IPT - Ci2 - Artigos em Revistas Científicas
URI permanente para esta coleção:
Navegar
Percorrer IPT - Ci2 - Artigos em Revistas Científicas por título
A mostrar 1 - 10 de 17
Resultados por página
Opções de ordenação
- Audio Features for Music Emotion Recognition: a SurveyPublication . Panda, Renato; Malheiro, Ricardo; Paiva, Rui PedroThe design of meaningful audio features is a key need to advance the state-of-the-art in Music Emotion Recognition (MER). This work presents a survey on the existing emotionally-relevant computational audio features, supported by the music psychology literature on the relations between eight musical dimensions (melody, harmony, rhythm, dynamics, tone color, expressivity, texture and form) and specific emotions. Based on this review, current gaps and needs are identified and strategies for future research on feature engineering for MER are proposed, namely ideas for computational audio features that capture elements of musical form, texture and expressivity that should be further researched. Finally, although the focus of this article is on classical feature engineering methodologies (based on handcrafted features), perspectives on deep learning-based approaches are discussed.
- Bioenergy routes for valorizing constructed wetland vegetation: An overviewPublication . Pinho, Henrique J. O.; Mateus, D. M. R.Valorizing constructed wetlands vegetation into biofuels can be a way to contribute to mitigating the increasing energy demand, avoiding the use of arable land, freshwater, and fertilizers consumption, while simultaneously treating wastewater with eco-friendly technology. This work shortly overviews the main genera of wetland plants and the main routes of vegetal biomass conversion into biofuels including biochemical and thermochemical processes, and through a cross-search, in the Scopus database, the research intensity in bioenergy application for each genus was assessed. A total of 283 genera of wetland plants were identified and classified into five groups, from very common to very rare genera. The very common group includes 10 genera and contributes to 62% of the literature hits, while the 147 genera classified as very rare contribute to only 3% of the hits. Concerning the bioenergy applications, four genera stand out from the remaining. The plants of the genus Sorghum are the most referred to in bioenergy applications, followed by the genera Brassica, Miscanthus, and Saccharum. Miscanthus is a less common wetland plant, while the other genera are rarely applied in constructed wetlands. The relevance of bioenergy routes depends on the plants' group. For common wetland plants, the most relevant applications are biogas production, followed by bio-ethanol production, and pyrolysis processing. As a recommendation for future research works the genera with high energy potential should be evaluated as wetland vegetation, and it is recommended that the goal to recover wetland vegetation for bioenergy applications be viewed as an integral step of the design and implementation of constructed wetlands facilities.
- A Comparison Study of Deep Learning Methodologies for Music Emotion RecognitionPublication . Louro, Pedro; Redinho, Hugo; Malheiro, Ricardo; Paiva, Rui Pedro; Panda, RenatoClassical machine learning techniques have dominated Music Emotion Recognition. However, improvements have slowed down due to the complex and time-consuming task of handcrafting new emotionally relevant audio features. Deep learning methods have recently gained popularity in the field because of their ability to automatically learn relevant features from spectral representations of songs, eliminating such necessity. Nonetheless, there are limitations, such as the need for large amounts of quality labeled data, a common problem in MER research. To understand the effectiveness of these techniques, a comparison study using various classical machine learning and deep learning methods was conducted. The results showed that using an ensemble of a Dense Neural Network and a Convolutional Neural Network architecture resulted in a state-of-the-art 80.20% F1 score, an improvement of around 5% considering the best baseline results, concluding that future research should take advantage of both paradigms, that is, combining handcrafted features with feature learning.
- Decoding Musical Valence and Arousal: Exploring the Neural Correlates of Music-Evoked Emotions and the Role of Expressivity FeaturesPublication . Sayal, Alexandre; Guedes, Ana Gabriela; Almeida, Inês A. T.; Jardim Pereira, Daniela; Lima, César F.; Panda, Renato; Paiva, Rui Pedro; Sousa, Teresa; Castelo-Branco, Miguel; Bernardino, Inês; Direito, BrunoMusic conveys both basic emotions, like joy and sadness, and complex ones, such as tenderness and nostalgia. Its effects on emotion regulation and reward have attracted much research attention, as the neural correlates of music-evoked emotions may inform neurorehabilitation interventions. Here, we used fMRI to decode and examine the neural correlates of perceived valence and arousal in music excerpts. Twenty participants were scanned while listening to 96 music excerpts, classified beforehand into four categories varying in valence and arousal. Music modulated activity in cortical regions, most noticeably in music-specific subregions of the auditory cortex, thalamus, and regions of the reward network such as the amygdala. Using multivoxel pattern analysis, we created a computational model to decode the perceived valence and arousal of the music excerpts with above-chance accuracy. We further explored associations between musical features and brain activity in valence-, arousal-, reward-, and auditory-related networks. The results emphasize the involvement of distinct musical features, notably expressive features such as vibrato and tonal and spectral dissonance in valence, arousal, and reward brain networks. Using ecologically valid music stimuli, we contribute to delineating the neural correlates of music-evoked emotions with potential implications in the development of novel music-based neurorehabilitation strategies.
