Generic system for human-computer gesture interaction: applications on sign language recognition and robotic soccer refereeing

Hand gestures are a powerful way for human communication, with lots of potential applications in the area of human computer interaction. Vision-based hand gesture recognition techniques have many proven advantages compared with traditional devices, giving users a simpler and more natural way to communicate with electronic devices. This work proposes a generic system architecture based in computer vision and machine learning, able to be used with any interface for human-computer interaction. The proposed solution is mainly composed of three modules: a pre-processing and hand segmentation module, a static gesture interface module and a dynamic gesture interface module. The experiments showed that the core of visionbased interaction systems could be the same for all applications and thus facilitate the implementation. For hand posture recognition, a SVM (Support Vector Machine) model was trained and used, able to achieve a final accuracy of 99.4%. For dynamic gestures, an HMM (Hidden Markov Model) model was trained for each gesture that the system could recognize with a final average accuracy of 93.7%. The proposed solution as the advantage of being generic enough with the trained models able to work in real-time, allowing its application in a wide range of human-machine applications. To validate the proposed framework two applications were implemented. The first one is a real-time system able to interpret the Portuguese Sign Language. The second one is an online system able to help a robotic soccer game referee judge a game in real time.

Generic system for human-computer gesture interaction

Hand gestures are a powerful way for human communication, with lots of potential applications in the area of human computer interaction. Vision-based hand gesture recognition techniques have many proven advantages compared with traditional devices, giving users a simpler and more natural way to communicate with electronic devices. This work proposes a generic system architecture based in computer vision and machine learning, able to be used with any interface for humancomputer interaction. The proposed solution is mainly composed of three modules: a pre-processing and hand segmentation module, a static gesture interface module and a dynamic gesture interface module. The experiments showed that the core of vision-based interaction systems can be the same for all applications and thus facilitate the implementation. In order to test the proposed solutions, three prototypes were implemented. For hand posture recognition, a SVM model was trained and used, able to achieve a final accuracy of 99.4%. For dynamic gestures, an HMM model was trained for each gesture that the system could recognize with a final average accuracy of 93.7%. The proposed solution as the advantage of being generic enough with the trained models able to work in real-time, allowing its application in a wide range of human-machine applications.

(Un)suitability of the use of pH buffers in biological, biochemical and environmental studies and their interaction with metal ions - a review

The use of buffers to maintain the pH within a desired range is a very common practice in chemical, biochemical and biological studies. Among them, zwitterionic N-substituted aminosulfonic acids, usually known as Good's buffers, although widely used, can complex metals and interact with biological systems. The present work reviews, discusses and updates the metal complexation characteristics of thirty one commercially available buffers. In addition, their impact on biological systems is also presented. The influences of these buffers on the results obtained in biological, biochemical and environmental studies, with special focus on their interaction with metal ions, are highlighted and critically reviewed. Using chemical speciation simulations, based on the current knowledge of the metal-buffer stability constants, a proposal of the most adequate buffer to employ for a given metal ion is presented.

Ion exchange dependent electroactive phase content and electrical properties of poly(vinylidene fluoride)/Na(M)Y composites

Different metal-ion exchanged NaY zeolite, Na(M)Y, were used to prepare poly(vinylidene fluoride) based composites by solvent casting and melting crystallization. The effect of different metal ion-exchanged zeolites on polymer crystallization and electrical properties was reported. Cation-framework interactions and hydration energy of the cations determined that K+ is the most efficient exchanged ion in NaY zeolite, followed by Cs+ and Li+. The electroactive phase crystallization strongly depends on the ions present in the zeolite, leading to variations of the surface energy characteristics of the Na(M)Y zeolites and the polymer chain ability of penetration in the zeolite. Thus, Na(Li)Y and NaY induces the complete electroactive -phase crystallization of the crystalline phase of PVDF, while Na(K)Y only induces it partly and Na(Cs)Y is not able to promote the crystallization of the electroactive phase. Furthermore, different ion size/weigh and different interaction with the zeolite framework results in significant variations in the electrical response of the composite. In this way, iinterfacial polarization effects in the zeolite cavities and zeolite-polymer interface, leads to strong increases of the dielectric constant on the composites with lightest ions weakly bound to the zeolite framework. Polymer composite with Na(Li)Y show the highest dielectric response, followed by NaY and Na(K)Y. Zeolite Na(Cs)Y contribute to a decrease of the dielectric constant of the composite. The results show the relevance of the materials for sensor development.

