Processing and interactive editing of huge point clouds from 3D scanners

This paper describes a new out-of-core multi-resolution data structure for real-time visualization, interactive editing and externally efficient processing of large point clouds. We describe an editing system that makes use of the novel data structure to provide interactive editing and preprocessing tools for large scanner data sets. Using the new data structure, we provide a complete tool chain for 3D scanner data processing, from data preprocessing and filtering to manual touch-up and real-time visualization. In particular, we describe an out-of-core outlier removal and bilateral geometry filtering algorithm, a toolset for interactive selection, painting, transformation, and filtering of huge out-of-core point-cloud data sets and a real-time rendering algorithm, which all use the same data structure as storage backend. The interactive tools work in real-time for small model modifications. For large scale editing operations, we employ a two-resolution approach where editing is planned in real-time and executed in an externally efficient offline computation afterwards. We evaluate our implementation on example data sets of sizes up to 63 GB, demonstrating that the proposed technique can be used effectively in real-world applications.     

Improving the structure MCMC sampler for Bayesian networks by introducing a new edge reversal move

Applications of Bayesian networks in systems biology are computationally demanding due to the large number of model parameters. Conventional MCMC schemes based on proposal moves in structure space tend to be too slow in mixing and convergence, and have recently been superseded by proposal moves in the space of node orders. A disadvantage of the latter approach is the intrinsic inability to specify the prior probability on network structures explicitly. The relative paucity of different experimental conditions in contemporary systems biology implies a strong influence of the prior probability on the posterior probability and, hence, the outcome of inference. Consequently, the paradigm of performing MCMC proposal moves in order rather than structure space is not entirely satisfactory. In the present article, we propose a new and more extensive edge reversal move in the original structure space, and we show that this significantly improves the convergence of the classical structure MCMC scheme.     

Result Merging Technique for Answering XPath Query over XSLT Transformed Data

Caching stores the results of previously answered queries in order to answer succeeding queries faster by reusing these results. We propose two different approaches for using caches of XSLT transformed XML data in order to answer queries. The first approach checks whether or not a current query Q can be directly answered from the result of a previously answered query Qi stored in the cache. The new query is otherwise submitted to the source over the network, the answer of the query is determined, transmitted back to the client, and stored in the cache. The second approach determines only the intersection Q−Qi and integrates the result of Q−-Qi into the previous results in the cache, which requires applying a numbering scheme for the output of the XSLT stylesheet. We show by experimental results that the second approach can significantly speed up the answering time in comparison to the first approach, but is not significantly slower in few worst cases than the second approach.     

Automated processing for map generalization using web services

In map generalization various operators are applied to the features of a map in order to maintain and improve the legibility of the map after the scale has been changed. These operators must be applied in the proper sequence and the quality of the results must be continuously evaluated. Cartographic constraints can be used to define the conditions that have to be met in order to make a map legible and compliant to the user needs. The combinatorial optimization approaches shown in this paper use cartographic constraints to control and restrict the selection and application of a variety of different independent generalization operators into an optimal sequence. Different optimization techniques including hill climbing, simulated annealing and genetic deep search are presented and evaluated experimentally by the example of the generalization of buildings in blocks. All algorithms used in this paper have been implemented in a web services framework. This allows the use of distributed and parallel processing in order to speed up the search for optimized generalization operator sequences.

Model‐based reconstruction for T1 mapping using single‐shot inversion‐recovery radial FLASH

Quantitative parameter mapping in MRI is typically performed as a two‐step procedure where serial imaging is followed by pixelwise model fitting. In contrast, model‐based reconstructions directly reconstruct parameter maps from raw data without explicit image reconstruction. Here, we propose a method that determines T1 maps directly from multi‐channel raw data as obtained by a single‐shot inversion‐recovery radial FLASH acquisition with a Golden Angle view order. Joint reconstruction of a T1, spin‐density and flip‐angle map is formulated as a nonlinear inverse problem and solved by the iteratively regularized Gauss‐Newton method. Coil sensitivity profiles are determined from the same data in a preparatory step of the reconstruction. Validations included numerical simulations, in vitro MRI studies of an experimental T1 phantom, and in vivo studies of brain and abdomen of healthy subjects at a field strength of 3 T. The results obtained for a numerical and experimental phantom demonstrated excellent accuracy and precision of model‐based T1 mapping. In vivo studies allowed for high‐resolution T1 mapping of human brain (0.5–0.75 mm in‐plane, 4 mm section thickness) and liver (1.0 mm, 5 mm section) within 3.6–5 s. In conclusion, the proposed method for model‐based T1 mapping may become an alternative to two‐step techniques, which rely on model fitting after serial image reconstruction. More extensive clinical trials now require accelerated computation and online implementation of the algorithm. © 2016 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 26, 254–263, 2016     

