Search results for "Pattern"

showing 10 items of 4203 documents

Análisis de métodos de validación cruzada para la obtención robusta de parámetros biofísicos

2015

[EN] Non-parametric regression methods are powerful statistical methods to retrieve biophysical parameters from remote sensing measurements. However, their performance can be affected by what has been presented during the training phase. To ensure robust retrievals, various cross-validation sub-sampling methods are often used, which allow to evaluate the model with subsets of the field dataset. Here, two types of cross-validation techniques were analyzed in the development of non-parametric regression models: hold-out and k-fold. Selected non-parametric linear regression methods were least squares Linear Regression (LR) and Partial Least Squares Regression (PLSR), and nonlinear methods were…

TeledeteccióGeography Planning and Developmentlcsh:G1-922Least squaresCross-validationValidación cruzadaProcesos gausianosHold-outAnàlisi de regressióLinear regressionStatisticsPartial least squares regressionEarth and Planetary Sciences (miscellaneous)MLRAbusiness.industryCross-validationRegression analysisPattern recognitionRegresión de Kernel RidgeAprendizaje automáticoRegressionK-foldHold-OutGeographyk-foldPrincipal component regressionArtificial intelligencebusinessKernel Ridge regressionNonlinear regressionGaussian process regressionlcsh:Geography (General)Revista de Teledetección
researchProduct

Metodo di Template Matching per l'Analisi di Immagini

2012

La presente invenzione si riferisce ad un metodo di Template Matching per l’analisi di immagini da ImmunoFluorescenza Indiretta (IFI) per la rivelazione e classificazione automatica di pattern autoanticorpali. L’invenzione qui presentata generalizza il metodo del Template Matching operando innovativamente il mapping del contenuto visuale dell’immagine con particolari funzioni discrete qui denominate “mappatori”; inoltre, utilizzando le informazioni provenienti dalla sovrapposizione dei vari mappatori con un metodo di confronto funzionale, realizza una funzione di correlazione originale. La metodologia descritta nel seguito presenta una flessibilità tale da renderla applicabile a qualsiasi p…

Template Matchingpattern recognitionImmunofluorescenza indirettabioimmaginiSettore FIS/07 - Fisica Applicata(Beni Culturali Ambientali Biol.e Medicin)
researchProduct

Segmentation d'images robuste appliqué à l'imagerie par résonance magnétique et l'échographie de la prostate

2012

Prostate segmentation in trans rectal ultrasound (TRUS) and magnetic resonanceimages (MRI) facilitates volume estimation, multi-modal image registration, surgicalplaning and image guided prostate biopsies. The objective of this thesis is to developshape and region prior deformable models for accurate, robust and computationallyefficient prostate segmentation in TRUS and MRI images. Primary contributionof this thesis is in adopting a probabilistic learning approach to achieve soft classificationof the prostate for automatic initialization and evolution of a shape andregion prior deformable models for prostate segmentation in TRUS images. Twodeformable models are developed for the purpose. An…

Tesis i dissertacions acadèmiques616.6[INFO.INFO-OH]Computer Science [cs]/Other [cs.OH]Physics::Medical PhysicsProstate segmentation68Segmentación de la próstataNo english key word[ SDV.MHEP ] Life Sciences [q-bio]/Human health and pathologySegmentació de la próstata616.6 - Patologia del sistema genitourinariImagen de ultrasonidoPas de mots-clés en français[SDV.MHEP] Life Sciences [q-bio]/Human health and pathologyProstate cancerImatge d'ultrasóComputer Science::Other[INFO.INFO-OH] Computer Science [cs]/Other [cs.OH]68 - Indústries oficis i comerç d'articles acabats. Tecnologia cibernètica i automàticaCancer de próstataImatges mèdiquesComputer Science::Computer Vision and Pattern RecognitionImágenes médicasImatge de ressonància magnèticaMRI imagingUS imaging[ INFO.INFO-OH ] Computer Science [cs]/Other [cs.OH]Imagen de resonancia magnéticaMedical imaging[SDV.MHEP]Life Sciences [q-bio]/Human health and pathology
researchProduct

