A Robust Multiple Feature Approach To Endpoint Detection In Car Environment Based On Advanced Classifiers

In this paper we propose an endpoint detection system based on the use of several features extracted from each speech frame, followed by a robust classifier (i.e Adaboost and Bagging of decision trees, and a multilayer perceptron) and a finite state automata (FSA). We present results for four different classifiers. The FSA module consisted of a 4-state decision logic that filtered false alarms and false positives. We compare the use of four different classifiers in this task. The look ahead of the method that we propose was of 7 frames, which are the number of frames that maximized the accuracy of the system. The system was tested with real signals recorded inside a car, with signal to noise ratio that ranged from 6 dB to 30dB. Finally we present experimental results demonstrating that the system yields robust endpoint detection.

3D noise power spectrum applied on clinical mdct scanners: effects of reconstruction algorithms and reconstruction filters

The noise power spectrum (NPS) is the reference metric for understanding the noise content in computed tomography (CT) images. To evaluate the noise properties of clinical multidetector (MDCT) scanners, local 2D and 3D NPSs were computed for different acquisition reconstruction parameters.A 64- and a 128-MDCT scanners were employed. Measurements were performed on a water phantom in axial and helical acquisition modes. CT dose index was identical for both installations. Influence of parameters such as the pitch, the reconstruction filter (soft, standard and bone) and the reconstruction algorithm (filtered-back projection (FBP), adaptive statistical iterative reconstruction (ASIR)) were investigated. Images were also reconstructed in the coronal plane using a reformat process. Then 2D and 3D NPS methods were computed.In axial acquisition mode, the 2D axial NPS showed an important magnitude variation as a function of the z-direction when measured at the phantom center. In helical mode, a directional dependency with lobular shape was observed while the magnitude of the NPS was kept constant. Important effects of the reconstruction filter, pitch and reconstruction algorithm were observed on 3D NPS results for both MDCTs. With ASIR, a reduction of the NPS magnitude and a shift of the NPS peak to the low frequency range were visible. 2D coronal NPS obtained from the reformat images was impacted by the interpolation when compared to 2D coronal NPS obtained from 3D measurements.The noise properties of volume measured in last generation MDCTs was studied using local 3D NPS metric. However, impact of the non-stationarity noise effect may need further investigations.

Biologia e tabela de vida de fertilidade de Agrotis ipsilon em dieta artificial

O objetivo deste trabalho foi desenvolver uma dieta artificial para criação de Agrotis ipsilon em laboratório com base em parâmetros biológicos e na tabela de vida de fertilidade. A dieta artificial utilizada continha feijão, caseína, proteína de soja, levedura e germe de trigo como fontes protéicas. Os parâmetros biológicos duração e viabilidade das fases larval e pupal, peso de pupas, de ambos os sexos, com 24 horas de idade, razão sexual, longevidade dos adultos, período de pré-oviposição e número de ovos produzidos por fêmea e a tabela de vida de fertilidade foram avaliados. Foram observados seis ínstares larvais com duração de 25,4 dias e viabilidade de 93%. A duração da fase pupal foi de 12,4 dias e viabilidade de 96%. A viabilidade de ciclo total foi 72%. O peso de pupas foi 387 mg (machos) e 484 mg (fêmeas). A razão sexual foi 0,46. O período de pré-oviposição foi de um dia, com 1.806 ovos por fêmea. Na tabela de vida verificou-se que a taxa líquida de reprodução e a razão finita de aumento foram 616,9 e 1,14, respectivamente. A dieta artificial é adequada à manutenção da criação de A. ipsilon, em laboratório.

Intersecciones entre visión artificial y mirada artística. Semantic Intersections between Artificial and Artistic Vision.

En el presente artículo se ha desarrollado un sistema capaz de categorizar de forma automática la base de datos de imágenes que sirven de punto de partida para la ideación y diseño en la producción artística del escultor M. Planas. La metodología utilizada está basada en características locales. Para la construcción de un vocabulario visual se sigue un procedimiento análogo al que se utiliza en el análisis automático de textos (modelo 'Bag-of-Words'-BOW) y en el ámbito de las imágenes nos referiremos a representaciones 'Bag-of-Visual Terms' (BOV). En este enfoque se analizan las imágenes como un conjunto de regiones, describiendo solamente su apariencia e ignorando su estructura espacial. Para superar los inconvenientes de polisemia y sinonimia que lleva asociados esta metodología, se utiliza el análisis probabilístico de aspectos latentes (PLSA) que detecta aspectos subyacentes en las imágenes, patrones formales. Los resultados obtenidos son prometedores y, además de la utilidad intrínseca de la categorización automática de imágenes, este método puede proporcionar al artista un punto de vista auxiliar muy interesante.