Investigation of Ablation Efficiency of Stainless Steel Using Pulsed Lasers in Burst Mode

Due to its high corrosion resistance and mechanical properties, stainless steel is commonly used in various industrial applications. Although different types of stainless steel are similar in their chemical composition, they can differ significantly in their thermal diffusivity. This property is relevant in the ability of a material to conduct heat and thus, in laser processing. In this frame, this study compares the ablation efficiency and characteristics of polished stainless steel samples of the alloys AISI 304, AISI 420, and AISI 316Ti. They are irradiated with single ultrashort pulses having pulse durations between 250 fs and 10 ps as well as using GHz burst modii. The goal is to investigate the differences in both the ablation threshold and the ablation rate to improve the ablation efficiency. The results show that shorter pulse durations lead to a more efficient ablation process. On the other hand, GHz bursts are found to be, in general, less efficient. In addition, there is a significant difference in the surface morphology depending on the process parameters. The differences in the thermal diffusivity do not significantly influence the ablation threshold fluence but surface morphology and the ablation rate.     

A Shock Tube Study of the Reactions of H Atoms with COS,CS2, and H2S

Reactions of H atoms with COS, CS₂, and H₂S were studied behind reflected shock waves at temperatures between 1170 K and 1830 K and pressures around 1.0 bar by applying atomic resonance absorption spectroscopy (ARAS) for time-resolved measurements of H atoms at L_α. The thermal decomposition of a few ppm ethyl iodide (C₂H₅I) was used as a H-atom source. In the presence of a large excess of the molecular reactant COS, CS₂, or H₂S, a consumption of H was observed which follows a pseudo first-order rate law. Rate coefficients for the reactions: were determined to be: k₁ = 2.4 × 10¹⁴exp(–3415 K/T) cm³mol⁻¹s⁻¹ k₂ = 1.4 × 10¹⁵exp(–9250 K/T) cm³mol⁻¹s⁻¹ k₃ = 2.5 × 10¹⁴exp(–2890 K/T) cm³mol⁻¹s⁻¹     

Nested multigrid methods for time-periodic,parabolic optimal control problems

We present a nested multigrid method to optimize time-periodic, parabolic, partial differential equations (PDE). We consider a quadratic tracking objective with a linear parabolic PDE constraint. The first order optimality conditions, given by a coupled system of boundary value problems can be rewritten as an Fredholm integral equation of the second kind, which is solved by a multigrid of the second kind. The evaluation of the integral operator consists of solving sequentially a boundary value problem for respectively the state and the adjoints. Both problems are solved efficiently by a time-periodic space-time multigrid method.     

Ensemble classification based on generalized additive models

Generalized additive models (GAMs) are a generalization of generalized linear models (GLMs) and constitute a powerful technique which has successfully proven its ability to capture nonlinear relationships between explanatory variables and a response variable in many domains. In this paper, GAMs are proposed as base classifiers for ensemble learning. Three alternative ensemble strategies for binary classification using GAMs as base classifiers are proposed: (i) GAMbag based on Bagging, (ii) GAMrsm based on the Random Subspace Method (RSM), and (iii) GAMens as a combination of both. In an experimental validation performed on 12 data sets from the UCI repository, the proposed algorithms are benchmarked to a single GAM and to decision tree based ensemble classifiers (i.e. RSM, Bagging, Random Forest, and the recently proposed Rotation Forest). From the results a number of conclusions can be drawn. Firstly, the use of an ensemble of GAMs instead of a single GAM always leads to improved prediction performance. Secondly, GAMrsm and GAMens perform comparably, while both versions outperform GAMbag. Finally, the value of using GAMs as base classifiers in an ensemble instead of standard decision trees is demonstrated. GAMbag demonstrates performance comparable to ordinary Bagging. Moreover, GAMrsm and GAMens outperform RSM and Bagging, while these two GAM ensemble variations perform comparably to Random Forest and Rotation Forest. Sensitivity analyses are included for the number of member classifiers in the ensemble, the number of variables included in a random feature subspace and the number of degrees of freedom for GAM spline estimation.