Hand gesture recognition for human computer interaction : a comparative study of different image features

Hand gesture recognition for human computer interaction, being a natural way of human computer interaction, is an area of active research in computer vision and machine learning. This is an area with many different possible applications, giving users a simpler and more natural way to communicate with robots/systems interfaces, without the need for extra devices. So, the primary goal of gesture recognition research is to create systems, which can identify specific human gestures and use them to convey information or for device control. For that, vision-based hand gesture interfaces require fast and extremely robust hand detection, and gesture recognition in real time. In this study we try to identify hand features that, isolated, respond better in various situations in human-computer interaction. The extracted features are used to train a set of classifiers with the help of RapidMiner in order to find the best learner. A dataset with our own gesture vocabulary consisted of 10 gestures, recorded from 20 users was created for later processing. Experimental results show that the radial signature and the centroid distance are the features that when used separately obtain better results, with an accuracy of 91% and 90,1% respectively obtained with a Neural Network classifier. These to methods have also the advantage of being simple in terms of computational complexity, which make them good candidates for real-time hand gesture recognition.

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.

Dispersion and re-agglomeration of graphite nanoplates in polypropylene melts under controlled flow conditions

The kinetics of GnP dispersion in polypropylene melt was studied using a prototype small scale modular extensional mixer. Its modular nature enabled the sequential application of a mixing step, melt relaxation, and a second mixing step. The latter could reproduce the flow conditions on the first mixing step, or generate milder flow conditions. The effect of these sequences of flow constraints upon GnP dispersion along the mixer length was studied for composites with 2 and 10 wt.% GnP. The samples collected along the first mixing zone showed a gradual decrease of number and size of GnP agglomerates, at a rate that was independent of the flow conditions imposed to the melt, but dependent on composition. The relaxation zone induced GnP re-agglomeration, and the application of a second mixing step caused variable dispersion results that were largely dependent on the hydrodynamic stresses generated.

Dynamic group formation in mobile computer supported collaborative learning environment

Forming suitable learning groups is one of the factors that determine the efficiency of collaborative learning activities. However, only a few studies were carried out to address this problem in the mobile learning environments. In this paper, we propose a new approach for an automatic, customized, and dynamic group formation in Mobile Computer Supported Collaborative Learning (MCSCL) contexts. The proposed solution is based on the combination of three types of grouping criteria: learner’s personal characteristics, learner’s behaviours, and context information. The instructors can freely select the type, the number, and the weight of grouping criteria, together with other settings such as the number, the size, and the type of learning groups (homogeneous or heterogeneous). Apart from a grouping mechanism, the proposed approach represents a flexible tool to control each learner, and to manage the learning processes from the beginning to the end of collaborative learning activities. In order to evaluate the quality of the implemented group formation algorithm, we compare its Average Intra-cluster Distance (AID) with the one of a random group formation method. The results show a higher effectiveness of the proposed algorithm in forming homogenous and heterogeneous groups compared to the random method.

The quantified-Self and wearable technologies in the workplace: implications and challenges for their implementations

Introduction of technologies in the workplace have led to a dramatic change. These changes have come with an increased capacity to gather data about one’s working performance (i.e. productivity), as well as the capacity to track one’s personal responses (i.e. emotional, physiological, etc.) to this changing workplace environment. This movement of self-monitoring or self-sensing using diverse types of wearable sensors combined with the use of computing has been identified as the Quantified-Self. Miniaturization of sensors, reduction in cost and a non-stop increase in the computer power capacity has led to a panacea of wearables and sensors to track and analyze all types of information. Utilized in the personal sphere to track information, a looming question remains, should employers use the information from the Quantified-Self to track their employees’ performance or well-being in the workplace and will this benefit employees? The aim of the present work is to layout the implications and challenges associated with the use of Quantified-Self information in the workplace. The Quantified-Self movement has enabled people to understand their personal life better by tracking multiple information and signals; such an approach could allow companies to gather knowledge on what drives productivity for their business and/or well-being of their employees. A discussion about the implications of this approach will cover 1) Monitoring health and well-being, 2) Oversight and safety, and 3) Mentoring and training. Challenges will address the question of 1) Privacy and Acceptability, 2) Scalability and 3) Creativity. Even though many questions remain regarding their use in the workplace, wearable technologies and Quantified-Self data in the workplace represent an exciting opportunity for the industry and health and safety practitioners who will be using them.

Predictors of dropout in a controlled clinical trial of psychotherapy for moderate depression

A significant number of psychotherapy clients remain untreated, and dropping out is one of the main reasons. Still, the literature around this subject is incoherent. The present study explores potential pre-treatment predictors of dropout in a sample of clients who took part in a clinical trial designed to test the efficacy of narrative therapy for major depressive disorder compared to cognitive-behavioral therapy. Logistic regression analysis showed that: (1) treatment assignment did not predict dropout, (2) clients taking psychiatric medication at intake were 80% less likely to drop out from therapy, compared to clients who were not taking medication, and (3) clients presenting anxious comorbidity at intake were 82% less likely to dropout compared to those clients not presenting anxious comorbidity. Results suggest that clinicians should pay attention to depressed clients who are not taking psychiatric medication or have no comorbid anxiety. More research is needed in order to understand this relationship.