Biologia e tabela de vida de fertilidade de Hypercompe indecisa em dieta artificial

O objetivo deste trabalho foi estudar a biologia de Hypercompe indecisa e construir a tabela de vida de fertilidade em dieta artificial, para criação em laboratório. Foram determinados: duração e sobrevivência das fases de ovo, larva e pupa; número de estádios larvais; razão sexual; peso de pupas; longevidade; fecundidade; e períodos de pré-oviposição e oviposição. Os períodos embrionário, larval e pupal foram de 6, 25,4 e 64,3 dias, com sobrevivência de 92,7, 92 e 71,9%, respectivamente. Foram observados seis estádios larvais com durações variáveis. O peso de pupas foi: 1,04 g para fêmeas e 0,726 g para machos. O ciclo biológico foi de 95,6 dias com sobrevivência total de 61,3%. As fêmeas colocaram, em média, 1.531 ovos durante 8,3 dias, e com um período de pré-oviposição de dois dias. A longevidade média de machos e fêmeas foi de 21,9 e 21,8 dias, respectivamente, e a razão sexual de machos para fêmeas foi 0,54. A espécie H. indecisa aumentou 283 vezes a cada geração, e a duração média de uma geração é de 98 dias, e a razão finita de aumento é de 1,0593. A dieta artificial utilizada foi adequada para a criação de H. indecisa, em laboratório.

Biologia floral e polinização artificial de pinhão-manso no norte de Minas Gerais

O objetivo deste trabalho foi avaliar alguns aspectos da biologia floral e do sistema reprodutivo de Jatropha curcas, em Janaúba, MG. Foram registrados: o número de flores femininas e masculinas; o intervalo de abertura das flores femininas; e a formação de frutos por apomixia, autofecundação, geitonogamia e xenogamia. A proporção de flores masculinas para femininas foi de 20:1. O intervalo de abertura das flores femininas variou de um a sete dias, conforme o número delas na inflorescência. No teste de apomixia, houve formação de frutos em apenas 5% das flores avaliadas. A percentagem de frutificação variou de 79 a 88% na autofecundação manual, na geitonogamia e na xenogamia. Na autofecundação sem a polinização manual a frutificação foi de 20%, e os frutos formados foram significativamente menores, com número inferior de sementes por fruto e menor índice de velocidade de emergência. As sementes foram semelhantes às formadas por polinização natural. é possível a realização de cruzamentos controlados em pinhão-manso, e não há autoincompatibilidade nesta espécie.

Health consultation : Exposure to fluff material from the Shine Brothers and noise concerns -- Spencer, Iowa (2011)

This letter has been prepared as a consultation to determine some potential health concerns raised by a resident of Spencer from exposure to wire shredder fluff material emitted by Shine Brothers metal salvaging facility in Spencer, Iowa on July 10, 2011 and, a concern regarding the health impacts from the level of noise measured by a resident near the metal salvaging facility.

Sidebranching induced by external noise in solutal dendritic growth

We have studied sidebranching induced by fluctuations in dendritic growth. The amplitude of sidebranching induced by internal (equilibrium) concentration fluctuations in the case of solidification with solutal diffusion is computed. This amplitude turns out to be significantly smaller than values reported in previous experiments. The effects of other possible sources of fluctuations (of an external origin) are examined by introducing nonconserved noise in a phase-field model. This reproduces the characteristics of sidebranching found in experiments. Results also show that sidebranching induced by external noise is qualitatively similar to that of internal noise, and it is only distinguished by its amplitude.

