Towards natural human computer interaction in BCI

BCI systems require correct classification of signals interpreted from the brain for useful operation. To this end this paper investigates a method proposed in [1] to correctly classify a series of images presented to a group of subjects in [2]. We show that it is possible to use the proposed methods to correctly recognise the original stimuli presented to a subject from analysis of their EEG. Additionally we use a verification set to show that the trained classification method can be applied to a different set of data. We go on to investigate the issue of invariance in EEG signals. That is, the brain representation of similar stimuli is recognisable across different subjects. Finally we consider the usefulness of the methods investigated towards an improved BCI system and discuss how it could potentially lead to great improvements in the ease of use for the end user by offering an alternative, more intuitive control based mode of operation.

The neuropsychology of male adults with High-Functioning Autism or Asperger Syndrome

Autism Spectrum Disorder (ASD) is diagnosed on the basis of behavioral symptoms, but cognitive abilities may also be useful in characterizing individuals with ASD. One hundred seventy-eight high-functioning male adults, half with ASD and half without, completed tasks assessing IQ, a broad range of cognitive skills, and autistic and comorbid symptomatology. The aims of the study were, first, to determine whether significant differences existed between cases and controls on cognitive tasks, and whether cognitive profiles, derived using a multivariate classification method with data from multiple cognitive tasks, could distinguish between the two groups. Second, to establish whether cognitive skill level was correlated with degree of autistic symptom severity, and third, whether cognitive skill level was correlated with degree of comorbid psychopathology. Fourth, cognitive characteristics of individuals with Asperger Syndrome (AS) and high-functioning autism (HFA) were compared. After controlling for IQ, ASD and control groups scored significantly differently on tasks of social cognition, motor performance, and executive function (P's < 0.05). To investigate cognitive profiles, 12 variables were entered into a support vector machine (SVM), which achieved good classification accuracy (81%) at a level significantly better than chance (P < 0.0001). After correcting for multiple correlations, there were no significant associations between cognitive performance and severity of either autistic or comorbid symptomatology. There were no significant differences between AS and HFA groups on the cognitive tasks. Cognitive classification models could be a useful aid to the diagnostic process when used in conjunction with other data sources-including clinical history.

Spatio-temporal variability of polynya dynamics and ice production in the Laptev Sea between the winters of 1979/80 and 2007/08

Polynyas in the Laptev Sea are examined with respect to recurrence and interannual wintertime ice production.We use a polynya classification method based on passive microwave satellite data to derive daily polynya area from long-term sea-ice concentrations. This provides insight into the spatial and temporal variability of open-water and thin-ice regions on the Laptev Sea Shelf. Using thermal infrared satellite data to derive an empirical thin-ice distribution within the thickness range from 0 to 20 cm, we calculate daily average surface heat loss and the resulting wintertime ice formation within the Laptev Sea polynyas between 1979 and 2008 using reanalysis data supplied by the National Centers for Environmental Prediction, USA, as atmospheric forcing. Results indicate that previous studies significantly overestimate the contribution of polynyas to the ice production in the Laptev Sea. Average wintertime ice production in polynyas amounts to approximately 55 km39 27% and is mostly determined by the polynya area, wind speed and associated large-scale circulation patterns. No trend in ice production could be detected in the period from 1979/80 to 2007/08.

