Search

This paper presents a fully analytical physics-based model for the current-voltage (I-V) characteristics of resonant tunneling diodes. We consider the whole electrical potential distribution in the structure, including the space charge regions in the emitter and collector layers. Additionally, we account for scattering mechanisms experienced by carriers during tunneling through the double barrier region as a function of the applied bias voltage. The model is validated with experimental and numerical data, yielding excellent agreement

COVID-19 is an infectious disease caused by a type of coronavirus recently discovered, called SARS-CoV-2. It has infected more than 20 million people worldwide and it is responsible for more than 737,000 deaths. This work presents a study that explores linear regression mechanisms combined with a sliding and cumulative time window approach to provide inputs to assist in decision making for public policies, within the scope of the COVID-19 pandemic evolution, whether they are hardening or easing the isolation. Data from five states of Brazil were collected and applied a Ridge regression to predict the curve behavior of cases and deaths of COVID-19. As a result, an Explained Variance Status (EVS) up to 0.998 and 0.999 is presented, considering cases and deaths, respectively. It was concluded that sliding time window bring more information about the infection than cumulative, since public policy changes in a few time-lapse.

Surveillance of overhead power line installations can be conveniently addressed using unmanned aerial vehicles (UAV). UAV are robotic platforms able to perform sophisticated tasks such as autonomous flight based on visual information. In this paper, we propose a novel solution to the problem of following a power line autonomously based on monocular vision. The method uses Deep Neural Networks (DNN) and the Hough transform to successfully discern power line images from environmental information, which is an essential result to accomplish fully autonomous vision-based navigation. A simulated navigation test demonstrates the efficiency of the proposed method, in the special condition of following right-angled changes of direction, which is a known restriction in many navigation methods reported in literature. The design of the proposed method is modular and can be incorporated in navigation strategies for automatic surveillance applications.

Existent video description approaches advocated in the literature rely on capturing the semantic relationships among concepts and visual features from training data specific to various datasets. Naturally, their success at generalizing the video descriptions for the domain is closely dependent on the availability, representativeness, size and annotation quality of the training data. Common issues are overfitting, the amount of training data and computational time required for the model. To overcome these issues, we propose to alleviate the learning of semantic knowledge from domain-specific datasets by leveraging general human knowledge sources such as ConceptNet. We propose the use of ConceptNet as the source of knowledge for generating video descriptions using Grenander's pattern theory formalism. Instead of relying on training data to estimate semantic compatibility of two concepts, we use weights in the ConceptNet that determines the degree of validity of the assertion between two concepts based on the knowledge sources. We test and compare this idea on the task of generating semantically coherent descriptions for videos from the Breakfast Actions and Carnegie Mellon's Multimodal activities dataset. In comparison with other approaches, the proposed method achieves comparable accuracy against state-of-the-art methods based on HMMs and CFGs and generate semantically coherent descriptions even when presented with inconsistent action and object labels. We are also able to show that the proposed approach performs comparably with models trained on domain-specific data.

Amorphous metals have been shown to have magnetomechanical properties which are superior to those of any other magnetic materials.

This paper presents the control design for the wind turbine in a hybrid photovoltaic/wind system. A Permanent Magnet Synchronous Generator (PMSG) connected to the wind turbine converts wind power into electricity. The speed is estimated without sensors by a PLL (Phase locked loop) algorithm. The proposed Maximum Power Point Tracking (MPPT) gives the speed reference based on the turbine power versus speed curve. A boost converter circuit is used to control the generator speed by changing the DC input voltage. PLECS Simulation Software and experimental results verify the performance of the proposed method in simulation.

Reporting-Guidelines in Medicine play an important role in promoting the quality of reports in health-related research. For instance, a poorly reported research may induce misinterpretation and inappropriate clinical settings. Basically, three major issues regarding Reporting-Guidelines in Medicine has to be addressed: (1) the existence of lots of reporting guidelines covering a broad spectrum of medical research increases the difficulties when reporting a medical research because it may not be clear which guidelines are applicable, or if all related guidelines are considered; (2) most guideline developers have not described how their guidance was developed, consequently tracking guidelines and features for evaluation, or even for correcting, may be not possible; (3) few guideline developers have evaluated the effectiveness of their guidance, then the real value that the guideline brings to reporting the research is not assured. Hence, we propose a solution to these three issues based on the Model-View-Controller Evolutionary Acquisition Interdisciplinary Research Project Management (MVC EA-IRPM) that is simultaneously a strategy and a web system for the development and management of reporting guidelines.

