Image Reconstruction Method Research Articles

Application of binocular image reconstruction method in the construction of 3D model of wooden arch corridor bridge structure

Application of binocular image reconstruction method in the construction of 3D model of wooden arch corridor bridge structure

Colour image reconstruction in indoor space based on machine vision

In order to solve the problems of low accuracy and long time consumption in colour image reconstruction in indoor space, a colour image reconstruction method in indoor space based on machine vision is proposed. Through obtaining the maximum and minimum grey value of colour image in indoor space, the colour image features are extracted according to the brightness contrast of colour image channel; the brightness gain value is set; the colour of the base colour and the initial image are mixed to enhance the colour image in indoor space. The SDF value of the colour image in indoor space is obtained by using the voxel mapping method of machine vision algorithm, and the overlapping parts of each frame of the image are fused to realise the colour image reconstruction in indoor space. The experimental results show that the highest accuracy of the proposed method is 98%.

A Hybrid Regularization-Based Multi-Frame Super-Resolution Using Bayesian Framework

The prime purpose for the image reconstruction of a multi-frame super-resolution is to reconstruct a higher-resolution image through incorporating the knowledge obtained from a series of relevant low-resolution images, which is useful in numerous fields. Nevertheless, super-resolution image reconstruction methods are usually damaged by undesirable restorative artifacts, which include blurring distortion, noises, and stair-casing effects. Consequently, it is always challenging to achieve balancing between image smoothness and preservation of the edges inside the image. In this research work, we seek to increase the effectiveness of multi-frame super-resolution image reconstruction by increasing the visual information and improving the automated machine perception, which improves human analysis and interpretation processes. Accordingly, we propose a new approach to the image reconstruction of multi-frame super-resolution, so that it is created through the use of the regularization framework. In the proposed approach, the bilateral edge preserving and bilateral total variation regularizations are employed to approximate a high-resolution image generated from a sequence of corresponding images with low-resolution to protect significant features of an image, including sharp image edges and texture details while preventing artifacts. The experimental results of the synthesized image demonstrate that the new proposed approach has improved efficacy both visually and numerically more than other approaches.

Quantitative comparison of planar coded aperture imaging reconstruction methods

Imaging distributions of radioactive sources plays a substantial role in nuclear medicine as well as in monitoring nuclear waste and its deposit. Coded Aperture Imaging (CAI) has been proposed as an alternative to parallel or pinhole collimators, but requires image reconstruction as an extra step. Multiple reconstruction methods with varying run time and computational complexity have been proposed. Yet, no quantitative comparison between the different reconstruction methods has been carried out so far.This paper focuses on a comparison based on three sets of hot-rod phantom images captured with an experimental γ-camera consisting of a Tungsten-based MURA mask with a 2 mm thick 256 × 256 pixelated CdTe semiconductor detector coupled to a Timepix© readout circuit.Analytical reconstruction methods, MURA Decoding, Wiener Filter and a convolutional Maximum Likelihood Expectation Maximization (MLEM) algorithm were compared to data-driven Convolutional Encoder-Decoder (CED) approaches. The comparison is based on the contrast-to-noise ratio as it has been previously used to assess reconstruction quality. For the given set-up, MURA Decoding, the most commonly used CAI reconstruction method, provides robust reconstructions despite the assumption of a linear model. For single image reconstruction, however, MLEM performed best of all analytical reconstruction methods, but took on average 45 times longer than MURA Decoding. The fastest reconstruction method is the Wiener Filter with a run time 4.3 times faster compared to MURA Decoding and a mediocre quality. The CED with a specifically tailored training set was able to succeed the most commonly used MURA decoding on average by a factor between 1.37 and 2.60 and an equal run time.