ENHANCING MULTISCALE FRACTAL DESCRIPTORS USING FUNCTIONAL DATA ANALYSIS

This work presents a novel approach in order to increase the recognition power of Multiscale Fractal Dimension (MFD) techniques, when applied to image classification. The proposal uses Functional Data Analysis (FDA) with the aim of enhancing the MFD technique precision achieving a more representative descriptors vector, capable of recognizing and characterizing more precisely objects in an image. FDA is applied to signatures extracted by using the Bouligand-Minkowsky MFD technique in the generation of a descriptors vector from them. For the evaluation of the obtained improvement, an experiment using two datasets of objects was carried out. A dataset was used of characters shapes (26 characters of the Latin alphabet) carrying different levels of controlled noise and a dataset of fish images contours. A comparison with the use of the well-known methods of Fourier and wavelets descriptors was performed with the aim of verifying the performance of FDA method. The descriptor vectors were submitted to Linear Discriminant Analysis (LDA) classification method and we compared the correctness rate in the classification process among the descriptors methods. The results demonstrate that FDA overcomes the literature methods (Fourier and wavelets) in the processing of information extracted from the MFD signature. In this way, the proposed method can be considered as an interesting choice for pattern recognition and image classification using fractal analysis.

ANFIS BASED MODELS FOR ACCESSING QUALITY OF WIKIPEDIA ARTICLES

Wikipedia is a free, web-based, collaborative, multilingual encyclopedia project supported by the non-profit Wikimedia Foundation. Due to the free nature of Wikipedia and allowing open access to everyone to edit articles the quality of articles may be affected. As all people don’t have equal level of knowledge and also different people have different opinions about a topic so there may be difference between the contributions made by different authors. To overcome this situation it is very important to classify the articles so that the articles of good quality can be separated from the poor quality articles and should be removed from the database. The aim of this study is to classify the articles of Wikipedia into two classes class 0 (poor quality) and class 1(good quality) using the Adaptive Neuro Fuzzy Inference System (ANFIS) and data mining techniques. Two ANFIS are built using the Fuzzy Logic Toolbox [1] available in Matlab. The first ANFIS is based on the rules obtained from J48 classifier in WEKA while the other one was built by using the expert’s knowledge. The data used for this research work contains 226 article’s records taken from the German version of Wikipedia. The dataset consists of 19 inputs and one output. The data was preprocessed to remove any similar attributes. The input variables are related to the editors, contributors, length of articles and the lifecycle of articles. In the end analysis of different methods implemented in this research is made to analyze the performance of each classification method used.

Utilizing Remote Sensing Imagery to Monitor Vegetation Change within World Heritage Sites

ABSTRACT World Heritage sites provide a glimpse into the stories and civilizations of the past. There are currently 1007 unique World Heritage properties with 779 being classified as cultural sites, 197 as natural sites, and 31 falling into the categories of both cultural and natural sites (UNESCO & World Heritage Centre, 1992-2015). However, of these 1007 World Heritage sites, at least 46 are categorized as in danger and this number continues to grow. These unique and irreplaceable sites are exceptional because of their universality. Consequently, since World Heritage sites belong to all the people of the world and provide inspiration and admiration to all who visit them, it is our responsibility to help preserve these sites. The key form of preservation involves the individual monitoring of each site over time. While traditional methods are still extremely valuable, more recent advances in the field of geographic and spatial technologies including geographic information systems (GIS), laser scanning, and remote sensing, are becoming more beneficial for the monitoring and overall safeguarding of World Heritage sites. Through the employment and analysis of more accurately detailed spatial data, World Heritage sites can be better managed. There is a strong urgency to protect these sites. The purpose of this thesis is to describe the importance of taking care of World Heritage sites and to depict a way in which spatial technologies can be used to monitor and in effect preserve World Heritage sites through the utilization of remote sensing imagery. The research conducted in this thesis centers on the Everglades National Park, a World Heritage site that is continually affected by changes in vegetation. Data used include Landsat satellite imagery that dates from 2001-2003, the Everglades' boundaries shapefile, and Google Earth imagery. In order to conduct the in-depth analysis of vegetation change within the selected World Heritage site, three main techniques were performed to study changes found within the imagery. These techniques consist of conducting supervised classification for each image, incorporating a vegetation index known as Normalized Vegetation Index (NDVI), and utilizing the change detection tool available in the Environment for Visualizing Images (ENVI) software. With the research and analysis conducted throughout this thesis, it has been shown that within the three year time span (2001-2003), there has been an overall increase in both areas of barren soil (5.760%) and areas of vegetation (1.263%) with a decrease in the percentage of areas classified as sparsely vegetated (-6.987%). These results were gathered through the use of the maximum likelihood classification process available in the ENVI software. The results produced by the change detection tool which further analyzed vegetation change correlate with the results produced by the classification method. As well, by utilizing the NDVI method, one is able to locate changes by selecting a specific area and comparing the vegetation index generated for each date. It has been found that through the utilization of remote sensing technology, it is possible to monitor and observe changes featured within a World Heritage site. Remote sensing is an extraordinary tool that can and should be used by all site managers and organizations whose goal it is to preserve and protect World Heritage sites. Remote sensing can be used to not only observe changes over time, but it can also be used to pinpoint threats within a World Heritage site. World Heritage sites are irreplaceable sources of beauty, culture, and inspiration. It is our responsibility, as citizens of this world, to guard these treasures.

