Gaussian-weighted Jensen–Shannon divergence as a robust fitness function for multi-model fitting

Model fitting is a fundamental component in computer vision for salient data selection, feature extraction and data parameterization. Conventional approaches such as the RANSAC family show limitations when dealing with data containing multiple models, high percentage of outliers or sample selection bias, commonly encountered in computer vision applications. In this paper, we present a novel model evaluation function based on Gaussian-weighted Jensen–Shannon divergence, and integrate into a particle swarm optimization (PSO) framework using ring topology. We avoid two problems from which most regression algorithms suffer, namely the requirements to specify inlier noise scale and the number of models. The novel evaluation method is generic and does not require any estimation of inlier noise. The continuous and meta-heuristic exploration facilitates estimation of each individual model while delivering the number of models automatically. Tests on datasets comprised of inlier noise and a large percentage of outliers (more than 90 % of the data) demonstrate that the proposed framework can efficiently estimate multiple models without prior information. Superior performance in terms of processing time and robustness to inlier noise is also demonstrated with respect to state of the art methods.     

Classification of small lesions in dynamic breast MRI: eliminating the need for precise lesion segmentation through spatio-temporal analysis of contrast enhancement

Characterizing the dignity of breast lesions as benign or malignant is specifically difficult for small lesions; they do not exhibit typical characteristics of malignancy and are harder to segment since margins are harder to visualize. Previous attempts at using dynamic or morphologic criteria to classify small lesions (mean lesion diameter of about 1 cm) have not yielded satisfactory results. The goal of this work was to improve the classification performance in such small diagnostically challenging lesions while concurrently eliminating the need for precise lesion segmentation. To this end, we introduce a method for topological characterization of lesion enhancement patterns over time. Three Minkowski Functionals were extracted from all five post-contrast images of 60 annotated lesions on dynamic breast MRI exams. For each Minkowski Functional, topological features extracted from each post-contrast image of the lesions were combined into a high-dimensional texture feature vector. These feature vectors were classified in a machine learning task with support vector regression. For comparison, conventional Haralick texture features derived from gray-level co-occurrence matrices (GLCM) were used. A new method for extracting thresholded GLCM features was also introduced and investigated here. The best classification performance was observed with Minkowski Functionals area and perimeter, thresholded GLCM features f8 and f9, and conventional GLCM features f4 and f6. However, both Minkowski Functionals and thresholded GLCM achieved such results without lesion segmentation while the performance of GLCM features significantly deteriorated when lesions were not segmented ( $p<0.05$ ). This suggests that such advanced spatio-temporal characterization can improve the classification performance achieved in such small lesions, while simultaneously eliminating the need for precise segmentation.     

mbeddr: instantiating a language workbench in the embedded software domain

Tools can boost software developer productivity, but building custom tools is prohibitively expensive, especially for small organizations. For example, embedded programmers often have to use low-level C with limited IDE support, and integrated into an off-the-shelf tool chain in an ad-hoc way. To address these challenges, we have built mbeddr, an extensible language and IDE for embedded software development based on C. mbeddr is a large-scale instantiation of the Jetbrains MPS language workbench. Exploiting its capabilities for language modularization and composition, projectional editing and multi-stage transformation, mbeddr is an open and modular framework that lets third parties add extensions to C with minimal effort and without invasive changes. End users can combine extensions in programs as needed. To illustrate the approach, in this paper we discuss mbeddr’s support for state machines, components, decision tables, requirements tracing, product line variability and program verification and outline their implementation. We also present our experience with building mbeddr, which shows that relying on language workbenches dramatically reduces the effort of building customized, modular and extensible languages and IDEs to the point where this is affordable by small organizations. Finally, we report on the experience of using mbeddr in a commercial project, which illustrates the benefits to end users.     

Multiparametric high-resolution imaging of barley embryos by multiphoton microscopy and magnetic resonance micro-imaging

Nonlinear optical microscopy and magnetic resonance imaging (MRI) address different properties of the sample and operate on different geometrical scales. MRI maps density and mobility of molecules tracking specific molecular signatures. Multiphoton imaging profits from the nonlinear absorption of light in the focus of a femtosecond laser source stimulating the autofluorescence of biomolecules. As this effect relies on a high light intensity, the accessible field of view is limited, but the resolution is very high (a few hundred nanometers). Here, we aim to link the different accessible scales and properties addressed in the different techniques to obtain a synoptic view. As model specimen we studied embryos of barley. Multiphoton stimulated autofluorescence images and images of second harmonic generation are achieved even down to low magnification (10x), low numerical aperture (N.A. 0.25) conditions. The overview images allowed morphological assignments and fluorescence lifetime imaging provides further information to identify accumulation of endogenous fluorophores. The second, complementary contribution from high-resolution MR images provides a 3D model and shows the embedding of the embryo in the grain. Images of the proton density were acquired using a standard 3D spin-echo imaging pulse sequence. Details directly comparable to the low magnification optical data are visible. Eventually, passing from the MR images of the whole grain via low magnification to high resolution autofluorescence data bridges the scale barrier, and might provide the possibility to trace transport and accumulation of, e.g., nutrients from large structure of the plant to the (sub-) cellular level.     

Note: Production of sharp gold tips with high surface quality

We present a simple method to produce sharp gold tips with excellent surface quality based on electrochemical etching with potassium chloride. Radii of curvature lie in the range of 20-40 nm and the surface roughness is measured to less than 0.8?nm. The tips are well suited for field emission, field ion microscopy, and likely for tip-enhanced Raman scattering as well as tip-enhanced near-field imaging.     