Deep learning-based models have recently been widely successful at outperforming traditional approaches in several computer vision applications such as image classification, object recognition and action recognition. However, those models are not naturally designed to learn structural information that can be important to tasks such as human pose estimation and structured semantic interpretation of video events. In this paper, we demonstrate how to build structured semantic understanding of audio-video events by reasoning on multiple-label decisions of deep visual models and auditory models using Grenander's structures for imposing semantic consistency. The proposed structured model does not require joint training of the structural semantic dependencies and deep models. Instead they are independent components linked by Grenander's structures. Furthermore, we exploited Grenander's structures as a means to facilitate and enrich the model with fusion of multimodal sensory data; in particular, auditory features with visual features. Overall, we observed improvements in the quality of semantic interpretations using deep models and auditory features in combination with Grenander's structures, reflecting as numerical improvements of up to 11.5% and 12.3% in precision and recall, respectively.

The novelty proposed in this work regarding teaching and training of software engineering is about how the Project-Based Learning approach is performed for developing a web system for managing academic projects. The supporting project is a real-life problem. Its development was planned in four phases: prototype, code refactoring, release-to-manu-facturing and deployment. We employed some strategies such as role-playing, a software house simulation, and the Model-View-Controller pattern combined with Evolutionary Acquisition and the Interdisciplinary Research Project Management framework.

Graph-theoretical methods have successfully provided semantic and structural interpretations of images and videos. A recent paper introduced a pattern-theoretic approach that allows construction of flexible graphs for representing interactions of actors with objects and inference is accomplished by an efficient annealing algorithm. Actions and objects are termed generators and their interactions are termed bonds; together they form high-probability configurations, or interpretations, of observed scenes. This work and other structural methods have generally been limited to analyzing short videos involving isolated actions. Here we provide an extension that uses additional temporal bonds across individual actions to enable semantic interpretations of longer videos. Longer temporal connections improve scene interpretations as they help discard (temporally) local solutions in favor of globally superior ones. Using this extension, we demonstrate improvements in understanding longer videos, compared to individual interpretations of non-overlapping time segments. We verified the success of our approach by generating interpretations for more than 700 video segments from the YouCook data set, with intricate videos that exhibit cluttered background, scenarios of occlusion, viewpoint variations and changing conditions of illumination. Interpretations for long video segments were able to yield performance increases of about 70% and, in addition, proved to be more robust to different severe scenarios of classification errors.

This paper presents studies and the implementation of a novel grid-connected PV system with storage. The goal of this project is to validate the operation and to determine the benefits and limitations of integrating storage with grid-connected PV systems. The objective is to evaluate the viability of using complementary electricity storage for meeting the local evening peak demands - “night-time peak shaving” in a typical load curve of CEMIG's feeder. A novel and versatile control unit—the Bidirectional Intelligent Conversion and Connection Unit (BICC)—has been developed to control the flow of energy from the electricity sources (i.e., PV generator, battery bank, and electric-power grid). In designing this electronic system, we adopted a topology in which the inverter and the maximum power tracking (MPPT) optimize the available power by incorporating two separate and independent converters. In the first phase of the project, a prototype of the system that includes PV with storage and the BICC unit has been installed at GREEN Solar-IPUC in PUC Minas and been under test to evaluate its performance. These trials have been used to optimize system parameters in order to improve the BICC and its interfacing parameters and ensure safe operation of PV-storage system with the grid. This paper reports on the design and advantages of the BICC, the procedures used in the parameter optimization of the prototype, the effectiveness of the BICC in improving the control and connection, conformity to standards, and the viability of scale up for the second phase testing connected to the electric grid (for the utility CEMIG) with storage at the football stadium, Arena do Jacare, located in Sete Lagoas, Minas Gerais.

