Hybrid incremental modeling based on least squares and fuzzy K-NN for monitoring tool wear in turning processes

There is now an emerging need for an efficient modeling strategy to develop a new generation of monitoring systems. One method of approaching the modeling of complex processes is to obtain a global model. It should be able to capture the basic or general behavior of the system, by means of a linear or quadratic regression, and then superimpose a local model on it that can capture the localized nonlinearities of the system. In this paper, a novel method based on a hybrid incremental modeling approach is designed and applied for tool wear detection in turning processes. It involves a two-step iterative process that combines a global model with a local model to take advantage of their underlying, complementary capacities. Thus, the first step constructs a global model using a least squares regression. A local model using the fuzzy k-nearest-neighbors smoothing algorithm is obtained in the second step. A comparative study then demonstrates that the hybrid incremental model provides better error-based performance indices for detecting tool wear than a transductive neurofuzzy model and an inductive neurofuzzy model.

Hybrid Approach Combining Machine Learning and a Rule-Based Expert System for Text Categorization

This paper discusses a novel hybrid approach for text categorization that combines a machine learning algorithm, which provides a base model trained with a labeled corpus, with a rule-based expert system, which is used to improve the results provided by the previous classifier, by filtering false positives and dealing with false negatives. The main advantage is that the system can be easily fine-tuned by adding specific rules for those noisy or conflicting categories that have not been successfully trained. We also describe an implementation based on k-Nearest Neighbor and a simple rule language to express lists of positive, negative and relevant (multiword) terms appearing in the input text. The system is evaluated in several scenarios, including the popular Reuters-21578 news corpus for comparison to other approaches, and categorization using IPTC metadata, EUROVOC thesaurus and others. Results show that this approach achieves a precision that is comparable to top ranked methods, with the added value that it does not require a demanding human expert workload to train

Towards hand biometrics in mobile devices

The increasing demand of security oriented to mobile applications has raised the attention to biometrics, as a proper and suitable solution for providing secure environment to mobile devices. With this aim, this document presents a biometric system based on hand geometry oriented to mobile devices, involving a high degree of freedom in terms of illumination, hand rotation and distance to camera. The user takes a picture of their own hand in the free space, without requiring any flat surface to locate the hand, and without removals of rings, bracelets or watches. The proposed biometric system relies on an accurate segmentation procedure, able to isolate hands from any background; a feature extraction, invariant to orientation, illumination, distance to camera and background; and a user classification, based on k-Nearest Neighbor approach, able to provide an accurate results on individual identification. The proposed method has been evaluated with two own databases collected with a HTC mobile. First database contains 120 individuals, with 20 acquisitions of both hands. Second database is a synthetic database, containing 408000 images of hand samples in different backgrounds: tiles, grass, water, sand, soil and the like. The system is able to identify individuals properly with False Reject Rate of 5.78% and False Acceptance Rate of 0.089%, using 60 features (15 features per finger)

Markov blanket-based approach for learning multi-dimensional Bayesian network classifiers: An application to predict the European Quality of Life-5 Dimensions (EQ-5D) from the 39-item Parkinson's Disease Questionnaire (PDQ-39)

Multi-dimensional Bayesian network classifiers (MBCs) are probabilistic graphical models recently proposed to deal with multi-dimensional classification problems, where each instance in the data set has to be assigned to more than one class variable. In this paper, we propose a Markov blanket-based approach for learning MBCs from data. Basically, it consists of determining the Markov blanket around each class variable using the HITON algorithm, then specifying the directionality over the MBC subgraphs. Our approach is applied to the prediction problem of the European Quality of Life-5 Dimensions (EQ-5D) from the 39-item Parkinson’s Disease Questionnaire (PDQ-39) in order to estimate the health-related quality of life of Parkinson’s patients. Fivefold cross-validation experiments were carried out on randomly generated synthetic data sets, Yeast data set, as well as on a real-world Parkinson’s disease data set containing 488 patients. The experimental study, including comparison with additional Bayesian network-based approaches, back propagation for multi-label learning, multi-label k-nearest neighbor, multinomial logistic regression, ordinary least squares, and censored least absolute deviations, shows encouraging results in terms of predictive accuracy as well as the identification of dependence relationships among class and feature variables.

