Automatic fault detection on BIPV systems without solar irradiation data

BIPV systems are small PV generation units spread out over the territory, and whose characteristics are very diverse. This makes difficult a cost-effective procedure for monitoring, fault detection, performance analyses, operation and maintenance. As a result, many problems affecting BIPV systems go undetected. In order to carry out effective automatic fault detection procedures, we need a performance indicator that is reliable and that can be applied on many PV systems at a very low cost. The existing approaches for analyzing the performance of PV systems are often based on the Performance Ratio (PR), whose accuracy depends on good solar irradiation data, which in turn can be very difficult to obtain or cost-prohibitive for the BIPV owner. We present an alternative fault detection procedure based on a performance indicator that can be constructed on the sole basis of the energy production data measured at the BIPV systems. This procedure does not require the input of operating conditions data, such as solar irradiation, air temperature, or wind speed. The performance indicator, called Performance to Peers (P2P), is constructed from spatial and temporal correlations between the energy output of neighboring and similar PV systems. This method was developed from the analysis of the energy production data of approximately 10,000 BIPV systems located in Europe. The results of our procedure are illustrated on the hourly, daily and monthly data monitored during one year at one BIPV system located in the South of Belgium. Our results confirm that it is possible to carry out automatic fault detection procedures without solar irradiation data. P2P proves to be more stable than PR most of the time, and thus constitutes a more reliable performance indicator for fault detection procedures. We also discuss the main limitations of this novel methodology, and we suggest several future lines of research that seem promising to improve on these procedures.

Identificación de áreas prioritarias para la conservación de razas de maíz en la sierra de Ecuador