Asymmetric stochastic switching driven by intrinsic molecular noise

Low-copy-number molecules are involved in many functions in cells. The intrinsic fluctuations of these numbers can enable stochastic switching between multiple steady states, inducing phenotypic variability. Herein we present a theoretical and computational study based on Master Equations and Fokker-Planck and Langevin descriptions of stochastic switching for a genetic circuit of autoactivation. We show that in this circuit the intrinsic fluctuations arising from low-copy numbers, which are inherently state-dependent, drive asymmetric switching. These theoretical results are consistent with experimental data that have been reported for the bistable system of the gallactose signaling network in yeast. Our study unravels that intrinsic fluctuations, while not required to describe bistability, are fundamental to understand stochastic switching and the dynamical relative stability of multiple states.

Biology and morphometrics of the immature stages of Epinotia aporema on artificial diet

The objective of this work was to establish a life table for the immature stages of Epinotia aporema, as part of a wider investigation on its biological control. Insects were reared on an artificial diet at 25±1ºC and a 16:8 (light:dark) hour photoperiod. For the identification of larval instars for the study of pathogen-insect interactions under laboratory conditions, head capsule widths (HCWs) were also determined. The egg incubation period was 4.13±0.30 days, larval stage took 11.64±0.49 days, and the development time of the pupal phase was sex-dependent with 8.51±0.69 days for the females and 9.41±0.65 days for the males. Five larval instars were identified.

Mineralogical alteration of artificial meteorites during atmospheric entry. The STONE-5 experiment

The generic concept of the artificial meteorite experiment STONE is to fix rock samples bearing microorganisms on the heat shield of a recoverable space capsule and to study their modifications during atmospheric re-entry. The STONE-5 experiment was performed mainly to answer astrobiological questions. The rock samples mounted on the heat shield were used (i) as a carrier for microorganisms and (ii) as internal control to verify whether physical conditions during atmospheric re-entry were comparable to those experienced by "real" meteorites. Samples of dolerite (an igneous rock), sandstone (a sedimentary rock), and gneiss impactite from Haughton Crater carrying endolithic cyanobacteria were fixed to the heat shield of the unmanned recoverable capsule FOTON-M2. Holes drilled on the back side of each rock sample were loaded with bacterial and fungal spores and with dried vegetative cryptoendoliths. The front of the gneissic sample was also soaked with cryptoendoliths. <p>The mineralogical differences between pre- and post-flight samples are detailed. Despite intense ablation resulting in deeply eroded samples, all rocks in part survived atmospheric re-entry. Temperatures attained during re-entry were high enough to melt dolerite, silica, and the gneiss impactite sample. The formation of fusion crusts in STONE-5 was a real novelty and strengthens the link with real meteorites. The exposed part of the dolerite is covered by a fusion crust consisting of silicate glass formed from the rock sample with an admixture of holder material (silica). Compositionally, the fusion crust varies from silica-rich areas (undissolved silica fibres of the holder material) to areas whose composition is "basaltic". Likewise, the fusion crust on the exposed gneiss surface was formed from gneiss with an admixture of holder material. The corresponding composition of the fusion crust varies from silica-rich areas to areas with "gneiss" composition (main component potassium-rich feldspar). The sandstone sample was retrieved intact and did not develop a fusion crust. Thermal decomposition of the calcite matrix followed by disintegration and liberation of the silicate grains prevented the formation of a melt.</p> <p>Furthermore, the non-exposed surface of all samples experienced strong thermal alterations. Hot gases released during ablation pervaded the empty space between sample and sample holder leading to intense local heating. The intense heating below the protective sample holder led to surface melting of the dolerite rock and to the formation of calcium-silicate rims on quartz grains in the sandstone sample. (c) 2008 Elsevier Ltd. All rights reserved.</p>

Crescimento de eucalipto sob efeito de desfolhamento artificial

O objetivo deste trabalho foi avaliar os efeitos do desfolhamento total, realizado após o plantio e ao longo do primeiro ano de cultivo, sobre o crescimento de Eucalyptus grandis, desde a implantação até ao corte do povoamento. Foram avaliados cinco tratamentos: sem desfolhamento; um desfolhamento aos 56 dias após o plantio (DAP); dois desfolhamentos, aos 56 e 143 DAP; dois desfolhamentos, aos 56 e 267 DAP; e três desfolhamentos, aos 56, 143 e 278 DAP. Foram mensurados os diâmetros do tronco a 1,3 m e a altura total de 60 árvores por tratamento, em oito avaliações, do 21º ao 92º mês de cultivo. O crescimento médio em cada tratamento foi descrito por modelos de regressão não lineares e comparados por testes de identidade para comparar as tendências entre a testemunha e os demais tratamentos. O desfolhamento causou reduções significativas nas taxas de crescimento em diâmetro e altura das plantas, e diminuição expressiva no faturamento ao final da rotação, mesmo quando realizado uma única vez, no início do plantio. Maiores danos, no entanto, foram verificados após consecutivos desfolhamentos ao longo do primeiro ano de cultivo. A manutenção de áreas que tenham sofrido desfolhamento total na fase inicial de plantio pode tornar-se uma medida economicamente inviável.

