Material Identification Algorithm Research Articles

An adaptive decomposition algorithm for quantitative spectral CT imaging

AbstractWith the development of photon counting detectors (PCDs), spectral CT achieves higher performance than in the past, thus has better application prospects. In this paper, an adaptive dual effect decomposition (ADED) algorithm is proposed to perform the quantitative estimation of effective atomic number and electron density. The algorithm is an empirical material identification algorithm that achieves calibration by combining polynomials on the projection data and constraining in the image domain. In addition, we also propose an innovative effective atomic number estimation model with an accurate physical model that adaptively adjusts the parameters according to the x‐ray energy and material type. The performance of the algorithm was verified by experiment and simulation in this paper. Compared to the conventional basis material decomposition algorithm, the proposed ADED algorithm reduces the relative error in the quantification of the effective atomic number from 4.5%–12.0% to 0.3%–4.5%. Overall, the results demonstrate that the proposed algorithm significantly improves the accuracy of quantifying the effective atomic number.

DIMAR: A Contactless Material Identification Algorithm for Complex Permittivity of Dielectrics via Moving RFID System

DIMAR: A Contactless Material Identification Algorithm for Complex Permittivity of Dielectrics via Moving RFID System

A 3D fluid-solid interaction model of the air puff test in the human cornea

We present a numerical model of a contactless test commonly used to assess the biomechanics of the human cornea. The test, consisting in a rapid air jet applied to the anterior surface of the cornea, is controversial. Although the numerous studies documented in the literature have not been able yet to clarify its relevance as a diagnostic tool, the test has the potential to be combined with inverse analysis procedures to characterize the parameters of numerical models of the cornea. With the final goal of employing the air puff test in advanced material identification algorithms, here we propose to model the cornea with standard finite elements and the fluids filling the anterior chamber of the eye with a meshfree discretization. The interaction between moving fluids and deforming cornea is accounted for by modifying the interface boundary conditions of both fluid and solid. The proposed model represents the first fully 3D example of an aqueous-cornea fluid-solid interaction analysis which uses a robust meshfree approach for the fluid. Although we restrict our scope to isotropic nonlinear materials, numerical results confirm the undeniable importance of including internal fluids in the simulation of the air puff test. Thus the proposed approach stands as a procedural paradigm for the identification of the mechanical parameters of the human cornea.

Spectral Radiance Modeling and Bayesian Model Averaging for Longwave Infrared Hyperspectral Imagery and Subpixel Target Identification

Hyperspectral imagery (HSI) exploitation typically requires spectral signatures for target detection and identification algorithms. As the longwave infrared (LWIR) region of the electromagnetic spectrum is dominated by thermal emission, spectral radiance measurements are influenced by object temperature, and thus, estimates of target temperature may be necessary for emissivity retrieval to support these algorithms. Therefore, lack of accurate temperature information poses a significant challenge for HSI target detection and identification. Previous studies have demonstrated LWIR hyperspectral unmixing in both radiance and emissivity domains using in-scene target signatures. Here, a radiance-domain LWIR material identification algorithm for subpixel target identification of solid materials is developed by combining spectral radiance and linear mixing models with Bayesian model averaging. Application to experimental LWIR HSI illustrates that the algorithm effectively distinguishes between solid materials with a high degree of spectral similarity and reduces the probability of false alarms by at least one order of magnitude over a standard adaptive coherence estimator detector. Limits of identification are inferred from the imagery and found to depend on material type, target size, and target geometry. For the sensor and materials in this paper, the results imply that targets of nominally 5 m2 in size with strong spectral features can be identified for ground sampling distances (GSDs) on the order of 5–10 m (with abundances as low as ~10%) whereas blackbody-like materials are difficult to distinguish for GSDs larger than approximately 3 m.

Automatic process parameter adaption for a hybrid workpiece during cylindrical operations

The combination of different materials in one workpiece in order to optimize the workpiece characteristics is a state of the art design method for high-performance components. These workpieces can be made of the most qualified materials according to local requirements. However, during machining of material compounds, the different materials have to be considered. This is due to the specific material properties, cutting characteristics, and chip formation mechanisms. Cutting parameters have to be adapted for each material in order to achieve the demanded workpiece quality and optimal processes in respect of tool life and material removal rate. The focus of this research is the development of an in-process material identification algorithm. Thus, a cylindrical turning process is investigated for friction welded aluminum/steel shafts (EN-AW6082/20MnCr5). A universal monitoring approach is presented which detects the different materials process-parallel. For this purpose, cutting forces and spindle torque are linked with a dexel-based material removal model to determine monitoring parameters. The design of experiment method is used to validate the approach for various process parameters. Cutting speed and feed velocity are adapted for cylindrical turning operations based on the monitoring algorithm. As a result, material-specific cutting parameters are adjusted during the machining in order to optimize the material removal rate. Based on this approach, further process optimization can be implemented, like the improvement of chip formation, while machining hybrid workpieces.