Codificador de voz MIDI

El presente proyecto tiene el objetivo de facilitar la composición de canciones mediante la creación de las distintas pistas MIDI que la forman. Se implementan dos controladores. El primero, con objeto de transcribir la parte melódica, convierte la voz cantada o tarareada a eventos MIDI. Para ello, y tras el estudio de las distintas técnicas del cálculo del tono (pitch), se implementará una técnica con ciertas variaciones basada en la autocorrelación. También se profundiza en el segmentado de eventos, en particular, una técnica basada en el análisis de la derivada de la envolvente. El segundo, dedicado a la base rítmica de la canción, permite la creación de la percusión mediante el golpe rítmico de objetos que disponga el usuario, que serán asignados a los distintos elementos de percusión elegidos. Los resultados de la grabación de estos impactos serán señales de corta duración, no lineales y no armónicas, dificultando su discriminación. La herramienta elegida para la clasificación de los distintos patrones serán las redes neuronales artificiales (RNA). Se realizara un estudio de la metodología de diseño de redes neuronales especifico para este tipo de señales, evaluando la importancia de las variables de diseño como son el número de capas ocultas y neuronas en cada una de ellas, algoritmo de entrenamiento y funciones de activación. El estudio concluirá con la implementación de dos redes de diferente naturaleza. Una red de Elman, cuyas propiedades de memoria permiten la clasificación de patrones temporales, procesará las cualidades temporales analizando el ataque de su forma de onda. Una red de propagación hacia adelante feed-forward, que necesitará de robustas características espectrales y temporales para su clasificación. Se proponen 26 descriptores como los derivados de los momentos del espectro: centroide, curtosis y simetría, los coeficientes cepstrales de la escala de Mel (MFCCs), y algunos temporales como son la tasa de cruces por cero y el centroide de la envolvente temporal. Las capacidades de discriminación inter e intra clase de estas características serán evaluadas mediante un algoritmo de selección, habiéndose elegido RELIEF, un método basado en el algoritmo de los k vecinos mas próximos (KNN). Ambos controladores tendrán función de trabajar en tiempo real y offline, permitiendo tanto la composición de canciones, como su utilización como un instrumento más junto con mas músicos. ABSTRACT. The aim of this project is to make song composition easier by creating each MIDI track that builds it. Two controllers are implemented. In order to transcribe the melody, the first controler converts singing voice or humming into MIDI files. To do this a technique based on autocorrelation is implemented after having studied different pitch detection methods. Event segmentation has also been dealt with, to be more precise a technique based on the analysis of the signal's envelope and it's derivative have been used. The second one, can be used to make the song's rhythm . It allows the user, to create percussive patterns by hitting different objects of his environment. These recordings results in short duration, non-linear and non-harmonic signals. Which makes the classification process more complicated in the traditional way. The tools to used are the artificial neural networks (ANN). We will study the neural network design to deal with this kind of signals. The goal is to get a design methodology, paying attention to the variables involved, as the number of hidden layers and neurons in each, transfer functions and training algorithm. The study will end implementing two neural networks with different nature. Elman network, which has memory properties, is capable to recognize sequences of data and analyse the impact's waveform, precisely, the attack portion. A feed-forward network, needs strong spectral and temporal features extracted from the hit. Some descriptors are proposed as the derivates from the spectrum moment as centroid, kurtosis and skewness, the Mel-frequency cepstral coefficients, and some temporal features as the zero crossing rate (zcr) and the temporal envelope's centroid. Intra and inter class discrimination abilities of those descriptors will be weighted using the selection algorithm RELIEF, a Knn (K-nearest neighbor) based algorithm. Both MIDI controllers can be used to compose, or play with other musicians as it works on real-time and offline.