Review of Non-English Corpora Annotated for Emotion Classification in Text

2020

In this paper we try to systematize the information about the available corpora for emotion classification in text for languages other than English with the goal to find what approaches could be used for low-resource languages with close to no existing works in the field. We analyze the corresponding volume, emotion classification schema, language of each corresponding corpus and methods employed for data preparation and annotation automation. We’ve systematized twenty-four papers representing the corpora and found that corpora were mostly for the most spoken world languages: Hindi, Chinese, Turkish, Arabic, Japanese etc. A typical corpus contained several thousand of manually-annotated ent…

Text corpusHindiArtificial neural networkTurkishComputer sciencebusiness.industryEmotion classificationcomputer.software_genrelanguage.human_languageAnnotationNaive Bayes classifierComputingMethodologies_PATTERNRECOGNITIONSchema (psychology)languageArtificial intelligencebusinesscomputerNatural language processing
researchProduct

Development of an Automatic Pollen Classification System Using Shape, Texture and Aperture Features

2015

International audience; Automatic detection and classification of pollen species has value for use inside of palynologic allergen studies. Traditional labeling of different pollen species requires an expert biologist to classify particles by sight, and is therefore time-consuming and expensive. Here, an automatic process is developed which segments the particle contour and uses the extracted features for the classification process. We consider shape features, texture features and aperture features and analyze which are useful. The texture features analyzed include: Gabor Filters, Fast Fourier Transform, Local Binary Patterns, Histogram of Oriented Gradients, and Haralick features. We have s…

Texture classification[INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]Image processingMachine learningComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONPollen[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]
researchProduct

Integrazione di celle fotovoltaiche ibride nel vetromattone

2012

L’invenzione riguarda l’integrazione del vetromattone con celle solari ibride (organiche/inorganiche), conosciute come Dye-sensitized Solar Cell (DSC). La combinazione del vetromattone con le DSC consente di migliorare le prestazioni del prodotto originario rendendolo in grado di produrre energia pulita. Le caratteristiche di trasparenza e isolamento termo-acustico del prodotto possono essere regolate agendo sulla configurazione delle DSC. Il prodotto può essere assemblato a formare pannelli fotovoltaici per la realizzazione di facciate traslucide energeticamente “attive” in varie condizioni luminose (condizioni di luce diffusa o artificiale) e indipendentemente dall’angolo di radiazione so…

The integration of Dye-sensitized Solar Cell (DSC) with the glassblock is disclosed. The combination of the two technological elements the glassblock and the DSC module enhance their own peculiarities. The final complete product is ready-made for installation owns the qualities of the glassblock and produces sustainable energy at the same time. The transparency and the thermo-acoustic insulation may be modulated according to the specific requirements of each case of installation. The product may be used for example to create PV panels for glazed and translucent façades that are able to perform efficiently in all light conditions (even in diffuse light also converting internal artificial light) that are not dependant on the angle of the solar radiation and do not suffer from the high operative temperatures of functioning due to the peculiarities of the DSCs. DSCs are made of economic and biocompatible materials and are produced with low-cost and highly productive printing methods on rigid or flexible substrates. With the integration of DSC modules with the glassblock PV panels coincide with the envelope itself so that they do not need any further structures to be installed on and allow the possibility of subtract from the price of the panel the price of the substituted building element allowing further economies. DSCs can be printed in different colours and with various patterns offering to designers and customers the possibility of intervening on the glassblock aesthetical aspect also taking advantage of the peculiar modularity of the glassblock for larger compositions of drawings on the building’s façade. Seven Hypotheses of integration are discussed in order to obtain DSC-integrated glassblock ready for the construction of a precast and dry-assembled PV panel which can use the horizontal and vertical cavities between the glassblocks to house plastic profiles where the electric interconnections and the steel bars -useful for improve the mechanical resistance of the panel- are integrated.Settore ICAR/10 - Architettura Tecnica
researchProduct

Canopy Architecture Appraisal by Fractal Dimension of 'Flordastar' Peach Trees Grafted onto Different Rootstocks

2007

The objective of this research was to evaluate the modification of canopy architecture of ''Flordastar'' peach (Prunus persica L. Batsch) grafted onto rootstocks with different vigour, by the use of fractal dimension (D). The hypothesis was that different vigour rootstocks are able to modify the complexity of the branching pattern and that this effect can be assessed by a geometric parameter such as the fractal dimension (D) of the 2D projection of tree branching structure. The observations were carried out in a four-year-old experimental orchard of cv. ''Flordastar'' peach trees grafted onto Ishtara, Barrier, GF677 and MrS 2/5 rootstocks. On digital pictures of leafless, dormant peach tree…

