Tavares P.,University of Porto |
Lima J.,INESC TEC |
Costa P.,INESC TEC |
Moreira A.P.,INESC TEC
ACM International Conference Proceeding Series | Year: 2015
The fast development in the field of Robotics has become more and more notorious throughout the past years. Nowadays technology in general and robotics in particular search for modular and flexible applications in order to answer the demands of a wide range of problems in an efficient manner. However there are several robotic solutions already implemented and some still available to be implemented that do not use modular tools. The Robotic Operative System (ROS) appears to be the unifying tool to connect all software developers allowing any developer in both education and professional areas to be able to develop complex software using small iterations of simple software. Still, despite of the several robotic solutions available, there are several robots that do not use the Robotic Operative System (ROS) and have limitations in terms of autonomously correct errors during their tasks. Moreover when developing new robots and software to the robotic area there is an important aspect to be consider: The selection of the methodology to be used. In this paper, it will be presented a challenge propose to college students using the ROS framework in a common robotic problem, the pick and place operations. The main aim for this challenge is to show how to produce software in a modular and flexible way using ROS can prompt the rapid development in all robotic applications. Moreover the challenge had one particular real end, the European Robotics Challenges (EUROC) - a challenge aiming to develop a robot for shop floor logistics and manipulation. Furthermore this challenge was based in the three tiers paradigm: 1 recognition/sensing tier, 2-effector tier and 3-The control tier and was built using the ROS framework. Another advantage of our proposed pick and place approach is the ability to have a robot safely and efficiently inserted in an unknown environment. This is possible due to the insertion of an adaptive control tier in our methodology. The proposed approach can be valuable in the field of robotics and can be potentially applied in multiple tasks and it has already allowed us to advance to the next stage of EUROC. Based on this information, the challenge propose to the students will primarily reinforce the need for modular and flexible software while showing how the ROS framework can be a simple tool for present and future developments. © 2015 ACM.
Costa P.,INESC TEC |
Campilho A.,INESC TEC
Proceedings of the 15th IAPR International Conference on Machine Vision Applications, MVA 2017 | Year: 2017
This paper describes a methodology for Diabetic Retinopathy detection from eye fundus images using a generalization of the Bag-of-Visual-Words (BoVW) method. We formulate the BoVW as two neural networks that can be trained jointly. Unlike the BoVW, our model is able to learn how to perform feature extraction, feature encoding and classification guided by the classification error. The model achieves 0.97 Area Under the Curve (AUC) on the DR2 dataset while the standard BoVW approach achieves 0.94 AUC. Also, it performs at the same level of the state-of-the-art on the Messidor dataset with 0.90 AUC. © 2017 MVA Organization All Rights Reserved.
Coelho F.,INESC TEC |
Paulo J.,INESC TEC |
Vilaca R.,LeanXcale |
Pereira J.,INESC TEC |
Oliveira R.,INESC TEC
ICPE 2017 - Proceedings of the 2017 ACM/SPEC International Conference on Performance Engineering | Year: 2017
The increasing demand for real-time analytics requires the fusion of Transactional (OLTP) and Analytical (OLAP) systems, eschewing ETL processes and introducing a plethora of proposals for the so-called Hybrid Analytical and Trans-actional Processing (HTAP) systems. Unfortunately, current benchmarking approaches are not able to comprehensively produce a unified metric from the assessment of an HTAP system. The evaluation of both engine types is done separately, leading to the use of disjoint sets of benchmarks such as TPC-C or TPC-H. In this paper we propose a new benchmark, HTAPBench, providing a unified metric for HTAP systems geared toward the execution of constantly increasing OLAP requests limited by an admissible impact on OLTP performance. To achieve this, a load balancer within HTAPBench regulates the coexistence of OLTP and OLAP workloads, proposing a method for the generation of both new data and requests, so that OLAP requests over freshly modified data are comparable across runs. We demonstrate the merit of our approach by validating it with different types of systems: OLTP, OLAP and HTAP; showing that the benchmark is able to highlight the differences between them, while producing queries with comparable complexity across experiments with negligible variability. © 2017 ACM.
Teixeira Lopes C.,INESC TEC |
Paiva D.,University of Porto |
Ribeiro C.,INESC TEC
Journal of the Association for Information Science and Technology | Year: 2017
Searching for health information is one of the most popular activities on the web. In this domain, users often misspell or lack knowledge of the proper medical terms to use in queries. To overcome these difficulties and attempt to retrieve higher-quality content, we developed a query suggestion system that provides alternative queries combining the Portuguese or English language with lay or medico-scientific terminology. Here we evaluate this system's impact on the medical accuracy of the knowledge acquired during the search. Evaluation shows that simply providing these suggestions contributes to reduce the quantity of incorrect content. This indicates that even when suggestions are not clicked, they are useful either for subsequent queries' formulation or for interpreting search results. Clicking on suggestions, regardless of type, leads to answers with more correct content. An analysis by type of suggestion and user characteristics showed that the benefits of certain languages and terminologies are more perceptible in users with certain levels of English proficiency and health literacy. This suggests a personalization of this suggestion system toward these characteristics. Overall, the effect of language is more preponderant than the effect of terminology. Clicks on English suggestions are clearly preferable to clicks on Portuguese ones. © 2017 ASIS&T.