Learned Reconstruction Methods With Convergence Guarantees: A survey of concepts and applications

In recent years, deep learning has achieved remarkable empirical success for image reconstruction. This has catalyzed an ongoing quest for the precise characterization of the correctness and reliability of data-driven methods in critical use cases, for instance, in medical imaging. Notwithstanding the excellent performance and efficacy of deep learning-based methods, concerns have been raised regarding the approaches’ stability, or lack thereof, with serious practical implications. Significant advances have been made in recent years to unravel the inner workings of data-driven image recovery methods, challenging their widely perceived black-box nature. In this article, we specify relevant notions of convergence for data-driven image reconstruction, which forms the basis of a survey of learned methods with mathematically rigorous reconstruction guarantees. An example that is highlighted is the role of input-convex neural networks (ICNNs), offering the possibility to combine the power of deep learning with classical convex regularization theory for devising methods that are provably convergent. This survey article is aimed at both methodological researchers seeking to advance the frontiers of our understanding of data-driven image reconstruction methods as well as practitioners by providing an accessible description of useful convergence concepts and by placing some of the existing empirical practices on a solid mathematical foundation.

Image Reconstruction of Planar Electrical Capacitance Tomography Based on DBSCAN and Self-Adaptive ADMM Algorithm

Planar electrical capacitance tomography (ECT) has broad application prospects in the defect detection of composite materials. The image reconstruction of planar ECT is a nonlinear and ill-posed inverse problem, which limits the accuracy of image reconstruction. An image reconstruction method based on the density-based spatial clustering of applications with noise cluster (DBSCAN) and self-adaptive alternating direction method of multipliers (SADMM) algorithms is proposed in this paper. Firstly, the characteristic capacitance values are extracted from the independent measurement capacitance values through the DBSCAN algorithm. Then, the original inverse problem is transformed into a set of local optimization sub-problems, and the SADMM algorithm with an adaptive penalty parameter is used to solve these subproblems to reconstruct the image. The experimental results show that the average relative error of the reconstructed images obtained by this algorithm is 0.086, which is significantly lower than other algorithms, the presented algorithm has higher anti-noise performance and can provide more accurate reconstructed images.

Particle field deconvolution multiplicative algebraic reconstruction technique for tomographic particle image velocimetry reconstruction

Precision tomographic reconstruction is critical for obtaining high-accuracy velocity measurements in tomographic particle image velocimetry. Traditional tomographic reconstruction methods, such as the multiplicative algebraic reconstruction technique (MART), can only be applied at low particle concentrations, limiting the spatial resolution of velocity measurements. In addition, the actual shape of the particles is not reconstructed well due to the limited views. In this study, we propose a novel method named particle field deconvolution MART (Deconv-MART) to repair the shape of actual particles while suppressing ghost particles reconstructed by MART iterations. This method first uses the Gaussian particle shape prior to estimate the convolution kernel obtained by MART reconstruction. Then, the estimated kernel is utilized to deconvolute the particle field and suppress ghost particles based on the prior information of the lower intensity of ghost particles as well as the sparsity of the particle field. Reconstruction fields are estimated with numerical and real experiments, and the results are compared with the results of advanced reconstruction methods. Comparisons of reconstruction demonstrate that the proposed method is effective at suppressing ghost particles and restoring the shape of actual particles. Comparisons of velocity measurements reveal that Deconv-MART has good performance and high measurement accuracy.

Evaluation of deep-learning image reconstruction for chest CT examinations at two different dose levels.