Robust Nonlinear Mapping of Soil Contamination Using Support Regression

Spatial data analysis mapping and visualization is of great importance in various fields: environment, pollution, natural hazards and risks, epidemiology, spatial econometrics, etc. A basic task of spatial mapping is to make predictions based on some empirical data (measurements). A number of state-of-the-art methods can be used for the task: deterministic interpolations, methods of geostatistics: the family of kriging estimators (Deutsch and Journel, 1997), machine learning algorithms such as artificial neural networks (ANN) of different architectures, hybrid ANN-geostatistics models (Kanevski and Maignan, 2004; Kanevski et al., 1996), etc. All the methods mentioned above can be used for solving the problem of spatial data mapping. Environmental empirical data are always contaminated/corrupted by noise, and often with noise of unknown nature. That's one of the reasons why deterministic models can be inconsistent, since they treat the measurements as values of some unknown function that should be interpolated. Kriging estimators treat the measurements as the realization of some spatial randomn process. To obtain the estimation with kriging one has to model the spatial structure of the data: spatial correlation function or (semi-)variogram. This task can be complicated if there is not sufficient number of measurements and variogram is sensitive to outliers and extremes. ANN is a powerful tool, but it also suffers from the number of reasons. of a special type ? multiplayer perceptrons ? are often used as a detrending tool in hybrid (ANN+geostatistics) models (Kanevski and Maignank, 2004). Therefore, development and adaptation of the method that would be nonlinear and robust to noise in measurements, would deal with the small empirical datasets and which has solid mathematical background is of great importance. The present paper deals with such model, based on Statistical Learning Theory (SLT) - Support Vector Regression. SLT is a general mathematical framework devoted to the problem of estimation of the dependencies from empirical data (Hastie et al, 2004; Vapnik, 1998). SLT models for classification - Support Vector Machines - have shown good results on different machine learning tasks. The results of SVM classification of spatial data are also promising (Kanevski et al, 2002). The properties of SVM for regression - Support Vector Regression (SVR) are less studied. First results of the application of SVR for spatial mapping of physical quantities were obtained by the authorsin for mapping of medium porosity (Kanevski et al, 1999), and for mapping of radioactively contaminated territories (Kanevski and Canu, 2000). The present paper is devoted to further understanding of the properties of SVR model for spatial data analysis and mapping. Detailed description of the SVR theory can be found in (Cristianini and Shawe-Taylor, 2000; Smola, 1996) and basic equations for the nonlinear modeling are given in section 2. Section 3 discusses the application of SVR for spatial data mapping on the real case study - soil pollution by Cs137 radionuclide. Section 4 discusses the properties of the modelapplied to noised data or data with outliers.

Comparison and validation of tissue modelization and statistical classification methods in T1-weighted MR brain images.

This paper presents a validation study on statistical nonsupervised brain tissue classification techniques in magnetic resonance (MR) images. Several image models assuming different hypotheses regarding the intensity distribution model, the spatial model and the number of classes are assessed. The methods are tested on simulated data for which the classification ground truth is known. Different noise and intensity nonuniformities are added to simulate real imaging conditions. No enhancement of the image quality is considered either before or during the classification process. This way, the accuracy of the methods and their robustness against image artifacts are tested. Classification is also performed on real data where a quantitative validation compares the methods' results with an estimated ground truth from manual segmentations by experts. Validity of the various classification methods in the labeling of the image as well as in the tissue volume is estimated with different local and global measures. Results demonstrate that methods relying on both intensity and spatial information are more robust to noise and field inhomogeneities. We also demonstrate that partial volume is not perfectly modeled, even though methods that account for mixture classes outperform methods that only consider pure Gaussian classes. Finally, we show that simulated data results can also be extended to real data.