Combinatorial preparation and characterization of thin-film multilayer electro-optical devices

In this article we present a setup for the combinatorial vapor deposition of thin-film multilayer devices as well as methods for the fast and efficient analytic screening of the libraries obtained. The preparation setup is based on a commercially available evaporation chamber equipped with various evaporation sources for both organic and metallic materials. The combinatorial approach is realized by the combination of a rotation stage for the substrate, a five-mask sampler, and an additional mask whose position can be deliberately varied along one axis during the evaporation process. The latter is used to evaporate linear as well as step gradients by continuous or stepwise movement of a shutter mask. The mask sampler allows to define the sectors of the library and to evaporate more complex structures, e.g., an electrode layout. Finally, the simultaneous evaporation of two or more materials enables us to produce layers of varying composition ratio in general and doped materials, in particular. For the control of the evaporation process we have developed an automation software, which is particularly helpful for complex library designs and which grants excellent repeatability of experiments. Efficient and fast characterization of the obtained libraries is realized by (i) a purely optical setup and (ii) an electro-optical setup. (i) The UV/vis reader FLASHScan 530 permits to map out the UV/vis absorbance or fluorescence of the whole library. The UV/vis absorbance is primarily used to determine layer thicknesses and to confirm thickness uniformity across larger regions. The fluorescence measurements are used to determine the composition of layers containing fluorescent dyes. (ii) For a detailed short- and long-term electro-optical analysis we have developed an automated measurement system, which allows the characterization of 8x8 optoelectronic devices and to study their degradation behavior. Both solar cells and organic light-emitting diodes can be tested. Finally, we have developed a data analysis software to extract characteristic values from the huge amount of data and with this facilitate the finding of systematic dependencies.     

Planning for Mixed Use: Affordable for Whom?

Problem, research strategy, and findings: Mixed-use zoning is widely advocated to increase density; promote active transportation; encourage economic development; and create lively, diverse neighborhoods. We know little, however, about whether mixed-use developments affect housing affordability. We question the impact of mixed-use zoning on housing affordability in Toronto (Canada) between 1991 and 2006 in the face of waning government support for affordable housing and increasing income inequality due to the occupational restructuring accompanying a shift to a knowledge-based economy. We fi nd that housing in mixed-use zones remained less affordable than housing in the rest of the city and in the metropolitan region. High-income service occupations experienced improved affordability while lower wage service, trade, and manufacturing occupations experienced stagnant or worsening affordability. Housing in mixed-use zones is increasingly affordable only to workers already able to pay higher housing costs. Our findings are limited to Canada's largest city but have lessons for large North American cities with similar urban economies and housing markets.

Takeaway for practice: Mixed-use developments may reduce housing affordability in core areas and inadvertently reinforce the sociospatial inequality resulting from occupational polarization unless supported by appropriate affordable housing policies. Planners should consider a range of policy measures to offset the unintentional outcomes of mixed-use developments and ensure affordability within mixed-use zones: inclusionary zoning, density bonuses linked to affordable housing, affordable housing trusts, and other relevant methods.     

Micromechanics of ITZ‐Aggregate Interaction in Concrete Part II: Strength Upscaling

Cracking in concrete typically starts in the immediate vicinity around the aggregates, i.e., in the region of the so‐called interfacial transition zones (ITZs), but the process is still not fully understood. Notably, crushing of concrete in compression results in fragments with interesting aggregate surface textures. Part of the aggregate surfaces is cleanly separated from the ITZ, while another part of the aggregate surfaces remains covered with a thin layer of cement paste. This suggests two different types of failure: ITZ‐aggregate separation and ITZ failure; which we here study based on the continuum micromechanics approach of the companion paper (part I). It provides access to both traction vectors acting on aggregate surfaces and three‐dimensional stress states within representative ITZ volumes for loading states below the elastic limit of concrete. When inserting these microtractions and microstresses into Rankine‐type strength criteria for the aggregate‐ITZ interface and for the ITZ, respectively, the micromechanics model allows for upscaling this microscopic failure behavior to concrete‐level criteria for crack onset. Comparing the latter to corresponding experimental results, reveals that under tension‐dominated loading both ITZ failure and ITZ‐aggregate separation appear to be realistic, while under compression‐dominated loading ITZ failure appears as the more likely mechanism. Also, comparing model and experiments shows that the ITZ‐aggregate separation strength amounts to at least half of the internal ITZ cohesion strength, but may be much larger than the latter.     

Assessing software product line potential: an exploratory industrial case study

Corporate organizations sometimes offer similar software products in certain domains due to former company mergers or due to the complexity of the organization. The functional overlap of such products is an opportunity for future systematic reuse to reduce software development and maintenance costs. Therefore, we have tailored existing domain analysis methods to our organization to identify commonalities and variabilities among such products and to assess the potential for software product line (SPL) approaches. As an exploratory case study, we report on our experiences and lessons learned from conducting the domain analysis in four application cases with large-scale software products. We learned that the outcome of a domain analysis was often a smaller integration scenario instead of an SPL and that business case calculations were less relevant for the stakeholders and managers from the business units during this phase. We also learned that architecture reconstruction using a simple block diagram notation aids domain analysis and that large parts of our approach were reusable across application cases.