En Ecuador el maíz es el cultivo más importante en superficie y es base de la alimentación para la población rural que vive en los Andes. A diferencia de lo que sucede en la Costa, en la región Sierra todavía se cultivan numerosas variedades tradicionales que se agrupan en veinticuatro razas. Mantener esta diversidad es, pues, de gran importancia no solo para la seguridad alimentaria, sino también como fuente de genes para tolerancia a factores abióticos que podrían ser incorporados a las variedades modernas. Si bien parte de esta diversidad fue recolectada a mediados del siglo pasado y está siendo conservada en distintos bancos de germoplasma, es deseable que su conservación in situ también esté asegurada, entre otras razones, porque de esta manera el cultivo puede seguir evolucionando. Para poder implementar un plan de conservación en finca que contribuya a preservar este patrimonio, resulta imprescindible identificar áreas idóneas donde concentrar los recursos y conocer las características y tipologías de los agricultores que manejan la diversidad actual. Generar esta información es el objetivo principal de esta investigación y para lograrlo se han llevado a cabo cuatro estudios: (1) Análisis de la diversidad a nivel de razas e identificación de áreas de alta riqueza de razas, alta diversidad morfológica y/o alta diversidad ecogeográfica en la Sierra de Ecuador, (2) Identificación del perfil y las características de los agricultores que conservan y manejan las variedades tradicionales de maíz en la Sierra de Ecuador, (3) Análisis del conocimiento local, manejo y usos de variedades tradicionales de maíz en la Sierra de Ecuador, y (4) Identificación de áreas de alta diversidad y bajo riesgo de pérdida para la conservación en finca de maíz en la Sierra de Ecuador. Para el primer estudio se visitaron 303 fincas distribuidas a lo largo de la Sierra y se recolectaron 636 muestras que fueron caracterizadas morfológicamente mediante 14 variables: 8 relacionadas con la mazorca (forma, longitud y diámetro de la mazorca, color y diámetro de olote y número y disposición de hileras) y 7 referidas el grano (número total de granos, color, forma, longitud, anchura y grosor de grano y tipo de endospermo). Adicionalmente, las fincas donde se tomaron las muestras fueron caracterizadas ecogeográficamente mediante 5 variables climáticas (temperatura media estacional, rango de temperatura media anual, temperatura mínima de diciembre, precipitación estacional y precipitación de octubre), 2 geofísicas (altitud y pendiente) y 5 edáficas (textura principal del suelo, profundidad a roca, pH, contenido en materia orgánica y fertilidad). A partir de esta información y mediante técnicas de sistemas de información geográfica (SIG), se generaron mapas de distribución por raza en formato vectorial y un mapa de riqueza de razas, un mapa de diversidad morfológica y un mapa de diversidad ecogeográfica en formato ráster con celdas de 10 km x 10 km. Los resultados permitieron constatar que, en los últimos 60 años, no se ha perdido ninguna raza. Sin embargo, Canguil, Chaucho y Clavito han dejado de cultivarse en algunas provincias con la consiguiente erosión genética del cultivo. La caracterización morfológica detectó diferencias en el grado de variabilidad intra-raza, siendo Patillo Ecuatoriano, Racimo de Uva y Uchima las razas más heterogéneas tanto para los caracteres cualitativos como cuantitativos. A nivel climático y geofísico, también se detectaron diferencias en el grado de variación intra-raza; Cuzco Ecuatoriano, Kcello Ecuatoriano y Montaña Ecuatoriana fueron las razas que en promedio presentaron mayores rangos y coeficientes de variación para estas variables ecogeográficas. En cuanto a las condiciones edáficas todas las razas, excepto Cónico Dentado, presentaron una gran heterogeneidad, pudiendo crecer tanto en suelos ricos como pobres, con valores de pH entre ácido y moderadamente alcalino. La comparación entre razas reveló diferencias significativas en los rangos ambientales de algunas razas como Cónico Dentado, que tiende a cultivarse a menor altitud y, por tanto, en ambientes menos fríos y de mayor precipitación que Blanco Blandito, Patillo Ecuatoriano, Sabanero Ecuatoriano, Uchima y Zhima. Para la mayoría de las razas se encontraron materiales potencialmente adaptados a condiciones de estrés (precipitación estacional inferior a 500 mm y suelos con pH entre 4.5 y 5.5). Finalmente, los mapas de riqueza, de diversidad morfológica y de diversidad ecogeográfica mostraron 36 celdas de alta diversidad repartidas en las 10 provincias de la Sierra: 11 celdas en las provincias del norte, 11 en las provincias del centro y 14 en las provincias del sur. Para la caracterización e identificación de las tipologías de los agricultores que cultivan maíz en la Sierra de Ecuador y el análisis de los posibles factores de riesgo de pérdida de diversidad, se realizaron entrevistas individuales y semiestructuradas a los agricultores dueños de las fincas donde se recolectaron las muestras para el estudio de diversidad (254 en total). Las preguntas que se formularon (11 abiertas y 5 cerradas) estuvieron organizadas en seis bloques: datos del agricultor, características de la finca, diversidad y conocimiento del cultivo, manejo del cultivo, usos y flujo de semillas. Los resultados indicaron que la diversidad de maíz que hay en la Sierra de Ecuador es manejada mayoritariamente por agricultores mestizos, de entre 30 y 55 años, que cultivan una o dos variedades tradicionales para autoconsumo, en parcelas de menos de 0.5 ha y en asocio con fréjol. El análisis de segmentación mediante el algoritmo Chi-square automatic interaction detection (CHAID) permitió identificar un pequeño grupo de agricultores indígenas con parcelas medianas (entre 0.5 ha y 1.5 ha) que conservan un mayor número de variedades tradicionales por finca que el agricultor promedio. Los análisis estadísticos no detectaron diferencias significativas entre etnias (mestizo vs. indígena), géneros (hombre vs. mujer) y grupos de edad (jóvenes menores de 30 años, adultos entre 30 y 55 años y adultos mayores de 55 años) en lo que respecta al conocimiento del cultivo (criterios de reconocimiento y razones de preferencia) y manejo (tipo de cultivo), pero sí detectaron diferencias entre regiones, principalmente en el modo de cultivar el maíz; mientras que en el norte y sur tienden a sembrarlo en asocio y con un mayor número de especies, en el centro acostumbran a cultivarlo preferentemente solo. En cuanto a los usos, se recopilaron hasta 39 modos diferentes de consumir maíz, siendo Kcello Ecuatoriano y Zhima las razas para las que se registró un mayor número de usos. La comparación del número medio de usos por variedad entre etnias evidenció que los agricultores mestizos utilizan sus variedades tradicionales de forma más variada que los indígenas. Entre los factores de riesgo que se analizaron, el bajo porcentaje de jóvenes agricultores que se ocupan de las fincas podría suponer una amenaza a medio plazo por falta de relevo generacional. Adicionalmente, las numerosas sinonimias y homonimias que se detectaron y el bajo intercambio de semillas también podrían ser causa de pérdida de diversidad, bien por reemplazo o por envejecimiento de la semilla. Finalmente, se concluyó que las razas Chaucho, Complejo Chillo-Huandango, Complejo Mishca-Huandango, Cónico Dentado, Montaña Ecuatoriana y Sabanero Ecuatoriano son particularmente vulnerables, no solo por su baja presencia, sino también por el color de grano que tienen (los mismos que la mayoría de las razas más comunes) y carecer de nombres y usos específicos. Finalmente, para la priorización de áreas de conservación en finca para maíz en la Sierra de Ecuador, se utilizaron 13 criterios de diferente naturaleza: 2 ecogeográficos (precipitación, diversidad ecogeográfica), 6 biológicos (grado de presencia del cultivo, riqueza de razas, diversidad morfológica, presencia de mezclas, presencia de razas locales y riesgo de erosión genética), 3 culturales (abundancia de variedades por finca, diversidad de usos y frecuencia de intercambio) y 2 demográficos (tamaño de la población y distancia a núcleos urbanos). Mediante técnicas SIG y de evaluación multicriterio, los valores originales de las capas-criterio fueron transformados a una escala de 0 a 100. Posteriormente, las capas-criterio normalizadas fueron sumadas utilizando tres métodos de ponderación: (1) mismo peso, (2) diferente peso según la puntuación otorgada por 72 expertos, y (3) diferente peso según el método de comparación entre pares de criterios. Los resultados permitieron identificar ocho celdas de 10 km x 10 km con alta puntuación (> 65): tres celdas en el norte (una en cada una de las provincias), una celda en el centro (en la provincia de Cotopaxi), y cuatro celdas en la región sur (dos en Azuay y otras dos en Loja). ABSTRACT In Ecuador, the maize is the most important cultivation in surface and it is a base of the feeding for the rural population who lives in the Andes. In contrast to what it happens on the Coast, in the Sierra region still there are cultivated numerous traditional varieties that are grouped into twenty-four races. Maintaining this diversity is, therefore, of great importance not only for food security, but also as a source of genes for tolerance to abiotic factors could be incorporated into modern varieties. Although part of this diversity was collected in the middle of the last century and is still preserved in various germplasm banks, it is desirable for the in situ conservation also is assured, among other reasons, because in this way the crop can continue to evolve. To be able to implement a conservation plan on farm that contribute to preserving this heritage, it is essential to identify suitable areas where to concentrate resources and know the characteristics and typology of farmer who managed the current diversity. To generate this information is the main target of this investigation and to achieve this, four studies have been carried out: (1) Analysis of the diversity at races and identification of areas of high richness of races, high morphological diversity and / or ecogeographical high diversity in the Sierra of Ecuador, (2) Identification of the profile and characteristics of farmers who conserve and manage traditional varieties of maize in the Sierra of Ecuador, (3) Analysis of local knowledge, management and use of traditional varieties of maize in the Sierra of Ecuador, and (4) Identification of areas of high diversity and low risk of loss for the conservation of maize in the Sierra of Ecuador. For the first study were visited 303 farms distributed along the Sierra and collected 636 samples that were characterized morphologically by 14 variables: 8 related to the ear (shape, length and diameter of the cob, colour, and diameter of cob and number and arrangement of rows) and 7 referred to the grain (total number of grain, colour, shape, length, width, and thickness and type of grain endosperm). In addition, the farms where the samples were taken were characterized ecogeographically through 5 climatic variables (seasonal average temperature, range of average annual temperature, minimum temperature for December, seasonal precipitation and precipitation of October), 2 geophysical (altitude and slope) and edaphic 5 (main texture of the soil, deep rock, pH, content of organic matter and fertility). From this information and techniques of geographic information systems (GIS), maps were generated for distribution by race in vector format and a map of richness of races, a map of morphological diversity and a map of ecogeographical diversity in raster format with cells of 10 km x 10 km. The results allowed observing that, over the past 60 years, it has not lost any race. Nevertheless, Canguil, Chaucho and Clavito have stopped being cultivated in some provinces with the consequent genetic erosion of the cultivation. The morphological characterization detected differences in the degree of variability intra-race, being Patillo Ecuatoriano, Racimo de Uva and Uchima races more heterogeneous both for the qualitative and quantitative characters. At climate and geophysical level, also detected differences in the degree of variation intra-race; Cuzco Ecuatoriano, Kcello Ecuatoriano and Montaña Ecuatoriana were races that, on average, showed higher ranges and coefficients of variation for these geographical characters. In terms of the edaphic conditions, all races, except Cónico Dentado, showed a great heterogeneity, and can grow both in rich and poor soils, with pH values between acid and moderately alkaline. The comparison between races revealed significant differences in the environmental ranges in some races as Cónico Dentado, which tends to be grown at lower elevations and, therefore, in environments less cold and greater precipitation than Blanco Blandito, Patillo Ecuatoriano, Sabanero Ecuatoriano, Uchima and Zhima. For most of the races were found materials potentially adapted to stress conditions (seasonal precipitation less than 500 mm and soil with a pH between 4.5 and 5.5). Finally, the maps of richness, morphologic diversity and ecogeographical diversity showed 36 cells high diversity distributed in 10 provinces of the Sierra: 11 cells in the northern provinces, 11 in the central provinces and 14 in the southern provinces. For the characterization and identification of the typology of the farmers who cultivate corn in the Sierra of Ecuador and the analysis of the possible factors of risk of loss of diversity, there were realized interviews individual and semistructured to the farmers’ owners of the farms where the samples were gathered for the study of diversity (254 in whole). The questions that were formulated (11 opened ones and 5 closed ones) were organized in six blocks: data of the farmer, characteristics of the farm, diversity and knowledge of the crop, crop management, uses and seed flow. The results indicated that the maize diversity that exist in the Sierra of Ecuador is managed mainly by mestizo farmers, aged between 30 and 55, who cultivate one or two traditional varieties for self-consumption, on plots of less than 0.5 has and in associated with beans. The segmentation analysis algorithm using the Chi-square automatic interaction detection (CHAID technique), allowed to identify a small group of indigenous farmers with medium-sized plots (between 0.5 there is and 1.5 it is) that a major number of traditional varieties preserves for farm that the average farmer. The statistical analysis did not detect significant differences between ethnic groups (mestizos vs. indigenous), genres (man vs. women) and age groups (young people under 30 years of age, adults between 30 and 55 years and adults over 55 years old) in regards to the knowledge of the cultivation (recognition criteria and reasons of preference) and management (type of crop), but if detected differences between regions, mainly on the mode of cultivating the maize; while in the north and south they tend to sow in associate and with a greater number of species, in the center accustomed to cultivate it preferably only. In regards to the uses, they were compiled up to 39 different ways of consuming maize, being Kcello Ecuatoriano and Zhima the races for which a major number of uses registered. The comparison of the average number of uses per variety between ethnic groups showed that the mestizo farmers used their traditional varieties of form more varied than the indigenous people. Between the factors of risk that were analyzed, the low percentage of young farmers who deal with the farms might suppose a medium-term threat for lack of generational relief. In addition, the numerous synonyms and homonyms that were detected and the low seed exchange could also be a cause of loss of diversity, either by replacement or by aging of the seed. Finally, it was concluded that the races Chaucho, Complex Chillo-Huandango, Complex Mishca-Huandango, Cónico Dentado, Montaña Ecuatoriana and Sabanero Ecuatoriano are particularly vulnerable, not only because of their low presence, but also by the grain color they have (the same as the majority of races more common) and lack of names and specific uses. Finally, for the prioritization of maize conservation areas on farm in the Sierra of Ecuador, used 13 criteria of different nature: 2 ecogeographic (precipitation, diversity ecogeographical), 6 biological (degree of presence of the crop, races richness, morphological diversity, the presence of mixtures, presence of local races and risk of genetic erosion), 3 cultural (abundance of varieties per farm, diversity of uses and frequency of exchange) and 2 demographic (population size and distance to urban centers). Using GIS techniques and multicriteria evaluation, the original values of the layers-criterion were transformed to a scale of 0 to 100. Later, the normalized layers - criteria were added using three weighting methods: (1) the same weight, (2) different weight according to the score given by 72 experts, and (3) different weight according to the method of comparison between pairs of criteria. The results allowed to identify eight 10 km cells x 10 km with high punctuation (> 65): three cells in the north (one in each of the provinces), a cell in the center (in the Cotopaxi province), and four cells in the south region (two in Azuay and other two in Loja).

