Scielo RSS <![CDATA[Polibits]]> vol. num. 50 lang. pt <![CDATA[SciELO Logo]]> <![CDATA[<b>Editorial</b>]]> <![CDATA[<b>Fast Intra Mode Decision in High Efficiency Video Coding</b>]]> In this paper a coding unit early termination algorithm resulting in a fast intra prediction is proposed that terminates complete full search prediction for the coding unit. This is followed by a prediction unit mode decision to find the optimal modes HEVC encoder 35 prediction modes. This includes a two-step process: firstly calculating the Sum of Absolute Differences (SAD) of all the modes by down sampling method and secondly applying a three-step search algorithm to remove unnecessary modes. This is followed by early RDOQ (Rate Distortion Optimization Quantization) termination algorithm to further reduce the encoding time. Experimental results based on several video test sequences for 30 frames from each test sequence show for HEVC a decrease of about 35%-48% in encoding, with negligible degradation in peak signal to noise ratio (PSNR). Metrics such as BD-bitrate (Bjentegaard Delta bitrate), BD-PSNR (Bjentegaard Delta Peak Signal to Noise Ratio) and RD plots (Rate Distortion) are also used. <![CDATA[<b>A Dynamic Gesture Recognition System based on CIPBR Algorithm</b>]]> Dynamic gesture recognition has been studied actually for it big application in several areas such as virtual reality, games and sign language. But some problems have to be solved in computer applications, such as response time and classification rate, which directly affect the real-time usage. This paper proposes a novel algorithm called Convex Invariant Position Based on Ransac which improved the good results in dynamic gesture recognition problem. The proposed method is combined with a adapted PSO variation to reduce features and a HMM and three DTW variations as classifiers. <![CDATA[<b>Design of High Accuracy Tracking Systems with H<sub>∞</sub> Preview Control</b>]]> Positioning and tracking control systems are an important component of autonomous robot applications. This paper presents the design method of tracking control systems based on H∞ preview control where the present and future desired positions of the robot are used to determine the control actions to be applied so that the robot describes the desired trajectory as close as possible. The performance improvements achieved with H∞ preview control have been examined in the frequency and time domains for different types of reference signals when applied to a one-dimensional positioning system. It was found that preview control improves the tracking performance by improving the phase response of the tracking system. <![CDATA[<b>A More Efficient Representation of Obscuration for VRCC-3D+ Relations</b>]]> VRCC-3D+ is an implementation of a region connection calculus that qualitatively determines the spatial relation between two 3D objects in terms of connectivity and obscuration. The eight connectivity relations are conceptually the same as RCC8, but calculated in 3D rather than 2D. The fifteen obscuration relations are calculated using the projection of the 3D objects on a particular 2D plane and the distance of the objects from the viewpoint. Herein we present a smaller, more precise set of VRCC-3D+ obscuration relations that retains the qualities of being jointly exhaustive and pairwise disjoint. However, this new set of relations overcomes two problems that existed in the previous set of fifteen relations: (1) lack of a precise mathematical definition for a key predicate, InFront, and (2) lack of an intuitive mapping of converse relations. <![CDATA[<b>Computational Model of Dialog Based on Rules Applied to a Robotic Mobile Guide</b>]]> En este artículo se presenta la descripción formal detallada del módulo de control del diálogo para un robot móvil que funciona como guía (en un museo). El módulo incluye el modelo proposicional del diálogo, la especificación de los actos de habla y los bloques del habla, así como el inventario de los patrones de habla correspondientes a todos los actos de habla en el modelo. El modelo del diálogo se implementa como una red de estados y transiciones entre estados condicionados por reglas que estipulan los factores verbales y visuales. La arquitectura del módulo no depende de un idioma particular y puede ser adaptado a cualquier lenguaje natural.<hr/>This paper presents a formal detailed description of the dialogue management module for a mobile robot functioning as a guide. The module includes a propositional dialogue model, specification of speech acts and speech blocks as well as the inventory of speech patterns corresponding to all speech acts of the model. The dialogue model is implemented as a network of states and transitions between states conditioned by rules, which include verbal and visual factors. The architecture of the module is language independent and can be adapted to any natural language. <![CDATA[<b>NoXperanto</b>: <b>Crowdsourced Polyglot Persistence</b>]]> This paper proposes NoXPERANTO, a novel crowdsourcing approach to address querying over data collections managed by polyglot persistence settings. The main contribution of NoXPERANTO is the ability to solve complex queries involving different data stores by exploiting queries