Desenvolvimento de um sistema hierárquico de classificação com aplicação baseada em abordagem Fuzzy

Land cover mappings represent important tools for the regional planning. However, the current mappings are related to very specific purposes and, consequently, they are limited in their capacity to define the wide variety of existing types of land cover. In that context, this paper aims at developing a wide and including hierarchical classification system for land cover mapping in regional scale, which should contribute for a future standardization of classes. Besides, it is intended to test that system for a study case that contemplates the use of a classification method based on fuzzy approach, which has shown to be more appropriate than conventional approaches. Therefore, it was proposed a hierarchical classification system with three detailing levels and a study case was defined with the specification of the test area and of the classification project. Then, the georreferencing of a TM/Landsat-5 image that comprises the test area was carried out. Later, it was applied a fuzzy classification approach in the TM/Landsat-5 image, starting from images of probability for the mapped classes and an uncertainty image were generated. Finally, it was produced a conventional output that represents the thematic mapping of the test area.

Metodologia para estudo do volume e densidade absoluta da placenta humana de termo

OBJETIVO: comparar duas metodologias para o cálculo do volume placentário em gestações normais de termo: a do princípio de Arquimedes e a do volume do cilindro, para estimar a densidade absoluta da placenta. Definir a metodologia mais adequada para o cálculo do volume e densidade placentários, que se relacione com o peso e classificação do recém-nascido. MÉTODOS: foram estudadas 50 placentas provenientes de gestações de termo sem complicações e calculados o volume e a densidade absoluta placentários: a) pelo princípio de Arquimedes e b) na suposição de que a placenta seria uma secção de cilindro com duas alturas diferentes do bolo placentário: com a altura média e com a altura da média aritmética do centro e extremidades. As densidades absolutas placentárias foram calculadas pelo quociente entre o peso ao ar da placenta e os diferentes volumes. RESULTADOS: a maioria das gestantes eram multíparas, idade média de 25,4 anos, volume placentário médio entre 547,8 e 610 cm³ e densidade média entre 0,94 e 1,14 g/cm³, dependendo da metodologia empregada. CONCLUSÕES: a metodologia mais adequada para estimar o volume placentário no termo foi a do princípio de Arquimedes, pela melhor correlação com o peso dos recém-nascidos, o índice placentário e a classificação do peso dos recém-nascidos em relação à idade gestacional.

Detecção de regiões de via integrando dados de varredura a LASER aerotransportado e imagens de alta resolução

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Terceiro setor: a gestão do pensar, sentir, agir

Pós-graduação em Serviço Social - FCHS

Identificação da dinâmica espaço-temporal para estimar área cultivada de soja a partir de imagens MODIS no Rio Grande do Sul