The aims of the present study were to, for both a full-dose protocol and an ultra-low dose (ULD) protocol, compare the image quality of chest CT examinations reconstructed using TrueFidelity (Standard kernel) with corresponding examinations reconstructed using ASIR-V (Lung kernel) and to evaluate if post-processing using an edge-enhancement filter affects the noise level, spatial resolution and subjective image quality of clinical images reconstructed using TrueFidelity. A total of 25 patients were examined with both a full-dose protocol and an ULD protocol using a GE Revolution APEX CT system (GE Healthcare, Milwaukee, USA). Three different reconstructions were included in the study: ASIR-V 40%, DLIR-H, and DLIR-H with additional post-processing using an edge-enhancement filter (DLIR-H + E2). Five observers assessed image quality in two separate visual grading characteristics (VGC) studies. The results from the studies were statistically analyzed using VGC Analyzer. Quantitative evaluations were based on determination of two-dimensional power spectrum (PS), contrast-to-noise ratio (CNR), and spatial resolution in the reconstructed patient images. For both protocols, examinations reconstructed using TrueFidelity were statistically rated equal to or significantly higher than examinations reconstructed using ASIR-V 40%, but the ULD protocol benefitted more from TrueFidelity. In general, no differences in observer ratings were found between DLIR-H and DLIR-H + E2. For the three investigated image reconstruction methods, ASIR-V 40% showed highest noise and spatial resolution and DLIR-H the lowest, while the CNR was highest in DLIR-H and lowest in ASIR-V 40%. The use of TrueFidelity for image reconstruction resulted in higher ratings on subjective image quality than ASIR-V 40%. The benefit of using TrueFidelity was larger for the ULD protocol than for the full-dose protocol. Post-processing of the TrueFidelity images using an edge-enhancement filter resulted in higher image noise and spatial resolution but did not affect the subjective image quality.

Temperature Field Reconstruction Method for Acoustic Tomography Based on Multi-Dictionary Learning.

A reconstruction algorithm is proposed, based on multi-dictionary learning (MDL), to improve the reconstruction quality of acoustic tomography for complex temperature fields. Its aim is to improve the under-determination of the inverse problem by the sparse representation of the sound slowness signal (i.e., reciprocal of sound velocity). In the MDL algorithm, the K-SVD dictionary learning algorithm is used to construct corresponding sparse dictionaries for sound slowness signals of different types of temperature fields; the KNN peak-type classifier is employed for the joint use of multiple dictionaries; the orthogonal matching pursuit (OMP) algorithm is used to obtain the sparse representation of sound slowness signal in the sparse domain; then, the temperature distribution is obtained by using the relationship between sound slowness and temperature. Simulation and actual temperature distribution reconstruction experiments show that the MDL algorithm has smaller reconstruction errors and provides more accurate information about the temperature field, compared with the compressed sensing and improved orthogonal matching pursuit (CS-IMOMP) algorithm, which is an algorithm based on compressed sensing and improved orthogonal matching pursuit (in the CS-IMOMP, DFT dictionary is used), the least square algorithm (LSA) and the simultaneous iterative reconstruction technique (SIRT).

Multi-weight respecification of scan-specific learning for parallel imaging

Parallel imaging is widely used in magnetic resonance imaging as an acceleration technology. Traditional linear reconstruction methods in parallel imaging often suffer from noise amplification. Recently, a non-linear robust artificial-neural-network for k-space interpolation (RAKI) exhibits superior noise resilience over other linear methods. However, RAKI performs poorly at high acceleration rates and needs a large number of autocalibration signals as the training samples. In order to tackle these issues, we propose a multi-weight method that implements multiple weighting matrices on the under-sampled data, named MW-RAKI. Enforcing multiple weighted matrices on the measurements can effectively reduce the influence of noise and increase the data constraints. Furthermore, we incorporate the strategy of multiple weighting matrixes into a residual version of RAKI, and form MW-rRAKI. Experimental comparisons with the alternative methods demonstrated noticeably better reconstruction performances, particularly at high acceleration rates. With only 12.5% of the k-space data is available, the PSNR of MW-RAKI and MW-rRAKI is improved by about 3 dB and 4 dB compared to RAKI and rRAKI, respectively.