Light-controlled resistive switching in laser-assisted annealed Ba0.8Sr0.2TiO3 thin films

In this work, Ba0.8Sr0.2TiO3 (BST)/ITO structures were grown on glass substrate and laser assisted annealing (LAA) was performed to promote the crystallization of BST. Atomic force microscopy and X-ray diffraction studies confirm the crack free and polycrystalline perovskite phase of BST. White light controlled resistive switching (RS) effect in Au/BST/ITO device is investigated. The device displays the electroforming-free bipolar RS characteristics and are explained by the modulationof the width and height of barrier at the BST/ITO interface via ferroelectric polarization. Moreover, the RS effect is signifi- cantly improved under white light illumination compared to that in the dark. The enhanced RS and photovoltaic effects are explained by considering depolarization field and charge distribution at the interface. The devices exhibit stable retention characteristics with low currents (mA), which make them attractive for non volatile memory devices.

Cognitive stimulation for portuguese older adults with cognitive impairment: a randomized controlled trial of efficacy, comparative duration, feasibility and experiential relevance

Although some studies point to cognitive stimulation as a beneficial therapy for older adults with cognitive impairments, this area of research and practice is still lacking dissemination and is underrepresented in many countries. Moreover, the comparative effects of different intervention durations remain to be established and, besides cognitive effects, pragmatic parameters, such as cost-effectiveness and experiential relevance to participants, are seldom explored. In this work, we present a randomized con- trolled wait-list trial evaluating 2 different intervention durations (standard 1⁄4 17 vs brief 1⁄4 11 sessions) of a cognitive stimulation program developed for older adults with cognitive impairments with or without dementia. 20 participants were randomly assigned to the standard duration intervention program (17 sessions, 1.5 months) or to a wait-list group. At postintervention of the standard intervention group, the wait-list group crossed over to receive the brief intervention program (11 sessions, 1 month). Changes in neuropsychological, functionality, quality of life, and caregiver outcomes were evaluated. Experience during intervention and costs and feasibility were also evaluated. The current cognitive stimulation programs (ie, standard and brief) showed high values of experiential relevance for both intervention durations. High adherence, completion rates, and reasonable costs were found for both formats. Further studies are needed to definitively establish the potential efficacy, optimal duration, cost-effectiveness, and experiential relevance for participants of cognitive intervention approaches.

Controlled aggregation of gold nanoparticles in a di-ureasilmatrix. Optical and micro indentation investigation

Nanocomposite materials with an organic-inorganic urea-silicate (di-ureasil) based matrix containing gold nanoparticles (NPs) were synthesized and characterized by optical (UV/Vis) spectroscopy and indentation measurement. The urea silicate gels were obtained by reaction between silicon alkoxyde modified by isocyanate group and polyethylene glycol oligomer with amine terminal groups in presence of catalyst. The latter ensures the successful incorporation of citrate-stabilized gold NPs in the matrix. It is shown that using a convenient destabilizing agent (AgNO3) and governing the preparative conditions, the aggregation degree of gold NPs can be controlled. The developed synthesis procedure significantly simplifies the preparative procedure of gold/urea silicate nanocomposites, compared to the procedure using gold NPs, preliminary covered with silica shells. Mechanical properties of the prepared sample were characterised using depth sensing indentation methods (DSI) and an idea about the type of aggregation structures was suggested.

Light responsive multilayer surfaces with controlled spatial extinction capability

Multilayer systems obtained using the Layer-by-Layer (LbL) technology have been proposed for a variety of biomedical applications in tissue engineering and regenerative medicine. LbL assembly is a simple and highly versatile method to modify surfaces and fabricate robust and highly-ordered nanostructured coatings over almost any type of substrates and with a wide range of substances. The incorporation of polyoxometalate (POM) inorganic salts as constituents of the layers presents a possibility of promoting light-stimuli responses in LbL substrates. We propose the design of a biocompatible photo-responsive multilayer system based on a Preyssler-type POM ([NaP5W30O110]14â ) and a natural origin polymer, chitosan, using the LbL methodology. The photo-reduction properties of the POM allow the spatially controlled disruption of the assembled layers due to the weakening of the electrostatic interactions between the layers. This system has found applicability in detaching devices, such as the cell sheet technology, which may solve the drawbacks actually found in other cell treatment proposals.

Biomimetic supramolecular designs for the controlled release of growth factors in bone regeneration

The extracellular matrix (ECM) of tissues is an assembly of insoluble macromolecules that specifically interact with soluble bioactive molecules and regulate their distribution and availability to cells. Recapitulating this ability has been an important target in controlled growth factor delivery strategies for tissue regeneration and requires the design of multifunctional carriers. This review describes the integration of supramolecular interactions on the design of delivery strategies that encompass self-assembling and engineered affinity components to construct advanced biomimetic carriers for growth factor delivery. Several glycan- and peptide-based self-assemblies reported in the literature are highlighted and commented upon. These examples demonstrate how molecular design and chemistry are successfully employed to create versatile multifunctional molecules which self-assemble/disassemble in a precisely predicted manner, thus controlling compartmentalization, transport and delivery. Finally, we discuss whether recent advances in the design and preparation of supramolecular delivery systems have been sufficient to drive real translation towards a clinical impact. 

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