When dealing with material classification in baggage at airports, Dual-Energy Computed Tomography (DECT) allows characterization of any given material with coefficients based on two attenuative effects: Compton scattering and photoelectric absorption. However, straightforward projection-domain decomposition methods for this characterization often yield poor reconstructions due to the high dynamic range of material properties encountered in an actual luggage scan. Hence, for better reconstruction quality under a timing constraint, we propose a splitting-based, GPU-accelerated, statistical DECT reconstruction algorithm. Compared to prior art, our main contribution lies in the significant acceleration made possible by separating reconstruction and decomposition within an Alternating Direction Method of Multipliers (ADMM) framework. Experimental results, on both synthetic and real-world baggage phantoms, demonstrate a significant reduction in time required for convergence.

Non-smooth regularization is widely used in image reconstruction to eliminate the noise while preserving subtle image structures. In this work, we investigate the use of proximal Newton (PN) method to solve an optimization problem with a smooth data-fidelity term and total variation (TV) regularization arising from image reconstruction applications. Specifically, we consider a nonlinear Poisson-modeled single-energy X-ray computed tomography reconstruction problem with the data-fidelity term given by the I-divergence. The PN algorithm is compared to state-of-the-art first-order proximal algorithms, such as the wellestablished fast iterative shrinkage and thresholding algorithm (FISTA), both in terms of number of iterations and time to solutions. We discuss the key factors that influence the performance of PN, including the strength of regularization, the stopping criterion for both sub-problem and main-problem, and the use of exact or approximated Hessian operators.

The three-dimensional (3D) shape reconstruction problem of an object is a task of high interest in autonomous vehicles, detection of moving objects, and precision agriculture. A common methodology to recover the 3D shape of an object is using its optical phase. However, this approach involves solving a non-convex computationally demanding inverse problem known as phase retrieval (PR) in a setup that records coded diffraction patterns (CDP). Usually, the acquisition of several snapshots from the scene is required to solve the PR problem. This work proposes a single-shot 3D shape estimation technique using the optical phase of the object from CDP. The presented approach consists on accurately estimating the optical phase of the object by low-passfiltering the leading eigenvector of a carefully constructed matrix. Then, the estimated phase is used to infer the 3D object shape. It is important to mention that the estimation procedure does not involve a full time demanding reconstruction of the objects. Numerical results on synthetic data demonstrate that the proposed methodology closely estimates the 3D surface of an object with a normalized Mean-Square-Error of up to 0.27, under both noiseless and noisy scenarios. Additionally, the proposed method requires up to 60% less measurements to accurately estimate the 3D surface compared to a state-of-the-art methodology.

X-ray phase contrast tomography (XPCT) is widely used for 3D imaging of objects with weak contrast in X-ray absorption index but strong contrast in refractive index decrement. To reconstruct an object imaged using XPCT, phase retrieval algorithms are first used to estimate the X-ray phase projections, which is the 2D projection of the refractive index decrement, at each view. Phase retrieval is followed by refractive index decrement reconstruction from the phase projections using an algorithm such as filtered back projection (FBP). In practice, phase retrieval is most commonly solved by approximating it as a linear inverse problem. However, this linear approximation often results in artifacts and blurring when the conditions for the approximation are violated. In this paper, we formulate phase retrieval as a non-linear inverse problem, where we solve for the transmission function, which is the negative exponential of the projections, from XPCT measurements. We use a constraint to enforce proportionality between phase and absorption projections. We do not use constraints such as large Fresnel number, slowly varying phase, or Born/Rytov approximations. Our approach also does not require any regularization parameter tuning since there is no explicit sparsity enforcing regularization function. We validate the performance of our non-linear phase retrieval (NLPR) method using both simulated and real synchrotron datasets. We compare NLPR with a popular linear phase retrieval (LPR) approach and show that NLPR achieves sharper reconstructions with higher quantitative accuracy.