- Design of a Mobile Augmented Reality Platform with Game-Based Learning PurposesPublication . Costa, Maria Cristina; Manso, António; Patrício, JoãoAugmented reality (AR) is an emergent technology that overlays virtual objects into the real environment. Lately, AR is gaining prominence in education because of its increasing affordability through computers or mobile technologies. In addition, research sustaining the benefits of AR to promote student’s engagement to learn is increasing every day. However, the literature identifies lack of studies about the use of AR in education, namely, studies focused on the development of AR games designed over specific learning objectives (game-based learning). This paper presents a mobile augmented reality platform with learning purposes. The platform includes a mobile application that consists of a location-based game targeted to promote learning about the universe. Furthermore, it includes a back-office that allows teachers to introduce information about celestial bodies and also develop a set of multiple-choice questions to assess student’s learning about the subject matters they teach. The mobile application provides the users with physical movement and social interaction in the real world, while playing the game and for this reason it is included in the pervasive games’ paradigm. Besides engaging the students to play the game, we argue that this platform may be used as a resource to be implemented in informal and formal learning environments.
- Envisaging a global infrastructure to exploit the potential of digitised collectionsPublication . Groom, Quentin; Dillen, Mathias; Addink, Wouter; Ariño, Arturo H.; Bölling, Christian; Bonnet, Pierre; Cecchi, Lorenzo; Ellwood, Elizabeth R.; Figueira, Rui; Gagnier, Pierre-Yves; Grace, Olwen; Güntsch, Anton; Hardy, Helen; Huybrechts, Pieter; Hyam, Roger; Joly, Alexis; Kommineni, Vamsi Krishna; Larridon, Isabel; Livermore, Laurence; Lopes, Ricardo Jorge; Meeus, Sofie; Miller, Jeremy; Milleville, Kenzo; Panda, Renato; Pignal, Marc; Poelen, Jorrit; Ristevski, Blagoj; Robertson, Tim; Rufino, Ana C.; Santos, Joaquim; Schermer, Maarten; Scott, Ben; Seltmann, Katja; Teixeira, Heliana; Trekels, Maarten; Gaikwad, JitendraTens of millions of images from biological collections have become available online over the last two decades. In parallel, there has been a dramatic increase in the capabilities of image analysis technologies, especially those involving machine learning and computer vision. While image analysis has become mainstream in consumer applications, it is still used only on an artisanal basis in the biological collections community, largely because the image corpora are dispersed. Yet, there is massive untapped potential for novel applications and research if images of collection objects could be made accessible in a single corpus. In this paper, we make the case for infrastructure that could support image analysis of collection objects. We show that such infrastructure is entirely feasible and well worth investing in.
- Evaluation of solid waste stratified mixtures as constructed wetland fillers under different operation modesPublication . Mateus, D.M.R.; Pinho, Henrique J. O.Based on the strategy that all processes can and should be modified to contribute to a circular economy, this work evaluates the recovery of waste solids as filler material in Constructed Wetlands (CWs) used for wastewater treatment. Five sets of lab-scale CWs were assembled with mixtures of five waste solids and operated to evaluate the removal of chemical oxygen demand (COD) and nutrients from urban and industrial-types of wastewater. The adaptation and growth of the macrophyte Phragmites australis in the mixed-filler CWs was also monitored. Although all evaluated waste solids showed to be acceptable substrates for macrophyte development and wastewater treatment, CWs assembled with mixtures of limestone waste and coal slag showed the best plant growth indicators and wastewater treatment efficiencies. The CWs assembled with mixtures of limestone waste and clay brick fragments or cork granulates showed to be suitable alternatives. With exception for CWs filled with mixtures of limestone and snail shells, pollutant removal efficiencies up to 95%, 86% and 83% were obtained respectively for COD, total phosphorous and total nitrogen, depending on the type of wastewater and mode of operation. Removal efficiencies were not significantly affected by increased hydraulic rate. The CWs can be operated to a hydraulic loading rate of 0.056 m/d, which corresponds to a retention time of 1.5 days. The valorisation of solid waste as a filler can contribute to CWs closely participating in the creation of circular flows for the reuse of waste solids.