The objective of this study was to define a method for estimating soybean crop area in the Northern Rio Grande do Sul state (Brazil). Overall, six different remote sensing methods were proposed based on spectral-temporal profile and minimum and maximum values of NDVI/MODIS related to the stages of sowing, maximum development and harvesting of soybean areas. The resulting estimates were compared to official crop area data provided by the Brazilian government, using statistical analysis and the fuzzy similarity method. The performance of each method depended on information such as crop size, type of crop management, and sowing/harvesting dates. Regression coefficients of determination and fuzzy agreement values were above 0.8 and 0.45, respectively, for all methods. For operational monitoring of soybean crop area, the empirical threshold applied to the image difference with inclusion of harvest image method was the most effective, producing estimates that matched closely the official data. For spatial analysis the application of multitemporal images classification method is recommended that generated a map of better quality. The efficiency of these methods should be evaluated in the areas of soybean expansion in the state.

Olfactory reference syndrome: a still open nosological and treatment debate

Objective: The objective was to report a case of olfactory reference syndrome (ORS) with several co-occurring disorders and to discuss ORS differential diagnoses, diagnostic criteria and classification.Method: Case report.Results: A 37-year-old married woman presented overvalued ideas of having bad breath since adolescence. Shemet current diagnostic criteria for social anxiety disorder, specific phobia, obsessive-compulsive disorder, generalized anxiety disorder, body dysmorphic disorder and major depressive disorder. ORS similarities and differences with some related disorders are discussed.Conclusion: Further studies regarding symptoms, biomarkers and outcomes are needed to fully disentangle ORS from existing depressive, anxiety and obsessive-compulsive spectrum disorders. (C) 2014 Elsevier Inc. All rights reserved.

A multisensor, multitemporal approach for monitoring herbaceous vegetation growth in the Amazon floodplain

The Amazon River floodplain is an important source of atmospheric CO2 and CH4. Aquatic herbaceous vegetation (macrophytes) have been shown to contribute significantly to floodplain net primary productivity (NPP) and methane emission in the region. Their fast growth rates under both flooded and dry conditions make herbaceous vegetation the most variable element in the Amazon floodplain NPP budget, and the most susceptible to environmental changes. The present study combines multitemporal Radarsat-1 and MODIS images to monitor spatial and temporal changes in herbaceous vegetation cover in the Amazon floodplain. Radarsat-1 images were acquired from Dec/2003 to Oct/2005, and MODIS daily surface reflectance products were acquired for the two cloud-free dates closest to each Radarsat-1 acquisition. An object-based, hierarchical algorithm was developed using the temporal SAR information to discriminate Permanent Open Water (OW), Floodplain (FP) and Upland (UL) classes at Level 1, and then subdivide the FP class into Woody Vegetation (WV) and Possible Macrophytes (PM) at Level 2. At Level 3, optical and SAR information were combined to discriminate actual herbaceous cover at each date. The resulting maps had accuracies ranging from 80% to 90% for Level 1 and 2 classifications, and from 60% to 70% for Level 3 classifications, with kappa values ranging between 0.7 and 0.9 for Levels 1 and 2 and between 0.5 and 0.6 for Level 3. All study sites had noticeable variations in the extent of herbaceous cover throughout the hydrological year, with maximum areas up to four times larger than minimum areas. The proposed classification method was able to capture the spatial pattern of macrophyte growth and development in the studied area, and the multitemporal information was essential for both separating vegetation cover types and assessing monthly variation in herbaceous cover extent.

Monitoramento da condição da ferramenta de dressagem usando sinais de vibração e modelos neurais

Pós-graduação em Engenharia Elétrica - FEB

Feature extraction in pressure signals for leak detection in water networks

Techniques based on signal analysis for leak detection in water supply systems typically use long pressure and/or flow data series of variable length. This paper presents the feature extraction from pressure signals and their application to the identification of changes related to the onset of a leak. Example signals were acquired from an experimental laboratory circuit, and features were extracted from temporal domain and from transformed signals. Statistical analysis of features values and a classification method were applied. It was verified the feasibility of using feature vectors for distinguish data acquired in the absence or presence of a leak.