On the performance of multi-compartment relaxometry for myelin water imaging (MCR-MWI) – test-retest repeatability and inter-protocol reproducibility

In this study, we optimized the variable flip angle (VFA) acquisition scheme using numerical simulations to shorten the acquisition time of multicompartment relaxometry for myelin water imaging (MCR-MWI) to a clinically practical range in the absence of advanced image reconstruction methods. As the primary objective of this study, the test-retest repeatability of myelin water fraction (MWF) measurements of MCR-MWI is evaluated on three gradient echo (GRE) sequence settings using the optimized VFA schemes with different echo times and repetition times, emulating various scanner setups. The cross-protocol reproducibility of MCR-MWI and MCR with diffusion-informed myelin water imaging (MCR-DIMWI) is also examined. As a secondary objective, we explore the bundle-specific profiles of various microstructural parameters from MCR-(DI)MWI and their cross-correlations to determine if these parameters possess supplementary microstructure information beyond myelin concentration.Numerical simulations indicate that MCR-MWI can be performed with a minimum of three flip angles covering a wide range of T1 weightings without adding significant bias. This is supported by the results of an in vivo experiment, allowing whole-brain 1.5 mm isotropic MWF maps to be acquired in 9 min, reducing the total scan time to 40% of the original implementation without significant quality degradation. Good test-retest repeatability is observed for MCR-MWI for all three GRE protocols. While good correlations can also be found in MWF across protocols, systematic differences are observed. Bundle-specific MWF analysis reveals that certain white matter bundles are similar in all participants. We also found that microstructure relaxation parameters have low linear correlations with MWF. MCR-MWI is a reproducible measure of myelin. However, attention should be paid to the protocol related MWF differences when comparing different studies, as the MWF bias up to 0.5% can be observed across the protocols examined in this work.

Large scale medical image online three-dimensional reconstruction based on WebGL using four tier client server architecture

The development of medical device technology has led to the rapid growth of medical imaging data. The reconstruction from two-dimensional images to three-dimensional volume visualization not only shows the location and shape of lesions from multiple views but also provides intuitive simulation for surgical treatment. However, the three-dimensional reconstruction process requires the high performance execution of image data acquisition and reconstruction algorithms, which limits the application to equipments with limited resources. Therefore, it is difficult to apply on many online scenarios, and mobile devices cannot meet high-performance hardware and software requirements. This paper proposes an online medical image rendering and real-time three-dimensional (3D) visualization method based on Web Graphics Library (WebGL). This method is based on a four-tier client-server architecture and uses the method of medical image data synchronization to reconstruct at both sides of the client and the server. The reconstruction method is designed to achieve the dual requirements of reconstruction speed and quality. The real-time 3D reconstruction visualization of large-scale medical images is tested in real environments. During the interaction with the reconstruction model, users can obtain the reconstructed results in real-time and observe and analyze it from all angles. The proposed four-tier client-server architecture will provide instant visual feedback and interactive information for many medical practitioners in collaborative therapy and tele-medicine applications. The experiments also show that the method of online 3D image reconstruction is applied in clinical practice on large scale image data while maintaining high reconstruction speed and quality.

JSENSE‐Pro: Joint sensitivity estimation and image reconstruction in parallel imaging using pre‐learned subspaces of coil sensitivity functions

To improve calibrationless parallel imaging using pre-learned subspaces of coil sensitivity functions. A subspace-based joint sensitivity estimation and image reconstruction method was developed for improved parallel imaging with no calibration data. Specifically, we proposed to use a probabilistic subspace model to capture prior information of the coil sensitivity functions from previous scans acquired using the same receiver system. Both the subspace basis and coefficient distributions were learned from a small set of training data. The learned subspace model was then incorporated into the regularized reconstruction formalism that includes a sparsity prior. The nonlinear optimization problem was solved using alternating minimization algorithm. Public fastMRI brain dataset was retrospectively undersampled by different schemes for performance evaluation of the proposed method. With no calibration data, the proposed method consistently produced the most accurate coil sensitivity estimation and highest quality image reconstructions at all acceleration factors tested in comparison with state-of-the-art methods including JSENSE, DeepSENSE, P-LORAKS, and Sparse BLIP. Our results are comparableto or even better than those from SparseSENSE, which used calibration data for sensitivity estimation. The work also demonstrated that the probabilistic subspace model learned from T2 w data can be generalized to aiding the reconstruction of FLAIR data acquired from the same receiver system. A subspace-based method named JSENSE-Pro has been proposed for accelerated parallel imaging without the acquisition of companion calibration data. The method is expected to further enhance the practical utility of parallel imaging, especially in applications where calibration data acquisition is not desirable or limited.

