversarial Networks

In current times, reconstruction of remote sensing images using super-resolution is a prominent topic of study. Remote sensing data have a complex spatial distribution. Compared with natural pictures, remote sensing pictures often contain subtler and more complicated information. Most super-resolution reconstruction algorithms cannot restore all the information contained in remote sensing images when reconstructing them. The content of some areas in the reconstructed images may be too smooth, and some areas may even have color changes, resulting in lower quality reconstructed images. In response to the problems presenting in current reconstruction algorithms about super-resolution, this article proposes the SRGAN-MSAM-DRC model (SRGAN model with multi-scale attention mechanism and dense residual connection). This model roots in generative adversarial networks and incorporates multi-scale attention mechanisms and dense residual connections into the generator. Furthermore, residual blocks are incorporated into the discriminator. We use some remote sensing image datasets of real-world data to evaluate this model, and the results indicate the SRGAN-MSAM-DRC model has shown enhancements in three evaluation metrics for reconstructed images about super-resolution. Compared to the basic SRGAN model, the SSIM (structural similarity), PSNR (peak signal-to-noise ratio), and IE (image entropy) increase by 5.0%, 4.0%, and 4.1%, respectively. From the results, we know the quality of the reconstructed images of remote sensing using the SRGAN-MSAM-DRC model is better than basic SRGAN model, and verifies that the model has good applicability and performance in reconstruction of remote sensing images using super-resolution.

Undulator linear taper control at the European X-Ray Free-Electron Laser facility

Undulator tapering controls the resonance properties of the free-electron laser (FEL) amplification process. Wakefield energy losses in an undulator’s vacuum chamber are one of the factors that determine the undulator’s linear taper. While another contribution to energy losses, namely the losses due to spontaneous radiation, can be calculated analytically, estimating wakefield energy losses requires detailed knowledge of the chamber geometry and the electron beam current profile. We introduce a method for the automatic estimation of wakefield energy losses, which leverages noninvasive THz diagnostics, a current profile reconstruction algorithm enhanced with machine learning, and a recently developed analytical wakefield function for the European XFEL’s undulator beamline. The correctness of this method was validated by directly measuring wakefield-induced electron beam energy losses in the undulator section. This, in turn, enables the prediction of the optimal linear taper in the undulator. Published by the American Physical Society 2024

Evaluation of 10 current image reconstruction algorithms for linear array photoacoustic imaging.

Various reconstruction algorithms have been implemented for linear array photoacoustic imaging systems with the goal of accurately reconstructing the strength absorbers within the tissue being imaged. Since the existing algorithms have been introduced by different research groups and the context of performance evaluation was not consistent, it is difficult to make a fair comparison between them. In this study, we systematically compared the performance of 10 published image reconstruction algorithms (DAS, UBP, pDAS, DMAS, MV, EIGMV, SLSC, GSC, TR, and FD) using in-vitro phantom data. Evaluations were conducted based on lateral resolution of the reconstructed images, computational time, target detectability, and noise sensitivity. We anticipate the outcome of this study will assist researchers in selecting appropriate algorithms for their linear array PA imaging applications.

Lensless Image Restoration Based on Multi-Stage Deep Neural Networks and Pix2pix Architecture

Lensless imaging represents a significant advancement in imaging technology, offering unique benefits over traditional optical systems due to its compact form factor, ideal for applications within the Internet of Things (IoT) ecosystem. Despite its potential, the intensive computational requirements of current lensless imaging reconstruction algorithms pose a challenge, often exceeding the resource constraints typical of IoT devices. To meet this challenge, a novel approach is introduced, merging multi-level image restoration with the pix2pix generative adversarial network architecture within the lensless imaging sphere. Building on the foundation provided by U-Net, a Multi-level Attention-based Lensless Image Restoration Network (MARN) is introduced to further augment the generator’s capabilities. In this methodology, images reconstructed through Tikhonov regularization are perceived as degraded images, forming the foundation for further refinement via the Pix2pix network. This process is enhanced by incorporating an attention-focused mechanism in the encoder--decoder structure and by implementing stage-wise supervised training within the deep convolutional network, contributing markedly to the improvement of the final image quality. Through detailed comparative evaluations, the superiority of the introduced method is affirmed, outperforming existing techniques and underscoring its suitability for addressing the computational challenges in lensless imaging within IoT environments. This method can produce excellent lensless image reconstructions when sufficient computational resources are available, and it consistently delivers optimal results across varying computational resource constraints. This algorithm enhances the applicability of lensless imaging in applications such as the Internet of Things, providing higher-quality image acquisition and processing capabilities for these domains.