Estudio de los factores que condicionan, a escala local, la presencia de Epidendrum rhopalostele (Orchidaceae): Implicaciones para su conservación

El análisis de los factores que determinan el establecimiento y supervivencia de orquídeas epífitas, incluyen: a) las condiciones microambientales de los bosques que las mantienen, b) preferencias por las características de los hospederos donde crecen, c) limitación en la dispersión de semillas, d) interacciones planta-planta, y e) asociaciones micorrízicas para la germinación y resultan esenciales para el desarrollo de estrategias para la conservación y manejo de este grupo de plantas. Este trabajo ha evaluado la importancia de estos factores en Epidendrum rhopalostele, orquídea epífita del bosque de niebla montano, a través de los análisis de los patrones espaciales de los árboles que la portan y de la propia orquídea, a escala de población, estudios de asociación y métodos moleculares. Estos últimos han consistido en el uso de marcadores AFLP para el análisis de la estructura genética de la orquídea y en la secuenciación-clonación de la región ITS para la identificación de los hongos micorrízicos asociados. El objetivo de esta tesis es, por tanto, una mejor comprensión de los factores que condicionan la presencia de orquídeas epífitas en los remanentes de bosque de niebla montano y una evaluación de las implicaciones para la conservación y mantenimiento de sus hábitats y la permanencia de sus poblaciones. El estudio fue realizado en un fragmento de bosque de niebla montano de sucesión secundaria situado al este de la Cordillera Real, en los Andes del sur de Ecuador, a 2250 m.s.n.m y caracterizado por una pendiente marcada, temperatura media anual de 20.8°C y precipitación anual de 2193 mm. En este fragmento se mapearon, identificaron y caracterizaron todos los árboles presentes con DBH > 1 cm y todos los individuos de Epidendrum rhopalostele. Así mismo se tomaron muestras de hoja para obtener ADN de todas las orquídeas registradas y muestras de raíces de individuos con flor de E. rhopalostele, uno por cada forófito, para el análisis filogenético de micorrizas. Análisis espaciales de patrones de puntos basados en la K de Ripley y la distancia al vecino más cercano fueron usados para los árboles, forófitos y la población de E. rhopalostele. Se observó que la distribución espacial de árboles y forófitos de E. rhopalostele no es aleatoria, ya que se ajusta a un proceso agregado de Poisson. De ahí se infiere una limitación en la dispersión de las semillas en el fragmento estudiado y en el establecimiento de la orquídea. El patrón de distribución de la población de E. rhopalostele en el fragmento muestra un agrupamiento a pequeña escala sugiriendo una preferencia por micro-sitios para el establecimiento de la orquídea con un kernel de dispersión de las semillas estimado de 0.4 m. Las características preferentes del micro-sitio como tipos de árboles (Clusia alata y árboles muertos), tolerancia a la sombra, corteza rugosa, distribución en los dos primeros metros sugieren una tendencia a distribuirse en el sotobosque. La existencia de una segregación espacial entre adultos y juveniles sugiere una competencia por recursos limitados condicionada por la preferencia de micro-sitio. La estructura genética de la población de E. rhopalostele analizada a través de Structure y PCoA evidencia la presencia de dos grupos genéticos coexistiendo en el fragmento y en los mismos forófitos, posiblemente por eventos de hibridización entre especies de Epidendrum simpátricas. Los resultados del análisis de autocorrelación espacial efectuados en GenAlex confirman una estructura genético-espacial a pequeña escala que es compatible con un mecanismo de dispersión de semillas a corta distancia ocasionada por gravedad o pequeñas escorrentías, frente a la dispersión a larga distancia promovida por el viento generalmente atribuida a las orquídeas. Para la identificación de los micobiontes se amplificó la región ITS1-5.8S-ITS2, y 47 secuencias fueron usadas para el análisis filogenético basado en neighborjoining, análisis bayesiano y máximum-likelihood que determinó que Epidendrum rhopalostele establece asociaciones micorrízicas con al menos dos especies diferentes de Tulasnella. Se registraron plantas que estaban asociadas con los dos clados de hongos encontrados, sugiriendo ausencia de limitación en la distribución del hongo. Con relación a las implicaciones para la conservación in situ resultado de este trabajo se recomienda la preservación de todo el fragmento de bosque así como de las interacciones existentes (polinizadores, micorrizas) a fin de conservar la diversidad genética de esta orquídea epífita. Si fuere necesaria una reintroducción se deben contemplar distancias entre los individuos en cada forófito dentro de un rango de 0.4 m. Para promover el reclutamiento y regeneración de E. rhopalostele, se recomienda que los forófitos correspondan preferentemente a árboles muertos o caídos y a especies, como Clusia alata, que posean además corteza rugosa, sean tolerantes a la sombra, y en el