INESCC-DL - Artigos em Revistas Internacionais
Permanent URI for this collection
Browse
Recent Submissions
- Synthetic image generation for effective deep learning model training for ceramic industry applicationsPublication . Gaspar, Fábio; Daniel Carreira; Rodrigues, Nuno; Miragaia, Rolando; Ribeiro, José; Costa, Paulo; Pereira, AntónioIn the rapidly evolving field of machine learning engineering, access to large, high-quality, and well-balanced labeled datasets is indispensable for accurate product classification. This necessity holds particular significance in sectors such as the ceramics industry, in which effective production line activities are paramount and deep learning classification mechanisms are particularly relevant for streamlining processes; but real-world image samples are scarce and difficult to obtain, hindering dataset building and consequently model training and deployment. This paper presents a novel approach for dataset building in the context of the ceramic industry, which involves employing synthetic images for building or complementing datasets for image classification problems. The proposed methodology was implemented in CeramicFlow, an innovative computer graphics rendering pipeline designed to create synthetic images by employing computer-aided design models of ceramic objects and incorporating domain randomization techniques. As a result, a fully synthetic image dataset named Synthetic CeramicNet was created and validated in real-world ceramic classification problems. The results demonstrate that synthetic images provide an adequate basis for datasets and can significantly reduce reliance on real-world data when developing deep learning approaches for image classification problems in the ceramic industry. Furthermore, the proposed approach can potentially be applied to other industrial fields.
- The role of beliefs, expectations and values in decision-making favoring climate change adaptation - Implications for communications with European forest professionalsPublication . Blennow, K.; Persson, J.; Gonçalves, Luísa M.S.; Borys, A.; Dutcă, I.; Hynynen, J.; Janeczko, E.; Lyubenova, M.; Merganič, J.; Merganičová, K.; Peltoniemi, M.; Petr, M.; Reboredo, F.; Vacchiano, G.; Reyer, C. P. O.Beliefs, expectations and values are often assumed to drive decisions about climate change adaptation. We tested hypotheses based on this assumption using survey responses from 508 European forest professionals in ten countries. We used the survey results to identify communication needs and the decision strategies at play, and to develop guidelines on adequate communications about climate change adaptation. We observed polarization in the positive and negative values associated with climate change impacts accepted by survey respondents. We identified a mechanism creating the polarization that we call the 'blocked belief' effect. We found that polarized values did not correlate with decisions about climate change adaptation. Strong belief in the local impacts of climate change on the forest was, however, a prerequisite of decision-making favoring adaptation. Decision-making in favor of adaptation to climate change also correlated with net values of expected specific impacts on the forest and generally increased with the absolute value of these in the absence of 'tipping point' behavior. Tipping point behavior occurs when adaptation is not pursued in spite of the strongly negative or positive net value of expected climate change impacts. We observed negative and positive tipping point behavior, mainly in SW Europe and N-NE Europe, respectively. In addition we found that advice on effective adaptation may inhibit adaptation when the receiver is aware of effective adaptation measures unless it is balanced with information explaining how climate change leads to negative impacts. Forest professionals with weak expectations of impacts require communications on climate change and its impacts on forests before any advice on adaptation measures can be effective. We develop evidence-based guidelines on communications using a new methodology which includes Bayesian machine learning modeling of the equivalent of an expected utility function for the adaptation decision problem.
- Autonomous Wireless Sensor with a Low Cost TEG for Application in Automobile VehiclesPublication . Costa, A.; Costa, D.; Morgado, J.; Santos, Helder; Ferreira, Carlos Daniel HenriquesThe present work consists in the development of an autonomous, low cost, reliable, energy scavenger sensor for automotive applications. Thermoelectric generators typically exhibit low efficiency but high reliability, making them suitable for autonomous, low average energy consumption, applications. A prototype sensor was developed for mounting in the engine exhaust pipe using a step-up voltage converter, a microcontroller, temperature and pressure sensing elements, conditioning electronics and a wireless transceiver, all powered by a low cost TEG (Peltier module TEC1-12706), through the scavenging of exhaust gases thermal energy. During the tests the prototype was able to sustain a regular signal transmission throughout the engine operation. The sensor was installed directly at the measuring point eliminating wired cables to hot and vibrating parts, thus, simplifying the installation of components and improving the reliability of the vehicle systems.
- High Sensitivity Micro-machined Piezoresistive Strain SensorPublication . Caseiro, D.; Santos, S.; Ferreira, Carlos Daniel Henriques; Neves, CarlosThis paper presents a micro-machined piezoresistive sensor capable of measuring very small strains. The sensor design, based on piezoresistive sensing technology, was optimized by the numerical method using Finite Element Method (FEM) to enhance sensibility. The high sensibility is achieved through a reduction of section and through the action of the bending moment. As a result, a sensor with a sensitivity of 569.4608 μV/V/με, which can be fabricated by the SensoNor MultiMEMS process, is proposed. Furthermore, practical essays with macro prototypes confirmed and validated the numerical analysis. Such a sensor can be a direct replacement for the strain gauges and its very high sensitivity opens the door to many other applications, that otherwise would not be possible.