Deep Learning of Diffuse Optical Tomography Based on Time-Domain Radiative Transfer Equation

Near infrared diffuse optical tomography (DOT) is a potential tool for diagnosing cancer by image reconstruction of tissue optical properties. A variety of image reconstruction methods for DOT have been attempted, in general, based on the diffusion equation (DE). However, the image quality is still insufficient to clinical use, which is mainly attributed to the fact that the DE is invalid in some regions, such as low-scattering regions, and the inverse problem is inherently ill-posed. In contrast, the radiative transfer equation (RTE) accurately describes light propagation in biological tissue and also the DOT by deep learning is recently thought to be an alternative approach to the inverse problem. Distribution of time of flight (DTOF) of photons estimated by the time-domain RTE lends itself to deep learning along a temporal sequence. In this study, we propose a new DOT image reconstruction algorithm based on a long-short-term memory and the time-domain RTE. In simulation studies, using this algorithm, we succeeded in detection of an absorbing inclusion with a diameter of 5 mm, an absorber mimicking cancer, which was embedded in a two-dimensional square model (4 cm × 4 cm) with an optically homogeneous background. Multiple absorbers and a bigger absorber embedded in this model were also detected. We also demonstrate that, if simulation data by beam injection from multiple directions are employed as a training set, the accuracy of detection is improved especially for multiple absorbers.

Image Inpainting with Bilateral Convolution

Due to sensor malfunctions and poor atmospheric conditions, remote sensing images often miss important information/pixels, which affects downstream tasks, therefore requiring reconstruction. Current image reconstruction methods use deep convolutional neural networks to improve inpainting performances as they have a powerful modeling capability. However, deep convolutional networks learn different features with the same group of convolutional kernels, which restricts their ability to handle diverse image corruptions and often results in color discrepancy and blurriness in the recovered images. To mitigate this problem, in this paper, we propose an operator called Bilateral Convolution (BC) to adaptively preserve and propagate information from known regions to missing data regions. On the basis of vanilla convolution, the BC dynamically propagates more confident features, which weights the input features of a patch according to their spatial location and feature value. Furthermore, to capture different range dependencies, we designed a Multi-range Window Attention (MWA) module, in which the input feature is divided into multiple sizes of non-overlapped patches for several heads, and then these feature patches are processed by the window self-attention. With BC and MWA, we designed a bilateral convolution network for image inpainting. We conducted experiments on remote sensing datasets and several typical image inpainting datasets to verify the effectiveness and generalization of our network. The results show that our network adaptively captures features between known and unknown regions, generates appropriate content for various corrupted images, and has a competitive performance compared with state-of-the-art methods.

Impact of Various Image Reconstruction Methods on Joint Compensation of Respiratory Motion and Partial Volume Effects in Whole-Body 18F-FDG PET/CT Imaging: Patients with Non-Small Cell Lung Cancer

Impact of Various Image Reconstruction Methods on Joint Compensation of Respiratory Motion and Partial Volume Effects in Whole-Body 18F-FDG PET/CT Imaging: Patients with Non-Small Cell Lung Cancer

Image Reconstruction in Ultrasound Reflection Tomography using Quick High-Resolution Method

This paper presents an image reconstruction method for reflective ultrasonic tomography. The variation of the time corresponding to the first peak for the transmitter-receiver pairs is used in the transmission tomography. Commonly, all the reflected packets are used in the reflective reconstruction, but here we assume that the inside of the object is either changing or has a high absorption coefficient. Classical tomography methods are based on an equation where the system matrix is not square. Therefore, inverting such a matrix is a complex task. The solution to this problem is our simple geometrical method, which is very fast and accurate. Moreover, a very high spatial resolution can be achieved. In this case, the amount of information obtained from the system is limited only by the number of transducers used for the ultrasonic measurement.