We present a novel framework, based on Germander's pattern theoretic concepts, for high-level interpretation of video activities. This framework allows us to elegantly integrate ontological constraints and machine learning classifiers in one formalism to construct high-level semantic interpretations that describe video activity. The unit of analysis is a generator that could represent either an ontological label as well as a group of features from a video. These generators are linked using bonds with different constraints. An interpretation of a video is a configuration of these connected generators, which results in a graph structure that is richer than conventional graphs used in computer vision. The quality of the interpretation is quantified by an energy function that is optimized using Markov Chain Monte Carlo based simulated annealing. We demonstrate the superiority of our approach over a purely machine learning based approach (SVM) using more than 650 video shots from the You Cook dataset. This dataset is very challenging in terms of complexity of background, presence of camera motion, object occlusion, clutter, and actor variability. We find significantly improved performance in nearly all cases. Our results show that the pattern theory inference process is able to construct the correct interpretation by leveraging the ontological constraints even when the machine learning classifier is poor and the most confident labels are wrong.

In statistical analysis of video sequences for speech recognition, and more generally activity recognition, it is natural to treat temporal evolutions of features as trajectories on Riemannian manifolds. However, different evolution patterns result in arbitrary parameterizations of these trajectories. We investigate a recent framework from statistics literature that handles this nuisance variability using a cost function/distance for temporal registration and statistical summarization & modeling of trajectories. It is based on a mathematical representation of trajectories, termed transported square-root vector field (TSRVF), and the L2 norm on the space of TSRVFs. We apply this framework to the problem of speech recognition using both audio and visual components. In each case, we extract features, form trajectories on corresponding manifolds, and compute parametrization-invariant distances using TSRVFs for speech classification. On the OuluVS database the classification performance under metric increases significantly, by nearly 100% under both modalities and for all choices of features. We obtained speaker-dependent classification rate of 70% and 96% for visual and audio components, respectively.

This paper presents the design and studies performed in the first-phase implementation of the R&D project D308, in partnership with Energetic Company of Minas Gerais (CEMIG-D). The objective of this project is to evaluate the benefits and limitations of using storage with a grid-connected PV system. Specifically to evaluate injecting the energy stored in batteries charge by the PV into the electric grid in order to flatten the peak of the load curve of a feeder system (with evening peak-electricity consumption experienced in this area). The PV/storage system is being installed in the football stadium Arena do Jacare, located in Sete Lagoas, Minas Gerais. Initial studies have been performed to define the characteristics of the PV array, including surveys of the region characteristics, and solar resource/ climate data. Additionally, the characteristics of the optimum PV generator configuration and module technology have been evaluated, taking into consideration the stadium's limited available area and defined orientation--and the major design criterion to eliminate any shadow projection on photovoltaic modules that could reduce its performance. With the initial design of the PV generator structure and geometry, the system was performance was simuulated under projected conditions. An innovative feature of this project is a bidirectional conversion and connection unit, which has the capacity to store the energy coming from the PV generator and/or the electrical network in a battery bank. The unit can deliver the energy to the grid (a feeder operated by CEMIG in our tests) at specific defined times—such as the peak time in this area between 18:30 and 20:30. At the conclusion of this protype project, a larger scale of the system (photovoltaic system connected to the network with storage) will be installed at GREEN-IPUC/PUCMinas, utilizing the lessons learned and experiences and the equipment developed at the Minas Gerais Arena do Jacare Stadium.

An important problem in speech, and generally activity, recognition is to develop analyses that are invariant to the execution rates. We introduce a theoretical framework that provides a parametrization-invariant metric for comparing parametrized paths on Riemannian manifolds. Treating instances of activities as parametrized paths on a Riemannian manifold of covariance matrices, we apply this framework to the problem of visual speech recognition from image sequences. We represent each sequence as a path on the space of covariance matrices, each covariance matrix capturing spatial variability of visual features in a frame, and perform simultaneous pairwise temporal alignment and comparison of paths. This removes the temporal variability and helps provide a robust metric for visual speech classification. We evaluated this idea on the OuluVS database and the rank-1 nearest neighbor classification rate improves from 32% to 57% due to temporal alignment.

