Overcoming Data Sparsity to Enable Deep Learning for Radiographic Non-Destructive Testing
Explore the source record for details and available documents.
Engineering topics
Publications and source records attributed to Jiang, M.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Abstract not provided
Purpose: Dynamic tracking of moving organs, such as lung and liver tumors, under radiation therapy requires prediction of organ motions prior to delivery. The shift of moving organ may change a lot due to huge transform of respiration at different periods. This study aims to reduce the influence of that changes using adjustable training signals and multi-layer perceptron neural network (ASMLP). Methods: Respiratory signals obtained using a Real-time Position Management(RPM) device were used for this study. The ASMLP uses two multi-layer perceptron neural networks(MLPs) to infer respiration position alternately and the training sample will be updated with time. Firstly, a Savitzky-Golay finite impulse response smoothing filter was established to smooth the respiratory signal. Secondly, two same MLPs were developed to estimate respiratory position from its previous positions separately. Weights and thresholds were updated to minimize network errors according to Leverberg-Marquart optimization algorithm through backward propagation method. Finally, MLP 1 was used to predict 120∼150s respiration position using 0∼120s training signals. At the same time, MLP 2 was trained using 30∼150s training signals. Then MLP is used to predict 150∼180s training signals according to 30∼150s training signals. The respiration position is predicted as this way until it was finished. Results: In this experiment, the two methods were used to predict 2.5 minute respiratory signals. For predicting 1s ahead of response time, correlation coefficient was improved from 0.8250(MLP method) to 0.8856(ASMLP method). Besides, a 30% improvement of mean absolute error between MLP(0.1798 on average) and ASMLP(0.1267 on average) was achieved. For predicting 2s ahead of response time, correlation coefficient was improved from 0.61415 to 0.7098.Mean absolute error of MLP method(0.3111 on average) was reduced by 35% using ASMLP method(0.2020 on average). Conclusion: The preliminary results demonstrate that the ASMLP respiratory prediction method is more accurate than MLP method and can improve the respiration forecast accuracy.
Abstract not provided
Detecting and predicting maneuvering satellites is an important problem for Space Situational Awareness. The spatial envelope of all possible locations within reach of such a maneuvering satellite is known as the Reachable Volume (RV). As soon as custody of a satellite is lost, calculating the RV and its subsequent time evolution is a critical component in the rapid recovery of the satellite. In this paper, we present a Monte Carlo approach to computing the RV for a given object. Essentially, our approach samples all possible trajectories by randomizing thrust-vectors, thrust magnitudes and time of burn. At any given instance, the distribution of the 'point-cloud' of the virtual particles defines the RV. For short orbital time-scales, the temporal evolution of the point-cloud can result in complex, multi-reentrant manifolds. Visualization plays an important role in gaining insight and understanding into this complex and evolving manifold. In the second part of this paper, we focus on how to effectively visualize the large number of virtual trajectories and the computed RV. We present a real-time out-of-core rendering technique for visualizing the large number of virtual trajectories. We also examine different techniques for visualizing the computed volume of probability density distribution, including volume slicing, convex hull and isosurfacing. We compare and contrast these techniques in terms of computational cost and visualization effectiveness, and describe the main implementation issues encountered during our development process. Finally, we will present some of the results from our end-to-end system for computing and visualizing RVs using examples of maneuvering satellites.
We study the creation of electron-positron pairs from the vacuum induced by two spatially displaced static electric fields. The strength and spatial width of each localized field is less than required for pair creation. If, however, the separation between the fields is less than the quantum-mechanical tunneling length associated with the corresponding quantum scattering system, the system produces a steady flux of electron-positron pairs. We compute the time dependence of the pair-creation probability by solving the Dirac equation numerically for various external field sequences. For the special case of two very narrow fields we provide an analytical expression for the pair-creation rate in the long-time limit.