Review of Ultrasonic Particle Manipulation Techniques: Applications and Research Advances.

Ultrasonic particle manipulation technique is a non-contact label-free method for manipulating micro- and nano-scale particles using ultrasound, which has obvious advantages over traditional optical, magnetic, and electrical micro-manipulation techniques; it has gained extensive attention in micro-nano manipulation in recent years. This paper introduces the basic principles and manipulation methods of ultrasonic particle manipulation techniques, provides a detailed overview of the current mainstream acoustic field generation methods, and also highlights, in particular, the applicable scenarios for different numbers and arrangements of ultrasonic transducer devices. Ultrasonic transducer arrays have been used extensively in various particle manipulation applications, and many sound field reconstruction algorithms based on ultrasonic transducer arrays have been proposed one after another. In this paper, unlike most other previous reviews on ultrasonic particle manipulation, we analyze and summarize the current reconstruction algorithms for generating sound fields based on ultrasonic transducer arrays and compare these algorithms. Finally, we explore the applications of ultrasonic particle manipulation technology in engineering and biological fields and summarize and forecast the research progress of ultrasonic particle manipulation technology. We believe that this review will provide superior guidance for ultrasonic particle manipulation methods based on the study of micro and nano operations.

A probabilistic framework for uncertainty quantification in positron emission particle tracking

Positron emission particle tracking (PEPT) is an imaging method for the visualization of fluid motion, capable of reconstructing three-dimensional trajectories of small tracer particles suspended in nearly any medium, including fluids that are opaque or contained within opaque vessels. The particles are labeled radioactively, and their positions are reconstructed from the detection of pairs of back-to-back photons emitted by positron annihilation. Current reconstruction algorithms are heuristic and typically based on minimizing the distance between the particles and the so-called lines of response (LoRs) joining the detection points, while accounting for spurious LoRs generated by scattering. Here we develop a probabilistic framework for the Bayesian inference and uncertainty quantification of particle positions from PEPT data. We formulate a likelihood by describing the emission of photons and their noisy detection as a Poisson process in the space of LoRs. We derive formulas for the corresponding Poisson rate in the case of cylindrical detectors, accounting for both undetected and scattered photons. We illustrate the formulation by quantifying the uncertainty in the reconstruction of the position of a single particle on a circular path from data generated by state-of-the-art Monte Carlo simulations. The results show how the observation time can be chosen optimally to balance the need for a large number of LoRs with the requirement of small particle displacement imposed by the assumption that the particle is static over . We further show how this assumption can be relaxed by inferring jointly the position and velocity of the particle, with clear benefits for the accuracy of the reconstruction.

Double total variation (DTV) regularization and Improved adaptive moment estimation (IADAM) optimization method for fast MR image reconstruction

Background and ObjectiveCompressed sensing has been extensively studied as an advanced technique for fast MR image reconstruction. Current reconstruction algorithms often use total variation as the regularization term. Traditional total variation can easily lead to a staircase effect because it only pays attention to the variational information of the horizontal and vertical subbands. MethodsIn this paper, we propose a novel algorithm to reduce the staircase effect by increasing the variational information of the two diagonal subbands, which named Double Total Variation (DTV). We optimize the conjugate gradient algorithm by Improved Adaptive Moment Estimation (IADAM) as the solution algorithm. ResultsMR images of three body parts (head, knee and ankle) were used for simulations under different acceleration factor conditions. The conjugate gradient and fast conjugate gradient series algorithms were selected for comparison experiments. The results showed that the improved adaptive moment estimation conjugate gradient combined with DTV achieves the best reconstruction performance, therefore proved the superiority of DTV. After that, 64 different MR images of the three body parts were further simulated and the results demonstrated the general superiority from the proposed algorithm. ConclusionsThe results of this study support that the proposed method may facilitate the development of the research field of image reconstruction algorithms and provide ideas for other algorithmic improvements.