área del sotobosque con menor luminosidad. Además es conveniente que las orquídeas en su distribución vertical estén ubicadas en los primeros metros. En conclusión, la limitación en la dispersión, las características del micro-sitio, las interacciones intraespecíficas y con especies congenéricas simpátricas y las preferencias micorrízicas condicionan la presencia de esta orquídea epífita en este tipo de bosque. ABSTRACT The analysis of factors that determine the establishment and survival of epiphytic depends on factors such as a) microenvironmental conditions of forest, b) preference for host characteristics where orchids grow, c) seed dispersal limitation, d) plant-plant interaction, e) priority mycorrhizal associations for germination, are essential for the development of strategies for management and conservation. This work evaluated the importance of these factors in Epidendrum rhopalostele, an epiphytic orchid of montane cloud forest through the analysis of spatial patterns of host trees and the orchid, in a more specific scale, with association studies and molecular methods, including AFLPs for orchid population genetic structure and the sequencing of the ITS region for associated mycorrhizal fungi. The aim of this thesis is to understand the factors that condition the presence of epiphytic orchids in the remnants of montane cloud forest and to assess the implications for the conservation and preservation of their habitats and the persistence of the orchid populations. The study was carried out in a fragment of montane cloud forest of secondary succession on the eastern slope of Cordillera Real in the Andes of southern Ecuador, located at 2250 m a.s.l. characterized by a steep slope, mean annual temperature of 20.8°C and annual precipitation of 2193 mm. All trees with DBH > 1 cm were mapped, characterized and identified. All E. rhopalostele individuals present were counted, marked, characterized and mapped. Leaf samples of all orchid individuals were collected for DNA analysis. Root samples of flowering E. rhopalostele individuals were collected for phylogenetic analysis of mycorrhizae, one per phorophyte. Spatial point pattern analysis based on Ripley`s K function and nearest neighbor function was used for trees, phorophytes and orchid population. We observed that spatial distribution of trees and phorophytes is not random, as it adjusts to a Poisson cluster process. This suggests a limitation for seed dispersal in the study fragment that is affecting orchid establishment. Furthermore, the small-scale spatial pattern of E. rhopalostele evidences a clustering that suggests a microsite preference for orchid establishment with a dispersal kernel of 0.4 m. Microsite features such as types of trees (dead trees or Clusia alata), shade tolerance trees, rough bark, distribution in the first meters suggest a tendency to prefer the understory for their establishment. Regarding plant-plant interaction a spatial segregation between adults and juveniles was present suggesting competition for limited resources conditioned for a microsite preference. Analysis of genetic structure of E. rhopalostele population through Structure and PCoA shows two genetic groups coexisting in this fragment and in the same phorophyte, possibly as a result of hybridization between sympatric species of Epidendrum. Our results of spatial autocorrelation analysis develop in GenAlex confirm a small-scale spatial-genetic structure within the genetic groups that is compatible with a short-distance dispersal mechanism caused by gravity or water run-off, instead of the long-distance seed dispersal promoted by wind generally attributed to orchids. For mycobionts identification ITS1-5.8S-ITS2 rDNA region was amplified. Phylogenetic analysis was performed with neighborjoining, Bayesian likelihood and maximum-likelihood for 47 sequences yielded two Tulasnella clades. This orchid establishes mycorrhizal associations with at least two different Tulasnella species. In some cases both fungi clades were present in same root, suggesting no limitation in fungal distribution. Concerning the implications for in situ conservation resulting from this work, the preservation of all forest fragment and their interactions (pollinators, mycorrhiza) is recommended to conserve the genetic diversity of this species. If a reintroduction were necessary, distances between individuals in each phorophyte within a range of 0.4 m, are recommended. To promote recruitment and regeneration of E. rhopalostele it is recommended that phorophytes correspond to dead or fallen trees or species, such as Clusia alata. Trees that have rough bark and are shade tolerant are also recommended. Furthermore, regarding vertical distribution, it is also convenient that orchids are located in the first meter (in understory, area with less light). In conclusion, limitation on seed dispersal, microsite characteristics, plant-plant interactions or interaction with cogeneric sympatric species and mycorrhizal preferences conditioned the presence of this epiphytic orchid in this fragment forest.