Miranda V.,INESC TEC |
Miranda V.,University of Porto |
Castro A.R.G.,Federal University of Pará |
Lima S.,Federal University of Maranhão
IEEE Transactions on Power Delivery | Year: 2012
This paper presents a new approach to incipient fault diagnosis in power transformers, based on the results of dissolved gas analysis. A set of autoassociative neural networks or autoencoders is trained, so that each becomes tuned with a particular fault mode or no fault condition. The scarce data available forms clusters that are densified using an Information Theoretic Mean Shift algorithm, allowing all real data to be used in the validation process. Then, a parallel model is built where the autoencoders compete with one another when a new input vector is entered and the closest recognition is taken as the diagnosis sought. A remarkable accuracy of 100% is achieved with this architecture, in a validation data set using all real information available. © 2012 IEEE.
Cota M.P.,University of Vigo |
Thomaschewski J.,Emden Leer University of Applied Sciences |
Schrepp M.,SAP |
Goncalves R.,INESC TEC
Procedia Computer Science | Year: 2013
For the international use of software products it is important to know the culture, language and behavior of the citizens. Means: internationalization and location. To be able to evaluate these products it is significant to know how people from a country behave and express their feelings. This article presents a questionnaire which was initially developed in German, English and Spanish and is now available after a complex transformation in Portuguese. © 2013 The Authors. Published by Elsevier B.V.
Pinto J.P.,INESC TEC |
Viana P.,Polytechnic Institute of Porto
ACM International Conference Proceeding Series | Year: 2015
This short paper presents a game for collecting metadata to describe video content. Tags, introduced by registered players on a given timecode of the video, are collected and validated based on a collaborative scoring mechanism that excludes erratic annotations. The system follows a gamification approach for motivating users and includes processes for semantically relating concepts. © 2015 ACM.
Lima S.L.,Federal University of Maranhão |
Saavedra O.R.,Federal University of Maranhão |
Miranda V.,INESC TEC |
Miranda V.,University of Porto
IEEE Transactions on Power Delivery | Year: 2015
Power transformers are important equipment of a substation from the generation, transmission, and distribution of electricity to end users. The costs associated with purchasing a power transformer in the voltage class of 500 kV (100 MVA) are a few million. The fines imposed by regulatory agencies are significant when there is unavailability of equipment for any defect or failure. Therefore, energy companies have been struggling in preventive and predictive maintenance in order to maintain this equipment in an effective maintenance program, avoiding the occurrence of failures. There are various techniques that are utilized for diagnosis and analysis of transformer failure, but little has been discussed about mechanisms that assist in decision making when it is necessary to overload the transformer, especially in emergency situations. In this paper, we present a framework that unifies the step of fault diagnosis of power transformers with the process of decision making, considering the current operating conditions as well as the life of the equipment. The assistance to the decision-making methodology is based on risk analysis with indicators derived from the failure rate and Arrhenius theory. The validation of the method is performed with a case study using data from a utility. © 1986-2012 IEEE.
Carvalho L.,University of Porto |
Campos R.,INESC TEC |
Ricardo M.,University of Porto
2013 IEEE 15th International Conference on e-Health Networking, Applications and Services, Healthcom 2013 | Year: 2013
Typical sensor networks are formed by low-end, battery operated devices, which rely on low-energy communication technologies, such as Bluetooth, Zigbee and ANT+, due to their energy efficiency. On the other hand, sensor networks increasingly need to be connected to the Internet, which implies adaptations of the TCP/IP stack to fit such wireless technologies. These adaptations bring additional complexity and imply new hardware, thus deployments are cumbersome and sub-optimal. Conversely, Wi-Fi is ubiquitous, can be seamlessly integrated with TCP/IP, and is energy-efficient with the right configurations; yet, its usage is still uncommon in e-health scenarios. For these reasons, we argue that a TCP/IP over Wi-Fi approach should be followed in e-health sensor networks. We propose a novel cross-layer, context-aware network configuration mechanism, which monitors the user and networking contexts and optimizes the configuration of the TCP/IP protocol stack accordingly. Our approach enables seamless integration between e-health wireless sensor networks and the TCP/IP backbone, while improving energy efficiency and reliability. © 2013 IEEE.
Almeida R.,INESC TEC |
Cruz N.,INESC TEC |
Matos A.,INESC TEC
OCEANS 2016 - Shanghai | Year: 2016
This paper presents a new generation of man portable acoustic navigation buoys. The aim of these buoys is to facilitate the deployment of an underwater acoustic positioning system for the operation of Autonomous Underwater Vehicles. Each buoy includes only the vital modules required for the most typical schemes of underwater acoustic navigation, packed in a small but dynamically stable platform for one day long operations in coastal waters. We will present an overview of the systems hardware and electronics, and also the key features of the deployment and operation of the beacons. © 2016 IEEE.