Hand gesture recognition system based in computer vision and machine learning

"Lecture notes in computational vision and biomechanics series, ISSN 2212-9391, vol. 19"

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.

Strategic planning of electricity systems: Integrating renewable energies

Tese de Doutoramento - Programa Doutoral em Engenharia Industrial e Sistemas (PDEIS)

Modelação de governo electrónico para países emergentes: O caso de São Tomé e Príncipe

Tese de Doutoramento em Tecnologias e Sistemas de Informação

Hand gesture recognition system based in computer vision and machine learning: Applications on human-machine interaction

Tese de Doutoramento em Engenharia de Eletrónica e de Computadores

BIOMedical search engine framework: lightweight and customized implementation of domain-specific biomedical search engines

The Smart Drug Search is publicly accessible at http://sing.ei.uvigo.es/sds/. The BIOMedical Search Engine Framework is freely available for non-commercial use at https://github.com/agjacome/biomsef

Desarrollo de sistemas nano/micro-particulados con aplicación en la fotodegradación selectiva de contaminantes orgánicos en solución. Development of nano / micro-particles systems with application to selective photodegradation of organic pollutants in solution.

IDENTIFICACIÓN DEL PROBLEMA DE ESTUDIO. Las sustancias orgánicas solubles en agua no biodegradables tales como ciertos herbicidas, colorantes industriales y metabolitos de fármacos de uso masivo son una de las principales fuentes de contaminación en aguas subterráneas de zonas agrícolas y en efluentes industriales y domésticos. Las reacciones fotocatalizadas por irradiación UV-visible y sensitizadores orgánicos e inorgánicos son uno de los métodos más económicos y convenientes para la descomposición de contaminantes en subproductos inocuos y/o biodegradables. En muchas aplicaciones es deseable un alto grado de especificidad, efectividad y velocidad de degradación de un dado agente contaminante que se encuentra presente en una mezcla compleja de sustancias orgánicas en solución. En particular son altamente deseables sistemas nano/micro -particulados que formen suspensiones acuosas estables debido a que estas permiten una fácil aplicación y una eficaz acción descontaminante en grandes volúmenes de fluidos. HIPÓTESIS Y PLANTEO DE LOS OBJETIVOS. El objetivo general de este proyecto es desarrollar sistemas nano/micro particulados formados por polímeros de impresión molecular (PIMs) y foto-sensibilizadores (FS). Un PIMs es un polímero especialmente sintetizado para que sea capaz de reconocer específicamente un analito (molécula plantilla) determinado. La actividad de unión específica de los PIMs en conjunto con la capacidad fotocatalizadora de los sensibilizadores pueden ser usadas para lograr la fotodescomposición específica de moléculas “plantilla” (en este caso un dado contaminante) en soluciones conteniendo mezclas complejas de sustancias orgánicas. MATERIALES Y MÉTODOS A UTILIZAR. Se utilizaran técnicas de polimerización en mini-emulsión para sintetizar los sistemas nano/micro PIM-FS para buscar la degradación de ciertos compuestos de interés. Para caracterizar eficiencias, mecanismos y especificidad de foto-degradación en dichos sistemas se utilizan diversas técnicas espectroscópicas (estacionarias y resueltas en el tiempo) y de cromatografía (HPLC y GC). Así mismo, para medir directamente distribuciones de afinidades de unión y eficiencia de foto-degradación se utilizaran técnicas de fluorescencia de molécula/partícula individual. Estas determinaciones permitirán obtener resultados importantes al momento de analizar los factores que afectan la eficiencia de foto-degradación (nano/micro escala), tales como cantidad y ubicación de foto- sensibilizadores en las matrices poliméricas y eficiencia de unión de la plantilla y los productos de degradación al PIM. RESULTADOS ESPERADOS. Los estudios propuestos apuntan a un mejor entendimiento de procesos foto-iniciados en entornos nano/micro-particulados para aplicar dichos conocimientos al diseño de sistemas optimizados para la foto-destrucción selectiva de contaminantes acuosos de relevancia social; tales como herbicidas, residuos industriales, metabolitos de fármacos de uso masivo, etc. IMPORTANCIA DEL PROYECTO. Los sistemas nano/micro-particulados PIM-FS que se propone desarrollar en este proyecto se presentan como candidatos ideales para tratamientos específicos de efluentes industriales y domésticos en los cuales se desea lograr la degradación selectiva de compuestos orgánicos. Los conocimientos adquiridos serán indispensables para construir una plataforma versátil de sistemas foto-catalíticos específicos para la degradación de diversos contaminantes orgánicos de interés social. En lo referente a la formación de recursos humanos, el proyecto propuesto contribuirá en forma directa a la formación de 3 estudiantes de postgrado y 2 estudiantes de grado. En las capacidades institucionales se contribuirá al acondicionamiento del Laboratorio para Microscopía Óptica Avanzada (LMOA) en el Dpto. de Química de la UNRC y al montaje de un sistema de microscopio de fluorescencia que permitirá la aplicación de técnicas avanzadas de espectroscopia de fluorescencia de molecula individual. Water-soluble organic molecules such as certain non-biodegradable herbicides, industrial dyes and metabolites of widespread use drugs are a major source of pollution in groundwater from agricultural areas and in industrial and domestic effluents. Photo-catalytic reactions by UV-visible irradiation and organic sensitizers are one of the most economical and convenient methods for the decomposition of pollutants into harmless byproducts. In many applications it is highly desirable a high degree of specificity, effectiveness and speed of degradation of specific pollutants present in a complex mixture. In particular nano/micro-particles systems that form stable aqueous suspensions are highly desirable because they allow for easy application and effective decontamination of large volumes of fluids. Herein we propose the development of nano/micro particles composed by molecularly imprinted polymers (MIP) and photo-sensitizers (PS). The specific binding of MIP and the photo-catalytic ability of the sensitizers are used to achieve the photo-decomposition of specific "template" molecules in complex mixtures. Mini-emulsion polymerization techniques will be used to synthesize nano/micro MIP-FS systems. Spectroscopy (steady-state and time resolved) and chromatography (GC and HPLC) will be used to characterize efficiency, mechanisms and specificity of photo-degradation in these systems. In addition single molecule/particle fluorescence spectroscopy techniques will be used to directly measure distributions of binding affinities and photo-degradation efficiency in individual particles. The proposed studies point to a more detailed understanding of the factors affecting the photo-degradation efficiency in nano/micro-particles and to apply that knowledge in the design of optimized systems for photo-selective destruction of socially relevant aqueous pollutants.

