Volumen 5 No. especial (2014) - MATCH'14: Congreso de Reconocimiento de Patrones, Control Inteligente y Comunicaciones

Permanent URI for this collectionhttps://dspace-test.ucuenca.edu.ec/handle/123456789/21328

Browse

Now showing 1 - 17 of 17

Adquisición y análisis de señales cerebrales utilizando el dispositivo MindWave
(Universidad de Cuenca, 2014) Torres, Felipe; Sánchez, Christian; Palacio Baus, Kenneth Samuel; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
Brain activity can be monitored through electroencephalography and used as a bioelectric indicator. This paper shows how a low cost device can be used in developing brain-computer interface based applications. The obtained results show that the MindWave device can be used for EEG signal acquisition, and that the Wavelet transform can be employed to improve the signal representation in order to apply artificial intelligence algorithms and pattern recognition techniques to classify brain responses.
Analyzing downloaded data from road side units
(Universidad de Cuenca, 2014) Astudillo S., Darwin; Barros G., Juan Gabriel; Chaput, Emmanuel; Beylot, André-Luc; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
In the context of vehicular networks a map updating application was analyzed, which enables vehicles to download relevant geographic data related to their position. Possible strategies for data sharing are the networks from infrastructure to a vehicle (I2V) and from vehicle to vehicle (V2V). In this study two chunking techniques, the Random Sort Strategy (RSS) and the Network Coding (NC), for the I2V network segment were compared. The study revealed that the distribution of different received chunks is independent of the file size when NC is used; when RSS is used the mean and standard deviation depend on the file size. The authors intend to integrate in a next step these results in the analysis of the V2V network segment.
Bioleta Mata: una planta que interactúa por medio de realidad aumentada
(Universidad de Cuenca, 2014) Chang, Carolina; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
This paper describes a real plant that features an intelligent vehicle. The plant navigates through indoor environments, responds to environmental stimuli, interacts with humans through augmented reality, detects fire, and asks for help via Twitter. Experiments show that there are no false positives in the detection of fire, and that the detection of fire is greater than 50%, for distances along the sight line between the flames and the sensor of less than 5 m. Communication by XBee radios in indoor environments is effective up to 25 m at least.
Caracterización de señales sísmicas del Volcán Cotopaxi utilizando estimadores espectrales clásicos y de máxima entropía
(Universidad de Cuenca, 2014) Jaramillo A., Carolina E.; León V., Rubén D.; Lara C., Román A.; Benítez M., Diego S.; Ruiz, Mario; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
A study of the detection and characterization of volcano-tectonic seismic type of events, using a long period of records of the Cotopaxi volcano, is presented. The sequential detection structure, proposed as part of this work, allows maximizing the likelihood of the presence of an event and minimizes the absence of this from a seismic record. The detection is performed in the near real time domain, maintaining a constant false alarm rate, enabling in a later stage a more detailed study of the spectral content of the events using classical spectral estimators such as the periodogram and the parametric maximum entropy method by Burg. The approach enabled to categorize phenomenon detected as volcano-tectonic from a long record of observations which even include activities of non-tectonic origin, such as the other two types of lightning.
Caracterización espectroscópica de impedancia de polímeros para el uso en electrodos de electrocardiogramas
(Universidad de Cuenca, 2014) Véliz, Bremnen; Huilcapi, Víctor; Cajo, Ricardo; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
A comparative impedance study of a polymer type Thermoplastic Polyolefin Elastomer with a commercial universal gel electrode has been made in order to know whether the polymer can be used as dry electrode for the acquisition of cardiac signals. The results of the impedance spectroscopy showed very similar characteristics and that both devices behave as resistors. For instance, at 120 Hz the polymer measured a resistance of 6,3 k and the commercial gel electrode 7,2 k . Further, it was getting a linear relation of current versus voltage, determining resistances of 7 k and 8,2 k for the polymer and commercial gelled electrode respectively. Finally we have concluded that the polymer is applicable to construct dry electrodes.
Clinical validation of footprint analysis using the low cost Photo-podoscope JHECA NAWE
(Universidad de Cuenca, 2014) La Cruz, Alexandra; Zeas P., Ana L.; Quishpi M., Estefanía A.; Campoverde V., Johanna L.; Guamán B., Heydi K.; Morocho, Villie; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
A photo-podoscope is a useful tool for assessing a footprint; it is performed by observing the supporting areas of the foot in standing position. This evaluation and diagnosis generally is made by a specialist, and without leaving medical evidence for future comparisons. Generally, this evaluation is made based on the form of the footprint by using a traditional podoscope or rudimentary process which carries several disadvantages. We present, a medical alternative equipment called JHECA NAWE photo-podoscope. This equipment allows assessment of a footprint in a fast an effective way with accurate diagnoses, including a record for future evaluations. Commercial equipment is expensive, reason why medical specialists prefer to use rudimentary processes. In order to evaluate clinically the effectiveness of JHECA NAWE, we analyzed a cross-sectional of attended patients by using this equipment in comparison with different medical specialist opinion.