Diffusion-weighted images are increasingly being employed in conjunction with T2-weighted images for accurate diagnosis of cervical cancer. However, these images suffer from eddy-current related and magnetic field inhomogeneity induced distortions. This paper presents an investigation into the reverse gradient technique to correct distortions in diffusion-weighted echo-planar images of human cervix. We hypothesize that the correction aligns the diffusion-weighted images with the T2-weighted images. To allow for any misalignments and to evaluate the correction, a normalized cross-correlation driven template matching strategy was employed to assess the correlation between the corrected image and the T2-weighted image. For b = 0 smm−2 images, mean correlation values of 0.72±0.14 for regions-of-interest drawn around the cervix for 10 patients were achieved. We also found that our hypothesis was most valid in regions encompassing the cervix with a misalignment of 4.4±3.4mm. Results are presented for images acquired at a field-of-view of 100mm using an endovaginal coil developed in-house for cervix imaging at 3T which is the first of its kind in the world.

In this paper we presented a violence detector built on the concept of visual codebooks using linear support vector machines. It differs from the existing works of violence detection in what concern the data representation, as none has considered local spatio-temporal features with bags of visual words. An evaluation of the importance of local spatio-temporal features for characterizing the multimedia content is conducted through the cross-validation method. The results obtained confirm that motion patterns are crucial to distinguish violence from regular activities in comparison with visual descriptors that rely solely on the space domain.

The great evolution of medias and information, that occurred during the XX century, had as one of the main consequences the increase of production speed and transmission of knowledge. The changes, associated to globalization, had modified world-wide the economic, commercial and human relations. The stimulation to the development of the creativity associated with engineering techniques are tools that will result benefits like the improvement of the quality of human's life. The objective of this paper is to present a project elaborated in one of the disciplines of the course of Industrial Engineering, looking, mainly, for the strategy of commercialization and the creative aspect of the product conceived for a group of students. It is considered the development of a product, its process of manufacture and its commercialization, objectifying the local and world-wide markets. The study and research about the uses and customs of some markets in the business is necessary to increase the project's possibility of succeeding. It is waited, as result of the work, that the students will absorb a global vision of the development processes and manufacture of a product that can supply the world-wide markets reaching through this the goal of crossing the borders of the local markets.

This paper presents a fuzzy inference system to work the problem of process scheduling within a computer network. The scheduling system, simulated in the network simulator, acts in two particularly points: first defining the discard priority for the applications according to its characteristics; and further, to redefine their transfer rate, also considering its particularities.

In the current national scene, many actions point at projects of digital inclusion and citizenship. In this context, providing access technologies as a requisite for the implementation of these actions is primordial. In this way, many innovative experiences have been presented in the past few years. This paper presents a study on the powerline communication - PLC technology; as a proposal for a feasible access network for Brazilian Amazon. First, the characteristics of the PLC technology are studied from an implanted indoor prototype at Federal University of Para. The measures used in this prototype serve as input for a created model, from which it is intended to study the system more widely, considering factors such as: scalability, reliability and the physical characteristics.

Internal and external computer network attacks or security threats occur according to standards and follow a set of subsequent steps, allowing us to establish profiles or patterns. This well-known behavior is the basis of signature analysis intrusion detection systems. This work presents a new attack signature model to be applied on network-based intrusion detection systems engines. The AISF (ACME! Intrusion Signature Format) model is built upon XML technology and works on intrusion signature handling and analysis, from storage to manipulation. Using this new model, the process of storing and analyzing information about intrusion signatures for further use by an IDS become a less difficult and standardized process.

Based on the notion of durability and culture proposed by Agenda 21 at the UN Conference in 1992, this text analyses the imposition of products manufactured by industry in the light of globalisation, considering this imposition as a source of environmental degradation: a cultural degradation. The process of industrial production involves design, a tool which looks to balance the interests of industry and the interests of the user. In the perspective of environment and culture, we ask what role design could play.