Can active proton interrogation find shielded nuclear threats at human-safe radiation levels?

A new method of low-dose proton radiography is presented. The system is composed of an 800MeV proton source, bending magnets, and compact detectors, and is designed for drive-through cargo scanning. The system has been simulated using GEANT4. Material identification algorithms and pixel sorting methods are presented that allow the system to perform imaging at doses low enough to scan passenger vehicles and people. Results are presented on imaging efficacy of various materials and cluttered cargoes. The identification of shielded nuclear materials at human-safe doses has been demonstrated.

Application of Muon Tomography to Detect Radioactive Sources Hidden in Scrap Metal Containers

The accidental melting of radioactive sources hidden inside metal scrap containers can produce severe environmental harm. Modern melting facilities are equipped with portals measuring radiation levels. Nonetheless, sources can pass undiscovered when shielded inside shells of high density material, such as lead. From time to time indeed some radioactive sources pass undetected through the controls at foundries entrance. Once molten they caused enormous damages to the steel mills, contaminating all the production line. The muon tomography technique allows to discriminate high-Z materials measuring multiple scattering of cosmic ray muons inside matter. Therefore this technique can be used to analyse a truck container searching for high-density source shields. We report here the results about simulation studies of a muon tomography portal. Within the Mu-Steel European project we developed the prototype design, the three-dimensional images reconstruction software and the high density material identification algorithm. MonteCarlo simulation was validated with data from a large volume demonstrator (~11 m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> ) built using spare muon drift-time chambers of the CMS high energy physics experiment operating at the Large Hadron Collider at CERN.

Material identification of loose particles in sealed electronic devices using PCA and SVM

Abstract The existence of loose particles left inside the sealed electronic devices is one of the main factors affecting the reliability of the whole system. It is important to identify the particle material for analyzing their source. The conventional material identification algorithms mainly rely on time, frequency and wavelet domain features. However, these features are usually overlapped and redundant, resulting in unsatisfactory material identification accuracy. The main objective of this paper is to improve the accuracy of material identification. First, the principal component analysis (PCA) is employed to reselect the nine features extracted from time and frequency domains, leading to six less correlated principal components. And then the reselected principal components are used for material identification using a support vector machine (SVM). Finally, the experimental results show that this new method can effectively distinguish the type of materials including wire, aluminum and tin particles.

3D dose distribution calculation in a voxelized human phantom by means of Monte Carlo method

The aim of this work is to provide the reconstruction of a real human voxelized phantom by means of a MatLab program and the simulation of the irradiation of such phantom with the photon beam generated in a Theratron 780 (MDS Nordion) (60)Co radiotherapy unit, by using the Monte Carlo transport code MCNP (Monte Carlo N-Particle), version 5. The project results in 3D dose mapping calculations inside the voxelized antropomorphic head phantom. The program provides the voxelization by first processing the CT slices; the process follows a two-dimensional pixel and material identification algorithm on each slice and three-dimensional interpolation in order to describe the phantom geometry via small cubic cells, resulting in an MCNP input deck format output. Dose rates are calculated by using the MCNP5 tool FMESH, superimposed mesh tally, which gives the track length estimation of the particle flux in units of particles/cm(2). Furthermore, the particle flux is converted into dose by using the conversion coefficients extracted from the NIST Physical Reference Data. The voxelization using a three-dimensional interpolation technique in combination with the use of the FMESH tool of the MCNP Monte Carlo code offers an optimal simulation which results in 3D dose mapping calculations inside anthropomorphic phantoms. This tool is very useful in radiation treatment assessments, in which voxelized phantoms are widely utilized.

Material identification algorithms for parallel systems

The direct binary hypercube interconnection network has been widely used in the design of parallel computer systems, because it has low diameter, and can effectively emulate commonly used structures such as binary trees, rings, and meshes. However, the hypercube has the disadvantage of high VLSI complexity due to the increase in the number of communication ports and channels per PE (processing element) with an increase in the dimension of the hypercube. This complexity is a major drawback of the hypercube because it limits its potential for scalability. Ziavras has introduced the reduced hypercube (RH) interconnection network. The RH reduces VLSI complexity without compromising performance. The RH is obtained from a regular hypercube by a uniform reduction in the number of edges for each hypercube node. The main objective of this paper is to demonstrate the feasibility of the RH in implementing material identification algorithms using X-ray fluorescence. Relevant algorithms are developed for the PRAM and the hypercube also, for comparative analysis. The algorithms rely heavily on the binary tree emulation capabilities of these systems. The results indicate that the RH is capable of providing hypercube-like performance at a significantly reduced complexity/cost.