Estudo de validação do Partner Interaction Questionnaire para fumantes e ex-fumantes

O presente artigo tem por objetivo validar uma versão adaptada do Partner Interaction Questionnaire (PIQ), instrumento que avalia o suporte para deixar de fumar dado pelo parceiro. É apresentada uma versão adaptada do PIQ para fumantes e abstinentes. Participaram no estudo 224 fumantes (52.7% do sexo feminino e 47.3% do sexo masculino; média de idades=28.6 anos; DP=8.69) e 169 ex-fumantes (68.6% do sexo masculino e 31.4% do sexo feminino; média de idades= 43.1 anos; DP=13.02). O instrumento apresentou boa fidelidade (alfa de Cronbach: 0.73) para a subescala positiva e para a subescala negativa (alfa de Cronbach: 0.77). O estudo de validade revelou a presença de dois fatores que explicam 43.36% da variância. O instrumento apresenta boa validade de construto, correlacionando-se positivamente com o ajustamento de casais e a qualidade de vida mental, e negativamente com a depressão. Esta versão adaptada do PIQ reuniu as condições necessárias para a sua aplicação em fumantes e ex-fumantes, podendo ser uma escala útil em programas de intervenção de cessação tabágica e de manutenção da abstinência.

Carbon materials as new generation of new electron shuttles for the anaerobic degradation of environmental xenobiotics

Tese de Doutoramento em Engenharia Química e Biológica

Analysis of physicochemical interaction between the components of asphalt mixtures with rubber

Tese de Doutoramento em Engenharia Civil

Flexible thin-film lithium battery

Tese de Doutoramento Programa Doutoral em Engenharia Electrónica e Computadores.

Production and characterization of ZrCN-Ag coatings deposited by mganetron sputtering

Tese de Doutoramento (Programa doutoral em Engenharia de Materiais)

Influence of chitosan coating on protein-based nanohydrogels properties and in vitro gastric digestibility

Chitosan coating was applied in Lactoferrin (Lf)-Glycomacropeptide (GMP) nanohydrogels by layer-by-layer coating process. A volume ratio of 0.1 of Lf-GMP nanohydrogels (0.2 mg.mL-1, at pH 5.0) to chitosan (1 mg.mL-1, at pH 3) demonstrated to be the optimal condition to obtain stable nanohydrogels with size of 230 ± 12 nm, a PdI of 0.22 ± 0.02 and a -potential of 30.0 ± 0.15 mV. Transmission electron microscopy (TEM) images showed that the application of chitosan coating in Lf-GMP did not affect the spherical shape of nanohydrogels and confirmed the low aggregation of nanohydrogels in solution. The analysis of chemical interactions between chitosan and Lf-GMP nanohydrogels were performed by Fourier transform infrared spectroscopy (FTIR) and by circular dichroism (CD) that revealed that a specific chemical interaction occurring between functional groups of protein-based nanohydrogels and active groups of the chitosan was established. The effect of chitosan coating on release mechanisms of Lf-GMP nanohydrogels at acid conditions (pH 2, 37 ºC) was evaluated by the encapsulation of a model compound (caffeine) in these systems. Linear Superposition Model was used to fit the experimental data and revealed that Fick and relaxation mechanisms are involved in caffeine release. It was also observed that the Fick contribution increase with the application of chitosan coating. In vitro gastric digestion was performed with Lf-GMP nanohydrogels and Lf-GMP nanohydrogels with chitosan coating and it was observed that the presence of chitosan improve the stability of Lf and GMP (proteins were hydrolysed at a slower rate and were present in solution by longer time). Native electrophoreses revealed that the nanohydrogels without coating remained intact in solution until 15 min and with chitosan coating remained intact until 60 min, during gastric digestion.