REST service testing based on inferred XML schemas

The concept of service oriented architecture has been extensively explored in software engineering, due to the fact that it produces architectures made up of several interconnected modules, easy to reuse when building new systems. This approach to design would be impossible without interconnection mechanisms such as REST (Representationa State Transfer) services, which allow module communication while minimizing coupling. . However, this low coupling brings disadvantages, such as the lack of transparency, which makes it difficult to sistematically create tests without knowledge of the inner working of a system. In this article, we present an automatic error detection system for REST services, based on a statistical analysis over responses produced at multiple service invocations. Thus, a service can be systematically tested without knowing its full specification. The method can find errors in REST services which could not be identified by means of traditional testing methods, and provides limited testing coverage for services whose response format is unknown. It can be also useful as a complement to other testing mechanisms.

An Efficient Multiple Object Detection and Tracking Framework for Automatic Counting and Video Surveillance Applications

Automatic visual object counting and video surveillance have important applications for home and business environments, such as security and management of access points. However, in order to obtain a satisfactory performance these technologies need professional and expensive hardware, complex installations and setups, and the supervision of qualified workers. In this paper, an efficient visual detection and tracking framework is proposed for the tasks of object counting and surveillance, which meets the requirements of the consumer electronics: off-the-shelf equipment, easy installation and configuration, and unsupervised working conditions. This is accomplished by a novel Bayesian tracking model that can manage multimodal distributions without explicitly computing the association between tracked objects and detections. In addition, it is robust to erroneous, distorted and missing detections. The proposed algorithm is compared with a recent work, also focused on consumer electronics, proving its superior performance.

Automatic Bandwidth Estimation Strategy for High-Quality Non-Parametric Modeling based Moving Object Detection

Here, a novel and efficient moving object detection strategy by non-parametric modeling is presented. Whereas the foreground is modeled by combining color and spatial information, the background model is constructed exclusively with color information, thus resulting in a great reduction of the computational and memory requirements. The estimation of the background and foreground covariance matrices, allows us to obtain compact moving regions while the number of false detections is reduced. Additionally, the application of a tracking strategy provides a priori knowledge about the spatial position of the moving objects, which improves the performance of the Bayesian classifier

Improving automatic detection of obstructive sleep apnea through nonlinear analysis of sustained speech

We present a novel approach for the detection of severe obstructive sleep apnea (OSA) based on patients' voices introducing nonlinear measures to describe sustained speech dynamics. Nonlinear features were combined with state-of-the-art speech recognition systems using statistical modeling techniques (Gaussian mixture models, GMMs) over cepstral parameterization (MFCC) for both continuous and sustained speech. Tests were performed on a database including speech records from both severe OSA and control speakers. A 10 % relative reduction in classification error was obtained for sustained speech when combining MFCC-GMM and nonlinear features, and 33 % when fusing nonlinear features with both sustained and continuous MFCC-GMM. Accuracy reached 88.5 % allowing the system to be used in OSA early detection. Tests showed that nonlinear features and MFCCs are lightly correlated on sustained speech, but uncorrelated on continuous speech. Results also suggest the existence of nonlinear effects in OSA patients' voices, which should be found in continuous speech.