A comparative study on unconstrained hand biometrics

Biometrics applied to mobile devices are of great interest for security applications. Daily scenarios can benefit of a combination of both the most secure systems and most simple and extended devices. This document presents a hand biometric system oriented to mobile devices, proposing a non-intrusive, contact-less acquisition process where final users should take a picture of their hand in free-space with a mobile device without removals of rings, bracelets or watches. The main contribution of this paper is threefold: firstly, a feature extraction method is proposed, providing invariant hand measurements to previous changes; second contribution consists of providing a template creation based on hand geometric distances, requiring information from only one individual, without considering data from the rest of individuals within the database; finally, a proposal for template matching is proposed, minimizing the intra-class similarity and maximizing the inter-class likeliness. The proposed method is evaluated using three publicly available contact-less, platform-free databases. In addition, the results obtained with these databases will be compared to the results provided by two competitive pattern recognition techniques, namely Support Vector Machines (SVM) and k-Nearest Neighbour, often employed within the literature. Therefore, this approach provides an appropriate solution to adapt hand biometrics to mobile devices, with an accurate results and a non-intrusive acquisition procedure which increases the overall acceptance from the final user.

Unconstrained and Contactless Hand Geometry Biometrics

This paper presents a hand biometric system for contact-less, platform-free scenarios, proposing innovative methods in feature extraction, template creation and template matching. The evaluation of the proposed method considers both the use of three contact-less publicly available hand databases, and the comparison of the performance to two competitive pattern recognition techniques existing in literature: namely Support Vector Machines (SVM) and k-Nearest Neighbour (k-NN). Results highlight the fact that the proposed method outcomes existing approaches in literature in terms of computational cost, accuracy in human identification, number of extracted features and number of samples for template creation. The proposed method is a suitable solution for human identification in contact-less scenarios based on hand biometrics, providing a feasible solution to devices with limited hardware requirements like mobile devices

Text categorization methods for automatic estimation of verbal intelligence

In this paper we investigate whether conventional text categorization methods may suffice to infer different verbal intelligence levels. This research goal relies on the hypothesis that the vocabulary that speakers make use of reflects their verbal intelligence levels. Automatic verbal intelligence estimation of users in a spoken language dialog system may be useful when defining an optimal dialog strategy by improving its adaptation capabilities. The work is based on a corpus containing descriptions (i.e. monologs) of a short film by test persons yielding different educational backgrounds and the verbal intelligence scores of the speakers. First, a one-way analysis of variance was performed to compare the monologs with the film transcription and to demonstrate that there are differences in the vocabulary used by the test persons yielding different verbal intelligence levels. Then, for the classification task, the monologs were represented as feature vectors using the classical TF–IDF weighting scheme. The Naive Bayes, k-nearest neighbors and Rocchio classifiers were tested. In this paper we describe and compare these classification approaches, define the optimal classification parameters and discuss the classification results obtained.