The objective of this research was to evaluate the modification of canopy architecture of ''Flordastar'' peach (Prunus persica L. Batsch) grafted onto rootstocks with different vigour by the use of fractal dimension (D). The hypothesis was that different vigour rootstocks are able to modify the complexity of the branching pattern and that this effect can be assessed by a geometric parameter such as the fractal dimension (D) of the 2D projection of tree branching structure. The observations were carried out in a four-year-old experimental orchard of cv. ''Flordastar'' peach trees grafted onto Ishtara Barrier GF677 and MrS 2/5 rootstocks. On digital pictures of leafless dormant peach trees the tree skeleton was manually depicted by an image analysis software to measure total length and angles of insertion of branches. Bitmaps of the skeleton were then subjected to a specific software (FDCALC) able to calculate D by the "box counting" method. To account for the error introduced by analyzing a three dimensional structure with its 2D projection measurements were done on two images per tree taken from two different viewing angles. Results are shown in relation to tree vigour of the scion/stock combination. Fractal dimension was related to the vigour ranking of the graft combinations tested. In particular vigorous scion/stock combinations showed the highest D.HorticultureCanopy architectureHorticultureRootstockFractal dimensionMathematics
researchProduct

A solution to the stochastic point location problem in metalevel nonstationary environments.

2008

This paper reports the first known solution to the stochastic point location (SPL) problem when the environment is nonstationary. The SPL problem involves a general learning problem in which the learning mechanism (which could be a robot, a learning automaton, or, in general, an algorithm) attempts to learn a "parameter," for example, lambda*, within a closed interval. However, unlike the earlier reported results, we consider the scenario when the learning is to be done in a nonstationary setting. For each guess, the environment essentially informs the mechanism, possibly erroneously (i.e., with probability p), which way it should move to reach the unknown point. Unlike the results availabl…

Theoretical computer scienceAutomatic controlDiscretizationComputer scienceInformation Storage and RetrievalDecision Support TechniquesPattern Recognition AutomatedArtificial IntelligenceComputer SimulationElectrical and Electronic EngineeringStochastic ProcessesModels StatisticalLearning automatabusiness.industryStochastic processSignal Processing Computer-AssistedGeneral MedicineRandom walkComputer Science ApplicationsAutomatonHuman-Computer InteractionControl and Systems EngineeringPoint locationArtificial intelligencebusinessSoftwareAlgorithmsInformation SystemsIEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society
researchProduct

Increasing the Inference and Learning Speed of Tsetlin Machines with Clause Indexing

2020

The Tsetlin Machine (TM) is a machine learning algorithm founded on the classical Tsetlin Automaton (TA) and game theory. It further leverages frequent pattern mining and resource allocation principles to extract common patterns in the data, rather than relying on minimizing output error, which is prone to overfitting. Unlike the intertwined nature of pattern representation in neural networks, a TM decomposes problems into self-contained patterns, represented as conjunctive clauses. The clause outputs, in turn, are combined into a classification decision through summation and thresholding, akin to a logistic regression function, however, with binary weights and a unit step output function. …

Theoretical computer scienceContextual image classificationArtificial neural networkLearning automataComputer scienceSentiment analysisSearch engine indexingPattern recognition (psychology)OverfittingMNIST database
researchProduct

Text Compression Using Antidictionaries

1999

International audience; We give a new text compression scheme based on Forbidden Words ("antidictionary"). We prove that our algorithms attain the entropy for balanced binary sources. They run in linear time. Moreover, one of the main advantages of this approach is that it produces very fast decompressors. A second advantage is a synchronization property that is helpful to search compressed data and allows parallel compression. Our algorithms can also be presented as "compilers" that create compressors dedicated to any previously fixed source. The techniques used in this paper are from Information Theory and Finite Automata.

Theoretical computer scienceFinite-state machineComputer science[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS]010102 general mathematicsforbidden wordData_CODINGANDINFORMATIONTHEORY0102 computer and information sciencesInformation theory01 natural sciencesfinite automatonParallel compressionpattern matching010201 computation theory & mathematicsEntropy (information theory)Pattern matching0101 mathematicsTime complexityAlgorithmdata compressioninformation theoryData compression
researchProduct