Intelligent automatic overtaking system using vision for vehicle detection

There is clear evidence that investment in intelligent transportation system technologies brings major social and economic benefits. Technological advances in the area of automatic systems in particular are becoming vital for the reduction of road deaths. We here describe our approach to automation of one the riskiest autonomous manœuvres involving vehicles – overtaking. The approach is based on a stereo vision system responsible for detecting any preceding vehicle and triggering the autonomous overtaking manœuvre. To this end, a fuzzy-logic based controller was developed to emulate how humans overtake. Its input is information from the vision system and from a positioning-based system consisting of a differential global positioning system (DGPS) and an inertial measurement unit (IMU). Its output is the generation of action on the vehicle’s actuators, i.e., the steering wheel and throttle and brake pedals. The system has been incorporated into a commercial Citroën car and tested on the private driving circuit at the facilities of our research center, CAR, with different preceding vehicles – a motorbike, car, and truck – with encouraging results.

Automatic detection of crop rows in maize fields with high weeds pressure

This paper proposes a new method, oriented to crop row detection in images from maize fields with high weed pressure. The vision system is designed to be installed onboard a mobile agricultural vehicle, i.e. submitted to gyros, vibrations and undesired movements. The images are captured under image perspective, being affected by the above undesired effects. The image processing consists of three main processes: image segmentation, double thresholding, based on the Otsu’s method, and crop row detection. Image segmentation is based on the application of a vegetation index, the double thresholding achieves the separation between weeds and crops and the crop row detection applies least squares linear regression for line adjustment. Crop and weed separation becomes effective and the crop row detection can be favorably compared against the classical approach based on the Hough transform. Both gain effectiveness and accuracy thanks to the double thresholding that makes the main finding of the paper.

Evaluación de la contribución y el impacto de las tecnologías del habla en la detección automática del Síndrome de la Apnea Obstructiva del Sueño = Contributions and impact assessment of speech technologies on the automatic detection of severe Obstructive Sleep Apnea syndrome