- Energy audit as an input for energy management and energy efficiency improvement in a gypsum manufacturing plantPublication . Bernardo, Hermano; Oliveira, Filipe Tadeu; Serrano, LuisThis paper aims at presenting the main results of an energy audit performed to a gypsum production plant, in Portugal, which due to the amount of energy consumed must comply with the Portuguese program SGCIE (Intensive Energy Consumption Management System). The program was created in 2008 to promote energy efficiency and energy consumption monitoring in intensive energy consuming facilities (energy consumption higher than 500 toe per year). Facilities operators are required to perform energy audits and take actions to draw up an action plan for energy efficiency, establishing targets for energy consumption reduction and greenhouse gases emissions indexes. An energy audit was carried out to identify potential energy conservation measures for improving energy efficiency, and also typical energy consumption patterns, sector/equipment load profiles and thermal equipment performance. This tool gives managers the information to support decision making on improving energy performance and reducing greenhouse gas emissions. A number of tangible targets and measures were devised and set to be implemented in the next few years. Results show that there is a considerable potential for reduction in the energy consumption and greenhouse gases emissions of gypsum manufacturing plants. Here, as elsewhere in the industrial sector, energy efficiency can only be achieved through a continuous energy monitoring and management system.
- Rotating speed stability and mechanical vibration analysis of one-stage inertia flexible rotor driven by variable speed drivesPublication . Oliveira, Filipe Tadeu; Donsión, M.P.; Peláez, G.This paper poses some questions as to the adequacy of the widespread use of variable speed drives. It presents a comparison of different variable speed drives, by studying their output wave forms and applying them to a mechanical one stage inertia flexible rotor specially designed for this purpose. The comparison of the output of the variable speed drives used reveals significant differences between them, which affect the way the load is driven and the overall behaviour of the mechanical system. A comparison against a sine wave reference shows that even the best of the variable speed drives used is outperformed, since the “perfect” sine wave ensures constant rotating speed, thus minimizing vibrations along the mechanical shaft.
- Simulation of the effect of voltage transients on an induction motor with ATP/EMTPPublication . Gonçalves, José; Baptista, José; Pires Neves, Luís; Oliveira, Filipe TadeuThe present study aims at understanding the behaviour of the induction motor when subject to different kinds of disturbances. The intention was to broaden the study of some disturbances by using a transient simulation software, ATP/EMTP, and comparing results with laboratory measurements. Obtained simulation results approached very closely the laboratory measurements, with small differences probably due to the model simplifying assumptions and uncertainties associated to the model parameter estimation process. However, although this theoretical model presents a response very similar to the actual model, improvements could be made by better modelling of the mechanical load, and by using different kinds of dynamic loads. Also, the use of more accurate laboratory equipment, and namely, programmable power supplies could lead to more accurate comparisons and better learning.
- Predictable impact of lighting control on the energy consumption of a building through computational simulationPublication . Bernardo, H.; Leitão, S.; Neves, L.; Amaral, P.; Bernardo, Hermano; Pires Neves, LuísBuilding energy simulation tools provide accurate predictions of the energetic performance of buildings and thermal comfort of its occupants, allowing an evaluation of the impact of proposed improvement measures, in order to support choice for the more economically viable. This paper aims at determining the energy saving potential that can be obtained by adequate measures and investments. It presents the simulated values of the impact on the energy consumptions of a building, caused by artificial lighting control systems set to maximize use of natural lighting. Results show that optimization measures have a significant impact on energy consumptions reduction, and lead to important economical savings.
- 2ARTs: A Platform for Exercise Prescriptions in Cardiac Recovery PatientsPublication . Pereira, Andreia; Martinho, Ricardo; Pinto, Rui; Rijo, Rui; Grilo, CarlosDue to limited access, increasing costs and an ageing population, the global healthcare system faces significant coverage problems that call for innovative approaches. Health professionals are actively seeking alternative methods to provide care to an increasingly needy population, without increasing human effort and associated costs. eHealth platforms, which use technology to provide patient care, are emerging as transformative solutions for addressing these problems. This study is centered on the demand for a Decision Support System (DSS) in cardiology to enable doctors to prescribe individualized care inside Cardiac Rehabilitation Programmes (CRPs). The 2ARTs project’s main objective is to include a cardiac rehabilitation platform with a DSS within the hospital infrastructure. This DSS uses models to classify patients into different groups, delivering crucial information to assist with decisions regarding treatment. Regarding the DSS, Principal Component Analysis (PCA) emerged as a standout technique for dimensionality reduction, due to its interoperability with clustering algorithms and superior evaluation metrics. The most appropriate clustering technique was determined to be the K-means algorithm, which was supported by the experts analysis. In accordance with the goals of the 2ARTs project, this integration of PCA and K-means provides meaningful insights that improve reasoned decision-making.
- Systematic Review of Emotion Detection with Computer Vision and Deep LearningPublication . Pereira, Rafael; Mendes, Carla; Ribeiro, José; Ribeiro, Roberto; Miragaia, Rolando; Rodrigues, Nuno; Costa, Nuno; Pereira, AntónioEmotion recognition has become increasingly important in the field of Deep Learning (DL) and computer vision due to its broad applicability by using human–computer interaction (HCI) in areas such as psychology, healthcare, and entertainment. In this paper, we conduct a systematic review of facial and pose emotion recognition using DL and computer vision, analyzing and evaluating 77 papers from different sources under Preferred Reporting Items for Systematic Reviews and MetaAnalyses (PRISMA) guidelines. Our review covers several topics, including the scope and purpose of the studies, the methods employed, and the used datasets. The scope of this work is to conduct a systematic review of facial and pose emotion recognition using DL methods and computer vision. The studies were categorized based on a proposed taxonomy that describes the type of expressions used for emotion detection, the testing environment, the currently relevant DL methods, and the datasets used. The taxonomy of methods in our review includes Convolutional Neural Network (CNN), Faster Region-based Convolutional Neural Network (R-CNN), Vision Transformer (ViT), and “Other NNs”, which are the most commonly used models in the analyzed studies, indicating their trendiness in the field. Hybrid and augmented models are not explicitly categorized within this taxonomy, but they are still important to the field. This review offers an understanding of state-of-the-art computer vision algorithms and datasets for emotion recognition through facial expressions and body poses, allowing researchers to understand its fundamental components and trends.