Motion Compensated Unsupervised Deep Learning for 5D MRI.

We propose an unsupervised deep learning algorithm for the motion-compensated reconstruction of 5D cardiac MRI data from 3D radial acquisitions. Ungated free-breathing 5D MRI simplifies the scan planning, improves patient comfort, and offers several clinical benefits over breath-held 2D exams, including isotropic spatial resolution and the ability to reslice the data to arbitrary views. However, the current reconstruction algorithms for 5D MRI take very long computational time, and their outcome is greatly dependent on the uniformity of the binning of the acquired data into different physiological phases. The proposed algorithm is a more data-efficient alternative to current motion-resolved reconstructions. This motion-compensated approach models the data in each cardiac/respiratory bin as Fourier samples of the deformed version of a 3D image template. The deformation maps are modeled by a convolutional neural network driven by the physiological phase information. The deformation maps and the template are then jointly estimated from the measured data. The cardiac and respiratory phases are estimated from 1D navigators using an auto-encoder. The proposed algorithm is validated on 5D bSSFP datasets acquired from two subjects.

A Graph-guided Hybrid Regularization Method For Bioluminescence Tomography

Background and objective: Bioluminescence tomography (BLT) is a powerful and sensitive imaging technique having great potential in preclinical application, such as tumor imaging, monitoring and therapy, etc. Regularization plays an important role in BLT reconstruction for considering the priori information to overcome the inherent ill-posedness of the inverse problem. Therefore, well-designed regularization term and sophisticated algorithm for solving the consequent optimization problem are key to improve the BLT quality.Methods: To balance the sparsity, smoothness and morphological characteristics of the bioluminescence targets, we constructed a novel Graph-Guided Hybrid Regularization (GGHR) method by combining graph-guided penalty term with L1 and L2 norm regularizer. To solve the corresponding minimization problem with hybrid penalties, the dual decomposition and Nesterov’s smoothing technique were adopted to decouple and transform the non-separable and non-smooth graph-guided penalty term into a differential smooth approximation form, which was solved by the fast iterative shrinkage thresholding algorithm.Results: The performance of the proposed GGHR method was verified and evaluated through a series of simulation, phantom and in vivo experiments. The comparison results demonstrated that the GGHR method outperformed current mainstream reconstruction algorithms in spatial localization, morphology recovery and in vivo practicality.Conclusions: The proposed GGHR method is a robust and practicality reconstruction algorithm for further highlighting the positive effect of hybrid regularization on BLT applications.

Three-dimensional iterative reconstruction of pulsed radiation sources using spherical harmonic decomposition.

Neutron and x-ray imaging are essential ways to diagnose a pulsed radiation source. The three-dimensional (3D) intensity distribution reconstructed from two-dimensional (2D) radiation images can significantly promote research regarding the generation and variation mechanisms of pulsed radiation sources. Only a few (≤5) projected images at one moment are available due to the difficulty in building imaging systems for high-radiation-intensity and short-pulsed sources. The reconstruction of a 3D source with a minimal number of 2D images is an ill-posed problem that leads to severe structural distortions and artifacts of the image reconstructed by conventional algorithms. In this paper, we present an iterative method to reconstruct a 3D source using spherical harmonic decomposition. Our algorithm improves the representation ability of spherical harmonic decomposition for 3D sources by enlarging the order of the expansion, which is limited in current analytical reconstruction algorithms. Prior knowledge of the source can be included to obtain a reasonable solution. Numerical simulations demonstrate that the reconstructed image quality of the iterative algorithm is better than that of the analytical algorithm. The iterative method can suppress the effect of noise in the integral projection image and has better robustness and adaptability than the analytical method.