La presente Tesis analiza las posibilidades que ofrecen en la actualidad las tecnologías del habla para la detección de patologías clínicas asociadas a la vía aérea superior. El estudio del habla que tradicionalmente cubre tanto la producción como el proceso de transformación del mensaje y las señales involucradas, desde el emisor hasta alcanzar al receptor, ofrece una vía de estudio alternativa para estas patologías. El hecho de que la señal emitida no solo contiene este mensaje, sino también información acerca del locutor, ha motivado el desarrollo de sistemas orientados a la identificación y verificación de la identidad de los locutores. Estos trabajos han recibido recientemente un nuevo impulso, orientándose tanto hacia la caracterización de rasgos que son comunes a varios locutores, como a las diferencias existentes entre grabaciones de un mismo locutor. Los primeros resultan especialmente relevantes para esta Tesis dado que estos rasgos podrían evidenciar la presencia de características relacionadas con una cierta condición común a varios locutores, independiente de su identidad. Tal es el caso que se enfrenta en esta Tesis, donde los rasgos identificados se relacionarían con una de la patología particular y directamente vinculada con el sistema de físico de conformación del habla. El caso del Síndrome de Apneas Hipopneas durante el Sueno (SAHS) resulta paradigmático. Se trata de una patología con una elevada prevalencia mundo, que aumenta con la edad. Los pacientes de esta patología experimentan episodios de cese involuntario de la respiración durante el sueño, que se prolongan durante varios segundos y que se reproducen a lo largo de la noche impidiendo el correcto descanso. En el caso de la apnea obstructiva, estos episodios se deben a la imposibilidad de mantener un camino abierto a través de la vía aérea, de forma que el flujo de aire se ve interrumpido. En la actualidad, el diagnostico de estos pacientes se realiza a través de un estudio polisomnográfico, que se centra en el análisis de los episodios de apnea durante el sueño, requiriendo que el paciente permanezca en el hospital durante una noche. La complejidad y el elevado coste de estos procedimientos, unidos a las crecientes listas de espera, han evidenciado la necesidad de contar con técnicas rápidas de detección, que si bien podrían no obtener tasas tan elevadas, permitirían reorganizar las listas de espera en función del grado de severidad de la patología en cada paciente. Entre otros, los sistemas de diagnostico por imagen, así como la caracterización antropométrica de los pacientes, han evidenciado la existencia de patrones anatómicos que tendrían influencia directa sobre el habla. Los trabajos dedicados al estudio del SAHS en lo relativo a como esta afecta al habla han sido escasos y algunos de ellos incluso contradictorios. Sin embargo, desde finales de la década de 1980 se conoce la existencia de patrones específicos relativos a la articulación, la fonación y la resonancia. Sin embargo, su descripción resultaba difícilmente aprovechable a través de un sistema de reconocimiento automático, pero apuntaba la existencia de un nexo entre voz y SAHS. En los últimos anos las técnicas de procesado automático han permitido el desarrollo de sistemas automáticos que ya son capaces de identificar diferencias significativas en el habla de los pacientes del SAHS, y que los distinguen de los locutores sanos. Por contra, poco se conoce acerca de la conexión entre estos nuevos resultados, los sé que habían obtenido en el pasado y la patogénesis del SAHS. Esta Tesis continua la labor desarrollada en este ámbito considerando específicamente: el estudio de la forma en que el SAHS afecta el habla de los pacientes, la mejora en las tasas de clasificación automática y la combinación de la información obtenida con los predictores utilizados por los especialistas clínicos en sus evaluaciones preliminares. Las dos primeras tareas plantean problemas simbióticos, pero diferentes. Mientras el estudio de la conexión entre el SAHS y el habla requiere de modelos acotados que puedan ser interpretados con facilidad, los sistemas de reconocimiento se sirven de un elevado número de dimensiones para la caracterización y posterior identificación de patrones. Así, la primera tarea debe permitirnos avanzar en la segunda, al igual que la incorporación de los predictores utilizados por los especialistas clínicos. La Tesis aborda el estudio tanto del habla continua como del habla sostenida, con el fin de aprovechar las sinergias y diferencias existentes entre ambas. En el análisis del habla continua se tomo como punto de partida un esquema que ya fue evaluado con anterioridad, y sobre el cual se ha tratado la evaluación y optimización de la representación del habla, así como la caracterización de los patrones específicos asociados al SAHS. Ello ha evidenciado la conexión entre el SAHS y los elementos fundamentales de la señal de voz: los formantes. Los resultados obtenidos demuestran que el éxito de estos sistemas se debe, fundamentalmente, a la capacidad de estas representaciones para describir dichas componentes, obviando las dimensiones ruidosas o con poca capacidad discriminativa. El esquema resultante ofrece una tasa de error por debajo del 18%, sirviéndose de clasificadores notablemente menos complejos que los descritos en el estado del arte y de una única grabación de voz de corta duración. En relación a la conexión entre el SAHS y los patrones observados, fue necesario considerar las diferencias inter- e intra-grupo, centrándonos en la articulación característica del locutor, sustituyendo los complejos modelos de clasificación por el estudio de los promedios espectrales. El resultado apunta con claridad hacia ciertas regiones del eje de frecuencias, sugiriendo la existencia de un estrechamiento sistemático en la sección del tracto en la región de la orofaringe, ya prevista en la patogénesis de este síndrome. En cuanto al habla sostenida, se han reproducido los estudios realizados sobre el habla continua en grabaciones de la vocal /a/ sostenida. Los resultados son cualitativamente análogos a los anteriores, si bien en este caso las tasas de clasificación resultan ser más bajas. Con el objetivo de identificar el sentido de este resultado se reprodujo el estudio de los promedios espectrales y de la variabilidad inter e intra-grupo. Ambos estudios mostraron importantes diferencias con los anteriores que podrían explicar estos resultados. Sin embargo, el habla sostenida ofrece otras oportunidades al establecer un entorno controlado para el estudio de la fonación, que también había sido identificada como una fuente de información para la detección del SAHS. De su estudio se pudo observar que, en el conjunto de datos disponibles, no existen variaciones que pudieran asociarse fácilmente con la fonación. Únicamente aquellas dimensiones que describen la distribución de energía a lo largo del eje de frecuencia evidenciaron diferencias significativas, apuntando, una vez más, en la dirección de las resonancias espectrales. Analizados los resultados anteriores, la Tesis afronta la fusión de ambas fuentes de información en un único sistema de clasificación. Con ello es posible mejorar las tasas de clasificación, bajo la hipótesis de que la información presente en el habla continua y el habla sostenida es fundamentalmente distinta. Esta tarea se realizo a través de un sencillo esquema de fusión que obtuvo un 88.6% de aciertos en clasificación (tasa de error del 11.4%), lo que representa una mejora significativa respecto al estado del arte. Finalmente, la combinación de este clasificador con los predictores utilizados por los especialistas clínicos ofreció una tasa del 91.3% (tasa de error de 8.7%), que se encuentra dentro del margen ofrecido por esquemas más costosos e intrusivos, y que a diferencia del propuesto, no pueden ser utilizados en la evaluación previa de los pacientes. Con todo, la Tesis ofrece una visión clara sobre la relación entre el SAHS y el habla, evidenciando el grado de madurez alcanzado por la tecnología del habla en la caracterización