Investigating verbal intelligence using the TF-IDF approach

In this paper we investigated differences in language use of speakers yielding different verbal intelligence when they describe the same event. The work is based on a corpus containing descriptions of a short film and verbal intelligence scores of the speakers. For analyzing the monologues and the film transcript, the number of reused words, lemmas, n-grams, cosine similarity and other features were calculated and compared to each other for different verbal intelligence groups. The results showed that the similarity of monologues of higher verbal intelligence speakers was greater than of lower and average verbal intelligence participants. A possible explanation of this phenomenon is that candidates yielding higher verbal intelligence have a better short-term memory. In this paper we also checked a hypothesis that differences in vocabulary of speakers yielding different verbal intelligence are sufficient enough for good classification results. For proving this hypothesis, the Nearest Neighbor classifier was trained using TF-IDF vocabulary measures. The maximum achieved accuracy was 92.86%.

Incorporating Group Recommendations to Recommender Systems: Alternatives and Performance

La importancia de los sistemas de recomendación ha experimentado un crecimiento exponencial como consecuencia del auge de las redes sociales. En esta tesis doctoral presentaré una amplia visión sobre el estado del arte de los sistemas de recomendación. Incialmente, estos estaba basados en fitrado demográfico, basado en contendio o colaborativo. En la actualidad, estos sistemas incorporan alguna información social al proceso de recomendación. En el futuro utilizarán información implicita, local y personal proveniente del Internet de las cosas. Los sistemas de recomendación basados en filtrado colaborativo se pueden modificar con el fin de realizar recomendaciones a grupos de usuarios. Existen trabajos previos que han incluido estas modificaciones en diferentes etapas del algoritmo de filtrado colaborativo: búsqueda de los vecinos, predicción de las votaciones y elección de las recomendaciones. En esta tesis doctoral proporcionaré un nuevo método que realizar el proceso de unficación (pasar de varios usuarios a un grupo) en el primer paso del algoritmo de filtrado colaborativo: cálculo de la métrica de similaridad. Proporcionaré una formalización completa del método propuesto. Explicaré cómo obtener el conjunto de k vecinos del grupo de usuarios y mostraré cómo obtener recomendaciones usando dichos vecinos. Asimismo, incluiré un ejemplo detallando cada paso del método propuesto en un sistema de recomendación compuesto por 8 usuarios y 10 items. Las principales características del método propuesto son: (a) es más rápido (más eficiente) que las alternativas proporcionadas por otros autores, y (b) es al menos tan exacto y preciso como otras soluciones estudiadas. Para contrastar esta hipótesis realizaré varios experimentos que miden la precisión, la exactitud y el rendimiento del método. Los resultados obtenidos se compararán con los resultados de otras alternativas utilizadas en la recomendación de grupos. Los experimentos se realizarán con las bases de datos de MovieLens y Netflix. ABSTRACT The importance of recommender systems has grown exponentially with the advent of social networks. In this PhD thesis I will provide a wide vision about the state of the art of recommender systems. They were initially based on demographic, contentbased and collaborative filtering. Currently, these systems incorporate some social information to the recommendation process. In the future, they will use implicit, local and personal information from the Internet of Things. As we will see here, recommender systems based on collaborative filtering can be used to perform recommendations to group of users. Previous works have made this modification in different stages of the collaborative filtering algorithm: establishing the neighborhood, prediction phase and determination of recommended items. In this PhD thesis I will provide a new method that carry out the unification process (many users to one group) in the first stage of the collaborative filtering algorithm: similarity metric computation. I will provide a full formalization of the proposed method. I will explain how to obtain the k nearest neighbors of the group of users and I will show how to get recommendations using those users. I will also include a running example of a recommender system with 8 users and 10 items detailing all the steps of the method I will present. The main highlights of the proposed method are: (a) it will be faster (more efficient) that the alternatives provided by other authors, and (b) it will be at least as precise and accurate as other studied solutions. To check this hypothesis I will conduct several experiments measuring the accuracy, the precision and the performance of my method. I will compare these results with the results generated by other methods of group recommendation. The experiments will be carried out using MovieLens and Netflix datasets.