Thermal Sensor Placement for Multicore Systems Based on Low-Complex Compressive Sensing Theory

As the complexity of the multicore system grows, the large workload diversity results in serious thermal problems. In a practical way, the number of placed thermal sensors is usually limited due to the manufacturing cost. In recent years, the compressive sensing (CS) theory is proven as an efficient way to reconstruct the original signal by using fewer sampling data. However, due to the high computational complexity during the signal reconstruction, the CS theory is not appropriate to apply to the real-time temperature monitoring in the current multicore system. In this article, we propose a grid-based sensor placement approach to placement the number-limited thermal sensors on the target multicore system. On the other hand, we adopt the matrix inversion bypass (MIB) property to reduce the computational complexity of two widely used signal reconstruction approaches in CS theory [i.e., the orthogonal matching pursuit (OMP) and stagewise OMP (StOMP)]. Due to the characteristic of random sampling in CS theory, the complexity of thermal sensor placement for multicore systems can be reduced significantly. In addition, the proposed MIB-based temperature reconstruction method helps to satisfy the requirement of real-time temperature estimation. The experimental results show that the proposed approach can reduce 57%–93% average full-system temperature reconstruction error compared with the previous non-CS-based approaches. Besides, we can also reduce 22%–41% computing latency compared with the current CS-based reconstruction algorithm. Due to the MIB-based operation, we can bypass the matrix inversion operation for temperature reconstruction. Therefore, the hardware overhead of the temperature reconstruction unit can be reduced significantly. Compared with the conventional approaches, we can reduce 24%–87% area overhead and improve 50%–220% hardware efficiency.

Research on the deep learning-based exposure invariant spectral reconstruction method.

The surface spectral reflectance of an object is the key factor for high-fidelity color reproduction and material analysis, and spectral acquisition is the basis of its applications. Based on the theoretical imaging model of a digital camera, the spectral reflectance of any pixels in the image can be obtained through spectral reconstruction technology. This technology can avoid the application limitations of spectral cameras in open scenarios and obtain high spatial resolution multispectral images. However, the current spectral reconstruction algorithms are sensitive to the exposure variant of the test images. That is, when the exposure of the test image is different from that of the training image, the reconstructed spectral curve of the test object will deviate from the real spectral to varying degrees, which will lead to the spectral data of the target object being accurately reconstructed. This article proposes an optimized method for spectral reconstruction based on data augmentation and attention mechanisms using the current deep learning-based spectral reconstruction framework. The proposed method is exposure invariant and will adapt to the open environment in which the light is easily changed and the illumination is non-uniform. Thus, the robustness and reconstruction accuracy of the spectral reconstruction model in practical applications are improved. The experiments show that the proposed method can accurately reconstruct the shape of the spectral reflectance curve of the test object under different test exposure levels. And the spectral reconstruction error of our method at different exposure levels is significantly lower than that of the existing methods, which verifies the proposed method's effectiveness and superiority.

Scintillator ageing of the T2K near detectors from 2010 to 2021

The T2K experiment widely uses plastic scintillator as a target for neutrino interactions and an active medium for the measurement of charged particles produced in neutrino interactions at its near detector complex. Over 10 years of operation the measured light yield recorded by the scintillator based subsystems has been observed to degrade by 0.9–2.2% per year. Extrapolation of the degradation rate through to 2040 indicates the recorded light yield should remain above the lower threshold used by the current reconstruction algorithms for all subsystems. This will allow the near detectors to continue contributing to important physics measurements during the T2K-II and Hyper-Kamiokande eras. Additionally, work to disentangle the degradation of the plastic scintillator and wavelength shifting fibres shows that the reduction in light yield can be attributed to the ageing of the plastic scintillator. The long component of the attenuation length of the wavelength shifting fibres was observed to degrade by 1.3–5.4% per year, while the short component of the attenuation length did not show any conclusive degradation.

Study of methods of resolved top quark reconstruction in semileptonic [formula omitted] decay: Erratum

Study of methods of resolved top quarks kinematic reconstruction in the tt̄→ℓ+jets channel is presented at the particle level as well as the fast-simulation detector level. Previous and current pseudo-top quark reconstruction algorithms are compared with suggestions presented on how to improve the reconstructed top-quark mass line shape, including the check of performance on physics observables in terms of correlations between detector, particle and parton levels, and in unfolding, with implications for current high energy physics experiments.