y detección del SAHS, poniendo de manifiesto que su uso para la evaluación de los pacientes ya sería posible, y dejando la puerta abierta a futuras investigaciones que continúen el trabajo aquí iniciado. ABSTRACT This Thesis explores the potential of speech technologies for the detection of clinical disorders connected to the upper airway. The study of speech traditionally covers both the production process and post processing of the signals involved, from the speaker up to the listener, offering an alternative path to study these pathologies. The fact that utterances embed not just the encoded message but also information about the speaker, has motivated the development of automatic systems oriented to the identification and verificaton the speaker’s identity. These have recently been boosted and reoriented either towards the characterization of traits that are common to several speakers, or to the differences between records of the same speaker collected under different conditions. The first are particularly relevant to this Thesis as these patterns could reveal the presence of features that are related to a common condition shared among different speakers, regardless of their identity. Such is the case faced in this Thesis, where the traits identified would relate to a particular pathology, directly connected to the speech production system. The Obstructive Sleep Apnea syndrome (OSA) is a paradigmatic case for analysis. It is a disorder with high prevalence among adults and affecting a larger number of them as they grow older. Patients suffering from this disorder experience episodes of involuntary cessation of breath during sleep that may last a few seconds and reproduce throughout the night, preventing proper rest. In the case of obstructive apnea, these episodes are related to the collapse of the pharynx, which interrupts the air flow. Currently, OSA diagnosis is done through a polysomnographic study, which focuses on the analysis of apnea episodes during sleep, requiring the patient to stay at the hospital for the whole night. The complexity and high cost of the procedures involved, combined with the waiting lists, have evidenced the need for screening techniques, which perhaps would not achieve outstanding performance rates but would allow clinicians to reorganize these lists ranking patients according to the severity of their condition. Among others, imaging diagnosis and anthropometric characterization of patients have evidenced the existence of anatomical patterns related to OSA that have direct influence on speech. Contributions devoted to the study of how this disorder affects scpeech are scarce and somehow contradictory. However, since the late 1980s the existence of specific patterns related to articulation, phonation and resonance is known. By that time these descriptions were virtually useless when coming to the development of an automatic system, but pointed out the existence of a link between speech and OSA. In recent years automatic processing techniques have evolved and are now able to identify significant differences in the speech of OSAS patients when compared to records from healthy subjects. Nevertheless, little is known about the connection between these new results with those published in the past and the pathogenesis of the OSA syndrome. This Thesis is aimed to progress beyond the previous research done in this area by addressing: the study of how OSA affects patients’ speech, the enhancement of automatic OSA classification based on speech analysis, and its integration with the information embedded in the predictors generally used by clinicians in preliminary patients’ examination. The first two tasks, though may appear symbiotic at first, are quite different. While studying the connection between speech and OSA requires simple narrow models that can be easily interpreted, classification requires larger models including a large number dimensions for the characterization and posterior identification of the observed patterns. Anyhow, it is clear that any progress made in the first task should allow us to improve our performance on the second one, and that the incorporation of the predictors used by clinicians shall contribute in this same direction. The Thesis considers both continuous and sustained speech analysis, to exploit the synergies and differences between them. On continuous speech analysis, a conventional speech processing scheme, designed and evaluated before this Thesis, was taken as a baseline. Over this initial system several alternative representations of the speech information were proposed, optimized and tested to select those more suitable for the characterization of OSA-specific patterns. Evidences were found on the existence of a connection between OSA and the fundamental constituents of the speech: the formants. Experimental results proved that the success of the proposed solution is well explained by the ability of speech representations to describe these specific OSA-related components, ignoring the noisy ones as well those presenting low discrimination capabilities. The resulting scheme obtained a 18% error rate, on a classification scheme significantly less complex than those described in the literature and operating on a single speech record. Regarding the connection between OSA and the observed patterns, it was necessary to consider inter-and intra-group differences for this analysis, and to focus on the articulation, replacing the complex classification models by the long-term average spectra. Results clearly point to certain regions on the frequency axis, suggesting the existence of a systematic narrowing in the vocal tract section at the oropharynx. This was already described in the pathogenesis of this syndrome. Regarding sustained speech, similar experiments as those conducted on continuous speech were reproduced on sustained phonations of vowel / a /. Results were qualitatively similar to the previous ones, though in this case perfomance rates were found to be noticeably lower. Trying to derive further knowledge from this result, experiments on the long-term average spectra and intraand inter-group variability ratios were also reproduced on sustained speech records. Results on both experiments showed significant differences from the previous ones obtained from continuous speech which could explain the differences observed on peformance. However, sustained speech also provided the opportunity to study phonation within the controlled framework it provides. This was also identified in the literature as a source of information for the detection of OSA. In this study it was found that, for the available dataset, no sistematic differences related to phonation could be found between the two groups of speakers. Only those dimensions which relate energy distribution along the frequency axis provided significant differences, pointing once again towards the direction of resonant components. Once classification schemes on both continuous and sustained speech were developed, the Thesis addressed their combination into a single classification system. Under the assumption that the information in continuous and sustained speech is fundamentally different, it should be possible to successfully merge the two of them. This was tested through a simple fusion scheme which obtained a 88.6% correct classification (11.4% error rate), which represents a significant improvement over the state of the art. Finally, the combination of this classifier with the variables used by clinicians obtained a 91.3% accuracy (8.7% error rate). This is within the range of alternative, but costly and intrusive schemes, which unlike the one proposed can not be used in the preliminary assessment of patients’ condition. In the end, this Thesis has shed new light on the underlying connection between OSA and speech, and evidenced the degree of maturity reached by speech technology on OSA characterization and detection, leaving the door open for future research which shall continue in the multiple directions that have been pointed out and left as future work.