Accurately and rapidly detecting the locations of the cores of large-scale dendrites from 2D sectioned microscopic images helps quantify the microstructure of material components. This provides a critical link between the processing and properties of the material. Such a tool could be a critical part of a quality control procedure for manufacturing these components. In this paper, we propose to use Faster R-CNN, a convolutional neural network (CNN) model that considers both the detection accuracy and computational efficiency, to detect the dendrite cores with complex shapes. However, training CNN models usually requires a large number of images annotated with ground-truth locations of dendrite cores, which are usually obtained by highly laborintensive manual annotations. In this paper, we leverage the crystallographic symmetry of dendrite cores for data augmentation – the cross sections of dendrite cores show, not perfect, but near four-fold rotation symmetry and we can rotate the image around the center of dendrite cores by specified angles to construct new training data without additional manual annotations. We conduct a series of experiments and the results show the effectiveness of the Faster R-CNN method with the proposed data augmentation strategy. Particularly, we find that we can reduce the number of the manually annotated training images by 75% while still maintaining the same detection accuracy of dendrite cores.

In this paper we propose a surrogate approach to extract fibers and voids from polymer matrix composites by combining results obtained from model-based methods to train convolutional neural networks. This approach focuses on microscopy images where labeled data is not readily available, but purely model based approaches can be too slow due to their computational complexity. In addition, we propose an encoder-decoder alternative to a fiber instance segmentation paradigm, showing a speedup in training and inference times without a significant decrease in accuracy with respect to alternative methods. The neural networks approach represent a significant speedup over model based approaches and can correctly capture most fibers and voids in large volumes for further statistical analysis of the data.

This work compares the material classification performance of Mueller matrix polarization imaging to RGB imaging. White painted wood and white fabric samples are selected to create a classification task that is challenging for RGB imaging. A Mueller Matrix Imaging Polarimeter with a 30° full field of view is used to capture the Mueller Matrix images at nominal red, green, and blue wavelengths across multiple specular scatter angles. A Bayesian ideal observer model is used to evaluate classification performance. Performance is quantified by the Area under (AUC) the Receiver Operating Characteristic (ROC) curve. An AUC = 1 is perfect detection and AUC = 0.5 is the performance of guessing. The ensemble average AUC does not exceed 0.70 for RGB irradiance data. The ensemble average AUC for all 16 individual Mueller elements is greater than 0.95. Various combinations of Mueller matrix elements are also tested. Elements related to diattenuation and polarizance are nearly perfect classifiers for large scatter angles but the AUC minimum is 0.60 at 20°. Depolarization index is the highest performing parameter out of all tested polarization parameters for scatter angles ≥70° where AUC ≥0.98.

Dual Energy Computed Tomography (DECT) is expected to become a significant tool for voxel-based detection of hazardous materials in airport baggage screening. The traditional approach to DECT imaging involves collecting the projection data using two different X-ray spectra and then decomposing the data thus collected into line integrals of two independent characterizations of the material properties. Typically, one of these characterizations involves the effective atomic number (Z_eff) of the materials. However, with the X-ray spectral energies typically used for DECT imaging, the current best-practice approaches for dualenergy decomposition yield Z_eff values whose accuracy range is limited to only a subset of the periodic-table elements, more specifically to (Z < 30). Although this estimation can be improved by using a system-independent ρ_e — Z_e (SIRZ) space, the SIRZ transformation does not efficiently model the polychromatic nature of the X-ray spectra typically used in physical CT scanners. In this paper, we present a new decomposition method, AdaSIRZ, that corrects this shortcoming by adapting the SIRZ decomposition to the entire spectrum of an X-ray source. The method reformulates the X-ray attenuation equations as direct functions of (ρ_e, Z_e) and solves for the coefficients using bounded nonlinear least-squares optimization. Performance comparison of AdaSIRZ with other Z_eff estimation methods on different sets of real DECT images shows that AdaSIRZ provides a higher output accuracy for Z_eff image reconstructions for a wider range of object materials.

Dual-energy imaging has emerged as a superior way to recognize materials in X-ray computed tomography. To estimate material properties such as effective atomic number and density, one often generates images in terms of basis functions. This requires decomposition of the dual-energy sinograms into basis sinograms, and subsequently reconstructing the basis images. However, the presence of metal can distort the reconstructed images. In this paper we investigate how photoelectric and Compton basis functions, and synthesized monochromatic basis (SMB) functions behave in the presence of metal and its effect on estimation of effective atomic number and density. Our results indicate that SMB functions, along with edge-preserving total variation regularization, show promise for improved material estimation in the presence of metal. The results are demonstrated using both simulated data as well as data collected from a dualenergy medical CT scanner.