The impact of tube current and iterative reconstruction algorithm on dose and image quality of infant CT head examination

Computed Tomography (CT) scan examination has been recognized for its potential in contributing high radiation doses, which raised a public concern towards pediatric exposure, especially infants. Thus, the optimization is vital through well-balanced levels of radiation and image quality. Hence, this study aimed to investigate optimized CT acquisition parameter on radiosensitive organs of infant CT head examination and its influence on the image quality. This study utilized a 1-year-old anthropomorphic phantom and thermoluminescence dosimeter (TLD) for organ dose assessment. The examination was performed by using 128 multi-slice CT (Ingenuity Core, Philips, Netherlands) and varying the tube current and iterative reconstruction algorithm (iDose4) levels. During the process, the reference mAs (ref. mAs) were altered while maintaining the tube potential at 100 kVp. Five protocols were established and denoted by P1, P2, P3, P4, and P5. As per measurement, 3 calibrated TLD chips were implanted into the phantom on slab numbers 2, 4, and 7 that represented brain, lens, and thyroid, respectively. Image quality was evaluated by utilizing the objective assessment in terms of signal to noise ratio (SNR) and contrast to noise ratio (CNR) and subjective image quality using the Visual Grading Analysis (VGA) technique in terms of sharpness, noise, and artefact were assessed from each protocol. Notably, the dose was decreased up to 30% along with the reduction of tube current and the improvement of SNR and CNR values for parameters that enable the iterative reconstruction algorithm. Hence, optimization of CT acquisition parameter should be evaluated and monitored accordingly.

High-Speed Control of AC Servo Motor Using High-Performance RBF Neural Network Terminal Sliding Mode Observer and Single Current Reconstructed Technique

This paper proposes a phase current reconstruction strategy based on a dc bus using a single current sensor for a surface permanent magnet synchronous motor (SPMSM). The method of a single current sensor reduces the number of mechanical hall sensors and shunt resistors by using a modified current reconstruction algorithm. The information of rotor position is estimated by the sliding mode observer for its rapid response and strong anti-interference ability, and the observer needs to detect voltage and current components from α − β coordinate system. In order to reduce the buffeting problem of sliding mode observers, an adaptive neural network is introduced, by the way of extracting angle speed estimated values from sliding mode observers, and these values are trained to obtain the compensate angular velocity and minus index value to suppress speed value. The performance of this sensorless speed regulation strategy in the high-speed region using a single current sensor with an optimized adaptive neural network is verified and evaluated by PSIM simulation and experiments.

Statistical and individual characteristics-based reconstruction for craniomaxillofacial surgery.

In craniomaxillofacial (CMF) surgery planning, a preoperative reconstruction of the CMF reference model is crucial for surgical restoration, especially the reconstruction of bilateral defects. Current reconstruction algorithms mainly generate reference models from the image analysis aspect, however, clinical indicators of the CMF reference model mostly consider the distribution of anatomical landmarks. Generating a reference model with optimal clinical evaluation helps promote the feasibility of an algorithm. We first build a dataset with 100 normal skull models and then calculate a statistical shape model (SSM) and the distribution of normal cephalometric values, which indicate the statistical features of a population. To further generate personalized reference models, we apply non-rigid registration to align the SSM with the defect skull model. An evaluation standard to select the optimal reference model considers both global performance and anatomical evaluation. Moreover, we develop a landmark detection network to improve the automatic level of the algorithm. The proposed method performs better than methods including Iterative Closest Point and SSM. From a global evaluation aspect, the results show that the RMSE between the reference model and the ground truth is [Formula: see text]mm, the percentage of vertices with error below 2mm is [Formula: see text]% and the average faces distance is [Formula: see text]mm (better than the state-of-the-art method). From the anatomical evaluation aspect, the target registration error between the landmark pairs is [Formula: see text]mm. In addition, the clinical application confirms that the reference model can meet clinical requirements. To the best of our knowledge, we propose the first CMF reconstruction method considering the global performance of reconstruction and anatomically local evaluation from clinical experience. Simulated experiments and clinical cases prove the general applicability and strength of the method.

The application of convolutional neural networks for tomographic reconstruction of hyperspectral images

A novel method, utilizing convolutional neural networks (CNNs), is proposed to reconstruct hyperspectral cubes from computed tomography imaging spectrometer (CTIS) images. Current reconstruction algorithms are usually subject to long reconstruction times and mediocre precision in cases of a large number of spectral channels. The constructed CNNs deliver higher precision and shorter reconstruction time than a sparse expectation maximization algorithm. In addition, the network can handle two different types of real-world images at the same time—specifically ColorChecker and carrot spectral images are considered. This work paves the way toward real-time reconstruction of hyperspectral cubes from CTIS images.