Automatic detection of PV systems failures from monitoring validated on 10,000 BIPV systems in Europe

The installers and owners show a growing interest in the follow-up of the performance of their photovoltaic (PV) systems. The owners are requesting reliable sources of information to ensure that their system is functioning properly, and the installers are actively looking for efficient ways of providing them the most useful possible information from the data available. Policy makers are becoming increasingly interested in the knowledge of the real performance of PV systems and the most frequent sources of problems that they suffer to be able to target the identified challenges properly. The scientific and industrial PV community is also requiring an access to massive operational data to pursue the technological improvements further.

Neuroanatomic-based detection algorithm for automatic labeling of brain structures in brain injury

The number and grade of injured neuroanatomic structures and the type of injury determine the degree of impairment after a brain injury event and the recovery options of the patient. However, the body of knowledge and clinical intervention guides are basically focused on functional disorder and they still do not take into account the location of injuries. The prognostic value of location information is not known in detail either. This paper proposes a feature-based detection algorithm, named Neuroanatomic-Based Detection Algorithm (NBDA), based on SURF (Speeded Up Robust Feature) to label anatomical brain structures on cortical and sub-cortical areas. Themain goal is to register injured neuroanatomic structures to generate a database containing patient?s structural impairment profile. This kind of information permits to establish a relation with functional disorders and the prognostic evolution during neurorehabilitation procedures.

GMM-based classifiers for the automatic detection of obstructive sleep apnea

The aim of automatic pathological voice detection systems is to serve as tools, to medical specialists, for a more objective, less invasive and improved diagnosis of diseases. In this respect, the gold standard for those system include the usage of a optimized representation of the spectral envelope, either based on cepstral coefficients from the mel-scaled Fourier spectral envelope (Mel-Frequency Cepstral Coefficients) or from an all-pole estimation (Linear Prediction Coding Cepstral Coefficients) forcharacterization, and Gaussian Mixture Models for posterior classification. However, the study of recently proposed GMM-based classifiers as well as Nuisance mitigation techniques, such as those employed in speaker recognition, has not been widely considered inpathology detection labours. The present work aims at testing whether or not the employment of such speaker recognition tools might contribute to improve system performance in pathology detection systems, specifically in the automatic detection of Obstructive Sleep Apnea. The testing procedure employs an Obstructive Sleep Apnea database, in conjunction with GMM-based classifiers looking for a better performance. The results show that an improved performance might be obtained by using such approach.

Automatic expert system based on images for accuracy crop row detection in maize fields

This paper proposes an automatic expert system for accuracy crop row detection in maize fields based on images acquired from a vision system. Different applications in maize, particularly those based on site specific treatments, require the identification of the crop rows. The vision system is designed with a defined geometry and installed onboard a mobile agricultural vehicle, i.e. submitted to vibrations, gyros or uncontrolled movements. Crop rows can be estimated by applying geometrical parameters under image perspective projection. Because of the above undesired effects, most often, the estimation results inaccurate as compared to the real crop rows. The proposed expert system exploits the human knowledge which is mapped into two modules based on image processing techniques. The first one is intended for separating green plants (crops and weeds) from the rest (soil, stones and others). The second one is based on the system geometry where the expected crop lines are mapped onto the image and then a correction is applied through the well-tested and robust Theil–Sen estimator in order to adjust them to the real ones. Its performance is favorably compared against the classical Pearson product–moment correlation coefficient.

Exploring differences between phonetic classes in Sleep Apnoea Syndrome Patients using automatic speech processing techniques

This work is part of an on-going collaborative project between the medical and signal processing communities to promote new research efforts on automatic OSA (Obstructive Apnea Syndrome) diagnosis. In this paper, we explore the differences noted in phonetic classes (interphoneme) across groups (control/apnoea) and analyze their utility for OSA detection

Detection and Tracking of Traffic Signs Using a Recursive Bayesian Decision Framework

In this paper we propose a new method for the automatic detection and tracking of road traffic signs using an on-board single camera. This method aims to increase the reliability of the detections such that it can boost the performance of any traffic sign recognition scheme. The proposed approach exploits a combination of different features, such as color, appearance, and tracking information. This information is introduced into a recursive Bayesian decision framework, in which prior probabilities are dynamically adapted to tracking results. This decision scheme obtains a number of candidate regions in the image, according to their HS (Hue-Saturation). Finally, a Kalman filter with an adaptive noise tuning provides the required time and spatial coherence to the estimates. Results have shown that the proposed method achieves high detection rates in challenging scenarios, including illumination changes, rapid motion and significant perspective distortion