Advanced optical methods for safe image reproduction on automotive displays

Modern cars are equipped with camera monitor systems (CMSs), such as a backup camera or side-mirror replacement. These systems are expected to perform optimally and achieve high safety levels (ASIL). Currently, only digital data are supervised in CMSs and safety mechanisms for such systems are individually derived on a case-by-case basis which is not effective. This study proposes generic optical supervision for displays of automotive CMS. This paper introduces “light-to-light” (camera to display output) protection for both in-car CMS and remote operator monitors used in autonomous car fleet operation centers. The first method is based on photodiodes attached to the display to optically supervise, for instance, the speedometer of vehicles. By combining intensities of photodiodes with calibration data, we can compare the measured speed with the value from CAN (Controller Area Network) data. The second method that entails capturing the display content using a camera enables top safety levels for both in-car displays and remote operator monitors. This safeguarding was successfully verified by conventional image processing and artificial intelligence (AI)-based analysis methods. Our results demonstrate that AI methods allow a substantial reduction in the wireless transmission bandwidth from a car to a remote operator compared with conventional image processing.     

Voronoi-based extraction and visualization of molecular paths

Visual analysis is widely used to study the behavior of molecules. Of particular interest are the analysis of molecular interactions and the investigation of binding sites. For large molecules, however, it is difficult to detect possible binding sites and paths leading to these sites by pure visual inspection. In this paper, we present new methods for the computation and visualization of potential molecular paths. Using a novel filtering method, we extract the significant paths from the Voronoi diagram of spheres. For the interactive visualization of molecules and their paths, we present several methods using deferred shading and other state-of-the-art techniques. To allow for a fast overview of reachable regions of the molecule, we illuminate the molecular surface using a large number of light sources placed on the extracted paths. We also provide a method to compute the extension surface of selected paths and visualize it using the skin surface. Furthermore, we use the extension surface to clip the molecule to allow easy visual tracking of even deeply buried paths. The methods are applied to several proteins to demonstrate their usefulness.     

Improved parallelism and scheduling in multi-core software routers

Recent technological advances in commodity server architectures, with multiple multi-core CPUs, integrated memory controllers, high-speed interconnects, and enhanced network interface cards, provide substantial computational capacity, and thus an attractive platform for packet forwarding. However, to exploit this available capacity, we need a suitable software platform that allows effective parallel packet processing and resource management. In this paper, we at first introduce an improved forwarding architecture for software routers that enhances parallelism by exploiting hardware classification and multi-queue support, already available in recent commodity network interface cards. After evaluating the original scheduling algorithm of the widely-used Click modular router, we propose solutions for extending this scheduler for improved fairness, throughput, and more precise resource management. To illustrate the potential benefits of our proposal, we implement and evaluate a few key elements of our overall design. Finally, we discuss how our improved forwarding architecture and resource management might be applied in virtualized software routers.     

Applying SVMs and weight-based factor analysis to unsupervised adaptation for speaker verification

This paper presents an extended study on the implementation of support vector machine (SVM) based speaker verification in systems that employ continuous progressive model adaptation using the weight-based factor analysis model. The weight-based factor analysis model compensates for session variations in unsupervised scenarios by incorporating trial confidence measures in the general statistics used in the inter-session variability modelling process. Employing weight-based factor analysis in Gaussian mixture models (GMMs) was recently found to provide significant performance gains to unsupervised classification. Further improvements in performance were found through the integration of SVM-based classification in the system by means of GMM supervectors.This study focuses particularly on the way in which a client is represented in the SVM kernel space using single and multiple target supervectors. Experimental results indicate that training client SVMs using a single target supervector maximises performance while exhibiting a certain robustness to the inclusion of impostor training data in the model. Furthermore, the inclusion of low-scoring target trials in the adaptation process is investigated where they were found to significantly aid performance.     

Quantifying structural attributes of system decompositions in 28 feature-oriented software product lines

A key idea of feature orientation is to decompose a software product line along the features it provides. Feature decomposition is orthogonal to object-oriented decomposition—it crosscuts the underlying package and class structure. It has been argued often that feature decomposition improves system structure by reducing coupling and by increasing cohesion. However, recent empirical findings suggest that this is not necessarily the case. In this exploratory, observational study, we investigate the decompositions of 28 feature-oriented software product lines into classes, features, and feature-specific class fragments. The product lines under investigation are implemented using the feature-oriented programming language Fuji. In particular, we quantify and compare the internal attributes import coupling and cohesion of the different product-line decompositions in a systematic, reproducible manner. For this purpose, we adopt three established software measures (e.g., coupling between units, CBU; internal-ratio unit dependency, IUD) as well as standard concentration statistics (e.g., Gini coefficient). In our study, we found that feature decomposition can be associated with higher levels of structural coupling in a product line than a decomposition into classes. Although coupling can be concentrated in very few features in most feature decompositions, there are not necessarily hot-spot features in all product lines. Interestingly, feature cohesion is not necessarily higher than class cohesion, whereas features are more equal in serving dependencies internally than classes of a product line. Our empirical study raises critical questions about alleged advantages of feature decomposition. At the same time, we demonstrate how our measurement approach of coupling and cohesion has potential to support static and dynamic analyses of software product lines (i.e., type checking and feature-interaction detection) by facilitating product sampling.     

<Emphasis Type="Italic">In situ</Emphasis> grazing-incidence small-angle X-ray scattering observation of block-copolymer templated formation of magnetic nanodot arrays and their magnetic properties

The fabrication of bit-patterned media (BPM) is crucial for new types of hard disk drives.The development of methods for the production of BPM is progressing rapidly.Conventional lithography reaches the limit regarding lateral resolution,and new routes are needed.In this study,we mainly focus on the dependence of the size and shape of magnetic nanodots on the Ar+-ion etching duration,using silica dots as masks.Two-dimensional (2D) arrays of magnetic nanostructures are created using silica-filled diblock-copolymer micelles as templates.After the self-assembly of the micelles into 2D hexagonal arrays,the polymer shell is removed,and the SiO2 cores are utilized to transform the morphology into a (Co/Pt)2-multilayer via ion etching under normal incidence.The number of preparation steps is kept as low as possible to simplify the formation of the nanostructure arrays.High-resolution in situ grazing-incidence small-angle X-ray scattering (GISAXS) investigations are performed during the Ar+-ion etching to monitor and control the fabrication process.The in situ investigation provides information on how the etching conditions can be improved for further ex situ experiments.The GISAXS patterns are compared with simulations.We observe that the dots change in shape from cylindrical to conical during the etching process.The magnetic behavior is studied by utilizing the magneto-optic Kerr effect.The Co/Pt dots exhibit different magnetic behaviors depending on their size,interparticle distance,and etching time.They show ferromagnetism with an easy axis of magnetization perpendicular to the film.A systematic dependence of the coercivity on the dot size is observed.     

新一代介质隔离式压力传感器

概述了三种不同介质隔离式压力传感器的温度性能比的测试结果。该测试不仅在指定的补偿温度范围内,而且还在拓宽的温度范围内进行。选用两种自产型号及一种其他品牌的传感器来完成此实验。实验结果表明,选用超稳定的压力芯片加以优秀封装设计的IC压力传感器在温度性能方面具有极大优越性。另外,对IC传感器中新型介质隔离式压力传感器进行了详尽的介绍。