Abstract not provided
A new fast reciprocating probe system (FRPS) has been built and installed on the outer midplane of the EAST tokamak to investigate the profiles of the boundary plasma parameters such as electron density and temperature. The system consists of a two-stage motion drive mechanism: slow motion and fast motion. The fast motion is powered by a servo motor, which drives the probe horizontally up to 50 cm to scan the edge region of the EAST tokamak. The maximum velocity achieved is 2 m/s. High velocity and flexible control of the fast motion are the remarkable features of this FRPS. A specially designed connector installed at the front end of the probe shaft makes it easy to install or replace the probe head on FRPS. During the latest experimental campaign in the spring of 2010, a probe head with seven tips, including two tips for a Mach probe, has been used. An example is given for simultaneous profile measurements of the plasma temperature, plasma density, and the plasma flow velocity.
This paper describes a systematic study on the epitaxial overgrowth of Pt on well-defined Pdnanocrystals with different shapes (and exposed facets), including regular octahedrons, truncated octahedrons, and cubes. Two different reducing agents, i.e., citric acid and L-ascorbic acid, were evaluated and compared for the reduction of K{sub 2}PtCl{sub 4} in an aqueous solution in the presence of Pdnanocrystal seeds. When citric acid was used as a reducing agent, conformal overgrowth of octahedral Pt shells on regular and truncated octahedrons of Pd led to the formation of Pd-Pt core-shell octahedrons, while non-conformal overgrowth of Pt on cubic Pd seeds resulted in the formation of an incomplete octahedral Pt shell. On the contrary, localized overgrowth of Pt branches was observed when L-ascorbic acid was used as a reducing agent regardless of the facets expressed on the surface of Pdnanocrystal seeds. This work shows that both the binding affinity of a reducing agent to the Pt surface and the reduction kinetics for a Pt precursor play important roles in determining the mode of Pt overgrowth on Pdnanocrystal surface.
Abstract not provided
Abstract not provided
Verification and validation (V&V) of geospatial image analysis algorithms is a difficult task and is becoming increasingly important. While there are many types of image analysis algorithms, we focus on developing V&V methodologies for algorithms designed to provide textual descriptions of geospatial imagery. In this paper, we present a novel methodological basis for V&V that employs a domain-specific ontology, which provides a naming convention for a domain-bounded set of objects and a set of named relationship between these objects. We describe a validation process that proceeds through objectively comparing benchmark imagery, produced using the ontology, with algorithm results. As an example, we describe how the proposed V&V methodology would be applied to algorithms designed to provide textual descriptions of facilities
This paper describes the application of a new, integrated modeling and simulation framework, encompassing the space situational awareness (SSA) enterprise, to the recent Cosmos-Iridium collision. This framework is based on a flexible, scalable architecture to enable efficient simulation of the current SSA enterprise, and to accommodate future advancements in SSA systems. In particular, the code is designed to take advantage of massively parallel, high-performance computer systems available, for example, at Lawrence Livermore National Laboratory. We will describe the application of this framework to the recent collision of the Cosmos and Iridium satellites, including (1) detailed hydrodynamic modeling of the satellite collision and resulting debris generation, (2) orbital propagation of the simulated debris and analysis of the increased risk to other satellites (3) calculation of the radar and optical signatures of the simulated debris and modeling of debris detection with space surveillance radar and optical systems (4) determination of simulated debris orbits from modeled space surveillance observations and analysis of the resulting orbital accuracy, (5) comparison of these modeling and simulation results with Space Surveillance Network observations. We will also discuss the use of this integrated modeling and simulation framework to analyze the risks and consequences of future satellite collisions and to assess strategies for mitigating or avoiding future incidents, including the addition of new sensor systems, used in conjunction with the Space Surveillance Network, for improving space situational awareness.
Controlling the morphology of Pt nanostructures can provide a great opportunity to improve their catalytic properties and increase their activity on a mass basis. We synthesized Pd-Pt bimetallic nanodendrites consisting of a dense array of Pt branches on a Pd core by reducing K{sub 2}PtCl{sub 4} with L-ascorbic acid in the presence of uniform Pd nanocrystal seeds in an aqueous solution. The Pt branches supported on faceted Pd nanocrystals exhibited relatively large surface areas and particularly active facets toward the oxygen reduction reaction (ORR), the rate-determining step in a proton-exchange membrane fuel cell. The Pd-Pt nanodendrites were two and a half times more active on the basis of equivalent Pt mass for the ORR than the state-of-the-art Pt/C catalyst and five times more active than the first-generation supportless Pt-black catalyst.