Maintaining distances with the engineer: patterns of coexistence in plant communities beyond the patch-bare dichotomy

Two-phase plant communities with an engineer conforming conspicuous patches and affecting the performance and patterns of coexisting species are the norm under stressful conditions. To unveil the mechanisms governing coexistence in these communities at multiple spatial scales, we have developed a new point-raster approach of spatial pattern analysis, which was applied to a Mediterranean high mountain grassland to show how Festuca curvifolia patches affect the local distribution of coexisting species. We recorded 22 111 individuals of 17 plant perennial species. Most coexisting species were negatively associated with F. curvifolia clumps. Nevertheless, bivariate nearest-neighbor analyses revealed that the majority of coexisting species were confined at relatively short distances from F. curvifolia borders (between 0-2 cm and up to 8 cm in some cases). Our study suggests the existence of a fine-scale effect of F. curvifolia for most species promoting coexistence through a mechanism we call 'facilitation in the halo'. Most coexisting species are displaced to an interphase area between patches, where two opposite forces reach equilibrium: attenuated severe conditions by proximity to the F. curvifolia canopy (nutrient-rich islands) and competitive exclusion mitigated by avoiding direct contact with F. curvifolia.

Age regression from soft aligned face images using low computational resources

The initial step in most facial age estimation systems consists of accurately aligning a model to the output of a face detector (e.g. an Active Appearance Model). This fitting process is very expensive in terms of computational resources and prone to get stuck in local minima. This makes it impractical for analysing faces in resource limited computing devices. In this paper we build a face age regressor that is able to work directly on faces cropped using a state-of-the-art face detector. Our procedure uses K nearest neighbours (K-NN) regression with a metric based on a properly tuned Fisher Linear Discriminant Analysis (LDA) projection matrix. On FG-NET we achieve a state-of-the-art Mean Absolute Error (MAE) of 5.72 years with manually aligned faces. Using face images cropped by a face detector we get a MAE of 6.87 years in the same database. Moreover, most of the algorithms presented in the literature have been evaluated on single database experiments and therefore, they report optimistically biased results. In our cross-database experiments we get a MAE of roughly 12 years, which would be the expected performance in a real world application.

Characterization of vineyard's canopy through fuzzy clustering and svm over color images

In this work we propose an image acquisition and processing methodology (framework) developed for performance in-field grapes and leaves detection and quantification, based on a six step methodology: 1) image segmentation through Fuzzy C-Means with Gustafson Kessel (FCM-GK) clustering; 2) obtaining of FCM-GK outputs (centroids) for acting as seeding for K-Means clustering; 3) Identification of the clusters generated by K-Means using a Support Vector Machine (SVM) classifier. 4) Performance of morphological operations over the grapes and leaves clusters in order to fill holes and to eliminate small pixels clusters; 5)Creation of a mosaic image by Scale-Invariant Feature Transform (SIFT) in order to avoid overlapping between images; 6) Calculation of the areas of leaves and grapes and finding of the centroids in the grape bunches. Image data are collected using a colour camera fixed to a mobile platform. This platform was developed to give a stabilized surface to guarantee that the images were acquired parallel to de vineyard rows. In this way, the platform avoids the distortion of the images that lead to poor estimation of the areas. Our preliminary results are promissory, although they still have shown that it is necessary to implement a camera stabilization system to avoid undesired camera movements, and also a parallel processing procedure in order to speed up the mosaicking process.

ANN and Fuzzy c-Means applied to environmental pollution prediction

Salamanca, situated in center of Mexico is among the cities which suffer most from the air pollution in Mexico. The vehicular park and the industry, as well as orography and climatic characteristics have propitiated the increment in pollutant concentration of Sulphur Dioxide (SO2). In this work, a Multilayer Perceptron Neural Network has been used to make the prediction of an hour ahead of pollutant concentration. A database used to train the Neural Network corresponds to historical time series of meteorological variables and air pollutant concentrations of SO2. Before the prediction, Fuzzy c-Means and K-means clustering algorithms have been implemented in order to find relationship among pollutant and meteorological variables. Our experiments with the proposed system show the importance of this set of meteorological variables on the prediction of SO2 pollutant concentrations and the neural network efficiency. The performance estimation is determined using the Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The results showed that the information obtained in the clustering step allows a prediction of an hour ahead, with data from past 2 hours.