Control HSMC para regulación de glucosa sanguínea
(Universidad de Cuenca, 2014) Darío Rojas, Rubén; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
This article proposes a new strategy of control based on continue measurements of glucose and a habituating sliding mode controller (HSMC). The HSMC is developed combining the sliding mode control law and the principles of habituating control. The HSMC applied to the regulation of blood glucose in the intensive care unit, includes both intravascular glucose and insulin infusion inputs to provide the nutrition supply and to improve the rejection to disturbances. The simulation based study (in silico), using a physiological model of the dynamic glucose-insulin, shows that proposed control strategy works properly. Finally, the performance of the proposed controller and a standard PID controller are compared.
Detección temprana del cáncer de mama mediante la termografía en Ecuador
(Universidad de Cuenca, 2014) Pérez, María G.; Conci, Aura; Aguilar, Aida; Sánchez, Angel; Andaluz, Víctor H.; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
Breast cancer is one of the most common cancer types among women in Quito, Ecuador. Although, awareness campaigns and researches have led to improved diagnosis and treatment, in less developed countries it still is an unresolved problem due to the diagnosis only in advanced stages and the lack of resources enabling early detection. In Ecuador, for example, about four out of ten women diagnosed with breast cancer are in advanced stages (III and IV), whose treatment is costly and complex. The World Health Organization (WHO) states that screening programs are the more efficient way to combat this disease. Therefore it is fundamental to address new researches on early detection that are cost-effective and present advantages over the current method (based on the self-examination and mammography). The identification of such disease in early stage increases the prognosis and the survival rate. This article proposes a technique to incorporate low-cost, non-invasive early diagnosis based on the use of thermal.
Implementación de un detector de coral utilizando filtros Gabor Wavelets y máquinas de aprendizaje
(Universidad de Cuenca, 2014) Tusa, Eduardo; Villegas, Hyxia; Reynolds, Alan; Lane, David M.; Robertson, Neil M.; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
This work focuses on the implementation of a fast coral reef detector that is used for an Autonomous Underwater Vehicle (AUV, its acronym in English). A fast detection of the presence of coral ensures the AUV stabilization in front of coral reef in the shortest possible time, avoiding collisions with coral. The coral detection is carried out on an image that captures the scene that the AUV’s camera perceives. A pixel-by-pixel classification is performed between two classes: coral reef and the background that is non-coral reef. Each pixel of the image is assigned to a feature vector, which is generated by using Gabor Wavelet filters. These are implemented in C++ and the OpenCV library. The feature vectors are classified using nine machine learning algorithms. The performance of each algorithm is compared with the accuracy and execution time. The Decision Tree algorithm proved to be the fastest and most accurate of all the algorithms. We created a database of 621 images of coral reefs in Belize (110 of training images and 511 of testing images).
Integración de Arbotix, Raspberry Pi y motores Dynamixel Ax-12+ para un robot humanoide que busca y patea pelotas
(Universidad de Cuenca, 2014) Dos Reis, Jennifer; León, Juliana; Chang, Carolina; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
This article presents DeBuPa (Detección Búsqueda Pateo), a small humanoid (38 cm tall) built with the Bioloid Premium Kit. The CM-510 card has been excluded to be replaced by the Arbotix controller card, which is used to control the 16 motors Dynamixel AX-12 + (for moving the robot) and 2 analog servo (to move the camera). The camera has been connected to a Raspberry Pi mini- computer, as to enable the robot to autonomous detect and track the ball movements. All those actions must be properly coordinated so that the robot accomplishes the tasks of detection, tracking and kicking of the ball, requiring fluent communication between the Arbotix and Raspberry Pi. The tool used for this is the ROS (Robot Operating System) framework. The C++ language is used in the Raspberry Pi, a program capturing, filtering and processing the camera images, to find the ball, decides the action to take, and requests the Arbotix to execute the movements of the motors. To capture the camera images. The RasPiCam CV library is used to capture the camera images and the OpenCV libraries for the filtering and processing of the images. In addition, to steering the engines, the Arbotix takes care of the balance of the robot using as input signals of the Robotis Gyro sensor. If Arbotix detects an imbalance and even if the robot has fallen, the mechanism tries to help the robot to stand up.
Lógica borrosa para la estimación de estados críticos de una pila de combustible PEM
(Universidad de Cuenca, 2014) Agila, Wilton E.; Huilcapi, Victor M.; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
The real time determination of the critical states of operation of the fuel cell proton exchange membrane (acronym in English, PEM) is one of the main challenges for the control systems of PEM fuel cells. In this paper, the development and implementation of a non-invasive low cost method based on fuzzy decision techniques to estimate the critical states of operation of the PEM fuel cell is presented. The estimation is performed by perturbations of the state of operation of PEM fuel cell and the subsequent analysis of the temporal evolution of the voltage generated by the cell. The implementation of this stimulation-perceived technique of the state of fuel cell for the detection of critical states is a novelty and a step towards autonomous control in optimal operation of PEM fuel cells.