from expert users (i.e. a crowd of database administrators, data engineers, domain experts, etc.), assuming that these users can submit meaningful queries. NoXPERANTO exploits the results of "meaningful queries" in order to facilitate the forthcoming query answering processes. In particular, queries results are used to: (i) help non-expert users in using the multi-database environment and (ii) improve performances of the multi-database environment, which not only uses disk and memory resources, but heavily rely on network bandwidth. NoXPERANTO employs a layer to keep track of the information produced by the crowd modeled as a Property Graph and managed in a Graph Database Management System (GDBMS). <![CDATA[<b>Haar Wavelet Neural Network for Multi-step-ahead Anchovy Catches Forecasting</b>]]> This paper proposes a hybrid multi-step-ahead forecasting model based on two stages to improve pelagic fish-catch time-series modeling. In the first stage, the Fourier power spectrum is used to analyze variations within a time series at multiple periodicities, while the stationary wavelet transform is used to extract a high frequency (HF) component of annual periodicity and a low frequency (LF) component of inter-annual periodicity. In the second stage, both the HF and LF components are the inputs into a single-hidden neural network model to predict the original non-stationary time series. We demonstrate the utility of the proposed forecasting model on monthly anchovy catches time-series of the coastal zone of northern Chile (18°S-24°S) for periods from January 1963 to December 2008. Empirical results obtained for 7-month ahead forecasting showed the effectiveness of the proposed hybrid forecasting strategy. <![CDATA[<b>A Comparison between Two Metaheuristics Applied to the Cell Formation Problem with Alternative Routings</b>]]> This work proposes a genetic algorithm for optimization of the cell formation problem with alternative routings. A series of test problems were generated and used to evaluate the performance of the proposed Genetic Algorithm and a Simulated Annealing algorithm as well. The novelty of the proposed work lies in the representation technique and the transformations that allow treating the original multidimensional problem as a two-dimensional one. That simplified the programming tasks and the resolution method. <![CDATA[<b>Traffic control based on intelligent agents</b>]]> La tecnología de agentes se ha demostrado ser una ciencia computacional avanzada capaz de lograr mejoras sustanciales en un rango de aplicaciones debido a su paradigma de la estructura de toma de decisiones basado en el razonamiento cognitivo. En este sentido, el artículo presenta el desarrollo de una metodología novedosa que permite incluir un modelo formal basado en agentes autónomos e inteligentes capaces de manipular las fases de los ciclos en una infraestructura de semáforos de acuerdo a las exigencias y limitaciones de la carretera. Este proceso mejora efectiva e inmediata de la calidad del servicio en una intersección, aumentando el rendimiento de la movilidad de los vehículos y mejorando la generación de emisiones, cuando los vehículos se paran en un semáforo rojo. Para corroborar esto, el artículo presenta algunos experimentos con el fin de comparar la metodología propuesta contra una infraestructura pre-programada. Por último, se presentan las conclusiones a destacar la eficacia y la utilidad de la metodología desarrollada con la intención de alcanzar el control de tráfico adecuado de una ciudad en expansión.<hr/>Agent technology has been demonstrated to be an advance computational science capable to achieve substantial improvements in a cover range of applications because of its paradigm of decision-making structure based on cognitive reasoning. In this sense, the paper introduces the development of a novel methodology that allows including a formal model founded on autonomous and intelligent agents capable to manipulate the phases of the cycles in a traffic lights infrastructure according to the requirements and constraints of the road. This process improves effectively and immediately the quality of the service in an intersection, increasing the performance of the vehicular mobility and the generation of emissions, when vehicles are stopped in a red light. To corroborate this, the article presents some experiments in order to compare the proposed methodology against a preprogrammed infrastructure. Finally, conclusions are presented to emphasize the effectiveness and usefulness of the developed methodology whit the main intention of achieving an adequate traffic control of an expanding city. <![CDATA[<b>Acoustic Fingerprint Recognition Using Artificial Neural Networks</b>]]> This paper presents an implementation of Artificial Neural Networks (ANN) for acoustic fingerprints recognition, applied to the identification of marine vessels. In many cases, the vessel recognition process from an audible signal is performed by human operators, which could lead to failures in the identification process. Before entering the ANN classification process, the signal is filtered and its spectral characteristics are extracted. A comparison of the classification process between three types of neural networks is presented.