Investigation and Analysis of EM Pulse Propagation Inside Human Head for High-Resolution UWB Elliptical SAR Imaging

In this paper, an ultra-wideband (UWB) microwave imaging system based on an elliptical synthetic aperture radar (ESAR) is proposed as a possible solution for human head imaging. The system along with the image reconstruction method operates in the time domain. We have adopted the global back-projection technique to an elliptical trajectory of data acquisition, to shape around the head. A miniaturized, low profile custom-designed, UWB patch antenna capable of operating in the proximity of the head tissues is used as a sensor. The antenna aperture impedance is matched to the head tissues to enable pumping radiating electromagnetic waves into the human head needles of any coupling liquid. The impact of operational bandwidth and radiated power on system key performance factors i.e., penetration depth, range-resolution, and safety aspects are fully investigated based on calculations as well as full-wave simulations. It is observed that by increasing the bandwidth of radiated pulse from 0.8∼2.8 GHz (which was used in most of the previous works) to 1.0∼5.2 GHz, the range resolution can be improved remarkably from 23.0 mm to 16.0 mm deep inside the head. However, further increase in operational bandwidth will not improve the resolution of the system and a significant fraction of the radiated power will be dissipated in the brain tissues. To demonstrate the system performance, a constellation of 12 antenna elements is placed in an elliptical shape directly attached to the 3D phantom to acquire the raw data using full-wave simulations. Finally, an image from a 3D MRI-derived phantom is reconstructed to validate the imaging capability of the designed system. The proposed system can be used as a tool for early diagnosis as well as for treatment monitoring while getting therapy like chemo or radiotherapies.

Fourier Ptychographic Microscopy Reconstruction Method Based on Residual Transfer Networks

Fourier ptychographic microscopy reconstruction mostly adopts the traditional alternating iterative phase recovery method and optimization method, which has high computational complexity, high redundancy of image acquisition data, low reconstruction quality and high time consumption. In this paper, the model of residual transfer networks based on Resnet152 is proposed for Fourier ptychographic microscopy reconstruction, the learning process of deep convolution neural network is introduced, and the image reconstruction method based on deep learning realizes the end-to-end reconstruction of low-resolution images to high-resolution images. Through comparative experiments and analysis, the residual network can overcome the gradient explosion, make the feature information more complete and efficient, and the incremental up-sampling reconstruction network has higher image quality, lower computational complexity and shorter running time.

Efficient Light Probes for Real-Time Global Illumination

Reproducing physically-based global illumination (GI) effects has been a long-standing demand for many real-time graphical applications. In pursuit of this goal, many recent engines resort to some form of light probes baked in a precomputation stage. Unfortunately, the GI effects stemming from the precomputed probes are rather limited due to the constraints in the probe storage, representation or query. In this paper, we propose a new method for probe-based GI rendering which can generate a wide range of GI effects, including glossy reflection with multiple bounces, in complex scenes. The key contributions behind our work include a gradient-based search algorithm and a neural image reconstruction method. The search algorithm is designed to reproject the probes' contents to any query viewpoint, without introducing parallax errors, and converges fast to the optimal solution. The neural image reconstruction method, based on a dedicated neural network and several G-buffers, tries to recover high-quality images from low-quality inputs due to limited resolution or (potential) low sampling rate of the probes. This neural method makes the generation of light probes efficient. Moreover, a temporal reprojection strategy and a temporal loss are employed to improve temporal stability for animation sequences. The whole pipeline runs in realtime (>30 frames per second) even for high-resolution (1920×1080) outputs, thanks to the fast convergence rate of the gradient-based search algorithm and a light-weight design of the neural network. Extensive experiments on multiple complex scenes have been conducted to show the superiority of our method over the state-of-the-arts.