- Isolation, Identification, and Characterization of Phosphate-Solubilizing Bacteria from Tunisian SoilsPublication . Amri, Marwa; Rjeibi, Mohamed Ridha; Gatrouni, Marwa; Mateus, D. M. R.; Asses, Nedra; Pinho, Henrique J. O.; Abbes, ChaabaneSoil microorganisms play an important role in maintaining natural ecological balance through active participation in carbon, nitrogen, sulfur, and phosphorous cycles. Phosphate-solubilizing bacteria (PSB) are of high importance in the rhizosphere, enhancing the solubilization of inorganic phosphorus complexes into soluble forms available for plant nutrition. The investigation of this species of bacteria is of major interest in agriculture, as they can be used as biofertilizers for crops. In the present study, 28 isolates of PSB were obtained after the phosphate enrichment of soil samples from five Tunisian regions. Five PSB species were identified by 16S rRNA gene sequencing including Pseudomonas fluorescens, P. putida, and P. taiwanensis, Stenotrophomonas maltophilia, and Pantoea agglomerans. Solid and liquid Pikovskaya’s (PVK) and National Botanical Research Institute’s (NBRIP) media containing insoluble tricalcium phosphate were used for the evaluation of the phosphate solubilization ability of the bacterial isolates by two methods: visual evaluation of the solubilization zone around colonies (halo) and determination of solubilized phosphates in liquid medium by the colorimetric method of the vanado-molybdate yellow. Based on the results of the halo method, the isolate of each species that showed the higher phosphate solubilization index was selected for evaluation of phosphate solubilization by the colorimetric method. In the liquid media, the bacterial isolates showed phosphate solubilization ranging from 535.70 to 618.57 µg mL−1 in the NBRIP medium, and 374.20 to 544.28 µg mL−1 in the PVK medium, with the highest values produced by P. fluorescens. The best phosphate solubilization ability and higher reduction in broth pH, which indicates higher organic acid production, were achieved in NBRIP broth for most of the PSB. Strong correlations were observed between the average capability of PSB to solubilize phosphates and both the pH and total phosphorous content in the soil. The production of the hormone indole acetic acid (IAA), which can promote plant growth, was observed for all five PSB species. Among them, P. fluorescens obtained from the forest soil of northern Tunisia showed the highest production of IAA (50.4 ± 0.9 µg mL−1).
- Participation of students in the project Valorbio: A case study to accelerate the implementation of sustainability principles in the curriculumPublication . Mateus, D. M. R.; Pinho, Henrique J. O.; Nogueira, Isabel; Rosa, Manuel; Cartaxo, Marco; Nunes, ValentimPurpose – The present work describes the case of the Valorbio research project, in which students of different high-level programs were involved in the experimental work and in the dissemination of results in collaboration with the research team. Design/methodology/approach – The inclusion in higher education curricula of content related to the sustainable development should be a preferred mechanism for the dissemination of good practices of sustainability. Another equally important way to achieve this is to involve students in research projects that seek solutions to the societal challenges related to sustainable growth. The Valorbio project aims to meet the needs for treating and reusing wastewater and solid waste. Its main goal was the development of modular systems for wastewater treatment based on constructed wetlands,exploring the possibility of the treatment systems being composed of solid waste and by-products from significant industrial sectors. Findings – The students’ contributionto the research workwas relevant, and simultaneously allowedthem to acquire skills on sustainable development. Additionally, the students contributed to the dissemination of the results. The Valorbio project can thus be considered a successful application of the concept of project-based learning,as a way to include sustainability issues content in the higher education curricula. Originality/value – The applied experimental work had an original approach regarding the equipment design, the waste materials valuation, as well as the integration of waste treatment processes in the circular economy paradigm.This paper is the first reported project-based learning experience involving students of short-cycle technical-professional programs in partnership with first and second-level students and a research team.
- Percussion and Instrumentation in Music Emotion Recognition: a Feature Engineering ApproachPublication . Redinho, Hugo; Lima Louro, Pedro Miguel; Santos, André C.; Malheiro, Ricardo; Pinto de Carvalho e Paiva, Rui Pedro; Panda, RenatoWe propose a new set of features for audio-based Music Emotion Recognition (MER) that are related to percussion and individual instrument information. One limitation of current feature engineering approaches in MER is that they primarily focus on melodic elements. However, the percussive elements and instrumentation are also essential for conveying and recognizing emotions in music. Our approach leverages the Demucs framework for music source separation (which enables drum channel separation) and the MT3 framework for automatic music transcription and instrument recognition. Building on the results of these frameworks, we created a new set of features that primarily capture information about musical texture, rhythm, dynamics, expressivity, tone color, and musical form. To validate our work, we utilized the MERGE dataset, which comprises over 3000 30-second audio clips annotated with Russell's emotion quadrants. To evaluate the impact of the new features, we compared classification results with those obtained using current state-of-the-art features, demonstrating statistically significant improvements in F1 score (from 71.1\% to 74.2\%). Moreover, the novel features helped to reduce the confusion between quadrants 3 and 4 (a common difficulty in MER models). The most significant finding of the present study is the impact of separately analyzing the drum channel, whose features proved particularly relevant.