Mecanismo de posicionamiento en Azimut y Elevación para la optimización de radioenlaces punto a punto basado en algoritmos de búsqueda espectral y técnicas de muestreo
(Universidad de Cuenca, 2014) Cajo, Ricardo A.; Zuñiga, Steve R.; Huilcapi, Víctor M.; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
In this paper an algorithm of spectral search in Azimuth and Elevation to obtain an optimal reception of RF signals in point-to-point radio links is described. An experimental mechanism that allows acquiring the radiation intensity in Azimuth and Elevation for various angular positions based on field sampling and measuring techniques, was built. These techniques provide a fast and simple method for acquiring data; software was developed to execute this measuring and spectral search procedure, which allows an efficient determination and space positioning in the located optimal point. An example of a radio link between two buildings with patch antennas aligned with the proposed mechanism was used, and the performance evaluated through the energy budget with theoretical well known models of propagation.
Nuevo método de identificación usando la respuesta escalón para sistemas de fase no mínima de segundo orden
(Universidad de Cuenca, 2014) Rojas, Rubén D.; Garcia-Gabin, Winston; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
This paper presents a novel identification method for non-minimum phase systems based on the step response. The proposed approach provides a linear second-order model approximation avoiding complex experimental design. The proposed method is a closed identification algorithm based on characteristic points of the second order non minimum phase system step response. It is validated using different linear models. They have inverse response between 3,5% and 38% of the final steady state response. In simulations, it has been shown that satisfactory results can be obtained using the proposed identification procedure, where the identified parameters present mean relative errors lower than those obtained by Balaguer’s method.
Simulación de un enlace óptico WDM (Wavelength Division Multiplexing) utilizando POFs (Polymer Optical Fiber) en el espectro de luz visible
(Universidad de Cuenca, 2014) Ocampo Dávila, Andrés; Jiménez J., María Soledad; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
In this project, the characteristics and the operation mode for polymer optical fibers were studied to use them over an optical link to transmit information over short distances. The project also analyzed the active and passive components needed to achieve this goal, as active components the LED sources were studied and as passive components the multiplexers, filters and splitters were considered. On the other hand all of the nonlinear effects caused by the transmission of WDM (Wavelength Division Multiplexing) signals were examined. Finally, using MATLAB, a simulation of an optical link in the frequency domain was made using LED sources emitting light in the visible spectrum, together with WDM multiplexers, absorption filters, splitters and POF (Plastic Optical Fiber).
Sistema de detección de incendios forestales mediante redes sensoriales inalámbricas (Zigbee)
(Universidad de Cuenca, 2014) Erazo P., Jennyfer K.; Hervas P., Carlos A.; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca
This document describes the use of Zigbee via Arduino technology. These technologies have enabled the development of a system called Natura Sys. The system consisting of sensor motes is able of receiving environmental information such as temperature, humidity and presence of smoke. All information is centralized by a central base connected wireless with the sensors. The information is visualized by a desktop application providing a warning if in a given area of the forest there is a fire. The desktop application contains also a website in which statistical reports of the monitored parameters in the forest area are visualized.
Sistema supervisor inteligente para procesos de producción de petróleo
(Universidad de Cuenca, 2014) Ramírez, Miguel; Colina, Eliezer; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
The production maximization of heavy and extra heavy oil wells is the main benefit to be obtained from the control systems that are currently operating on oil companies. Given the complex and changing nature over time of existing artificial lift methods for oil extraction, meeting the specifications set for the processing of crude using conventional regulatory control is a difficult task. Taking this into account, this paper presents an intelligent supervisory system that detects changes in the operating conditions of the production process and makes automatic adjustments in its set points. Furthermore, the proposed supervisory system has the ability to detect faults in the sensors involved in the control loops, thus ensuring reliable operation of the process. The proposed supervisory system was tested in a real oil well, yielding results that exceeded expectations.
Tratamiento químico y biológico de efluentes mineros cianurados a escala laboratorio
(Universidad de Cuenca, 2014) Jumbo Pacheco, Pablo Xavier; Nieto Monteros, Diego Alejandro; Universidad de Cuenca; Dirección de Investigación de la Universidad de Cuenca; DIUC
Cyanide is a chemical compound used per excellence for gold leaching in the mining industry. However, it is highly toxic for the environment and its organisms. This work focused on determining the optimum conditions for the degradation of CN- using chemical compounds and a microbial consortium. Chemical and biological assays were performed separately using samples of mining effluents at different cyanide concentrations (280 and 10 mg/l CN-). For chemical degradation, three different oxidizing compounds were used: sodium hypochlorite, Caro’s acid and hydrogen peroxide at different concentrations, pH (10-11), and degradation times (4,71 and 20,75 h). For the biodegradation assays, a microbial consortium was cultivated in flasks containing the cyanide effluent, liquid medium, pH (11), agitation (200 rpm) and temperature (20±5°C). Free cyanide concentration, pH and the biomass concentration were measured. Chemical treatment results showed that the best oxidizing compound was hydrogen peroxide (8:1 g H2O2/g CN-) and pH (10), thus obtaining a 92,7% removal of free cyanide in 45 minutes (280 mg/l CN-) and 91,0% removal in 25 minutes (10 mg/l CN-). While in biodegradation the removal was 73,7% (280 mg/l CN-) in 384 h and 78,6% (10 mg/l CN-) in 240 h.