Quantum phase transitions in magnetic systems: application of coupled cluster method

Magdeburg, Univ., Fak. für Naturwiss., Diss., 2009

Application of Fuzzy Variables for Systems of Management Decision Support

The following article describes an approach covering the variety of opinions and uncertainties of estimates within the chosen technique of decision support. Mathematical operations used for assessment of options are traced to operations of working with functions that are used for assessment of possible options of decision-making. Approach proposed could be used within any technique of decision support based on elementary mathematical operations. In this article the above-mentioned approach is described under analytical hierarchy process.

Minimal-invasive provenance integration into data-intensive systems

Magdeburg, Univ., Fak. für Informatik, Diss., 2014

Model-based diagnosis system: application to constructed wetlands

Report for the scientific sojourn carried out at the Model-based Systems and Qualitative Reasoning Group (Technical University of Munich), from September until December 2005. Constructed wetlands (CWs), or modified natural wetlands, are used all over the world as wastewater treatment systems for small communities because they can provide high treatment efficiency with low energy consumption and low construction, operation and maintenance costs. Their treatment process is very complex because it includes physical, chemical and biological mechanisms like microorganism oxidation, microorganism reduction, filtration, sedimentation and chemical precipitation. Besides, these processes can be influenced by different factors. In order to guarantee the performance of CWs, an operation and maintenance program must be defined for each Wastewater Treatment Plant (WWTP). The main objective of this project is to provide a computer support to the definition of the most appropriate operation and maintenance protocols to guarantee the correct performance of CWs. To reach them, the definition of models which represent the knowledge about CW has been proposed: components involved in the sanitation process, relation among these units and processes to remove pollutants. Horizontal Subsurface Flow CWs are chosen as a case study and the filtration process is selected as first modelling-process application. However, the goal is to represent the process knowledge in such a way that it can be reused for other types of WWTP.

Numerical simulation of combined heat transfer phenomena (conduction, convection, and radiation). Application to he optimisation of thermal systems

Thermal systems interchanging heat and mass by conduction, convection, radiation (solar and thermal ) occur in many engineering applications like energy storage by solar collectors, window glazing in buildings, refrigeration of plastic moulds, air handling units etc. Often these thermal systems are composed of various elements for example a building with wall, windows, rooms, etc. It would be of particular interest to have a modular thermal system which is formed by connecting different modules for the elements, flexibility to use and change models for individual elements, add or remove elements without changing the entire code. A numerical approach to handle the heat transfer and fluid flow in such systems helps in saving the full scale experiment time, cost and also aids optimisation of parameters of the system. In subsequent sections are presented a short summary of the work done until now on the orientation of the thesis in the field of numerical methods for heat transfer and fluid flow applications, the work in process and the future work.

Hydrogeophysical data integration at larger scales: Application of Bayesian sequential simulation for the characterization of heterogeneous alluvial aquifers

Knowledge of the spatial distribution of hydraulic conductivity (K) within an aquifer is critical for reliable predictions of solute transport and the development of effective groundwater management and/or remediation strategies. While core analyses and hydraulic logging can provide highly detailed information, such information is inherently localized around boreholes that tend to be sparsely distributed throughout the aquifer volume. Conversely, larger-scale hydraulic experiments like pumping and tracer tests provide relatively low-resolution estimates of K in the investigated subsurface region. As a result, traditional hydrogeological measurement techniques contain a gap in terms of spatial resolution and coverage, and they are often alone inadequate for characterizing heterogeneous aquifers. Geophysical methods have the potential to bridge this gap. The recent increased interest in the application of geophysical methods to hydrogeological problems is clearly evidenced by the formation and rapid growth of the domain of hydrogeophysics over the past decade (e.g., Rubin and Hubbard, 2005).