Robot visual localization through local feature fusion: an evaluation of multiple classifiers combination approaches

In the last decade, local image features have been widely used in robot visual localization. In order to assess image similarity, a strategy exploiting these features compares raw descriptors extracted from the current image with those in the models of places. This paper addresses the ensuing step in this process, where a combining function must be used to aggregate results and assign each place a score. Casting the problem in the multiple classifier systems framework, in this paper we compare several candidate combiners with respect to their performance in the visual localization task. For this evaluation, we selected the most popular methods in the class of non-trained combiners, namely the sum rule and product rule. A deeper insight into the potential of these combiners is provided through a discriminativity analysis involving the algebraic rules and two extensions of these methods: the threshold, as well as the weighted modifications. In addition, a voting method, previously used in robot visual localization, is assessed. Furthermore, we address the process of constructing a model of the environment by describing how the model granularity impacts upon performance. All combiners are tested on a visual localization task, carried out on a public dataset. It is experimentally demonstrated that the sum rule extensions globally achieve the best performance, confirming the general agreement on the robustness of this rule in other classification problems. The voting method, whilst competitive with the product rule in its standard form, is shown to be outperformed by its modified versions.

GPU implementation of the simplex identification via split augmented Lagrangian

Hyperspectral imaging can be used for object detection and for discriminating between different objects based on their spectral characteristics. One of the main problems of hyperspectral data analysis is the presence of mixed pixels, due to the low spatial resolution of such images. This means that several spectrally pure signatures (endmembers) are combined into the same mixed pixel. Linear spectral unmixing follows an unsupervised approach which aims at inferring pure spectral signatures and their material fractions at each pixel of the scene. The huge data volumes acquired by such sensors put stringent requirements on processing and unmixing methods. This paper proposes an efficient implementation of a unsupervised linear unmixing method on GPUs using CUDA. The method finds the smallest simplex by solving a sequence of nonsmooth convex subproblems using variable splitting to obtain a constraint formulation, and then applying an augmented Lagrangian technique. The parallel implementation of SISAL presented in this work exploits the GPU architecture at low level, using shared memory and coalesced accesses to memory. The results herein presented indicate that the GPU implementation can significantly accelerate the method's execution over big datasets while maintaining the methods accuracy.

Parallel GPU architecture for hyperspectral unmixing based on augmented Lagrangian method

Hyperspectral imaging has become one of the main topics in remote sensing applications, which comprise hundreds of spectral bands at different (almost contiguous) wavelength channels over the same area generating large data volumes comprising several GBs per flight. This high spectral resolution can be used for object detection and for discriminate between different objects based on their spectral characteristics. One of the main problems involved in hyperspectral analysis is the presence of mixed pixels, which arise when the spacial resolution of the sensor is not able to separate spectrally distinct materials. Spectral unmixing is one of the most important task for hyperspectral data exploitation. However, the unmixing algorithms can be computationally very expensive, and even high power consuming, which compromises the use in applications under on-board constraints. In recent years, graphics processing units (GPUs) have evolved into highly parallel and programmable systems. Specifically, several hyperspectral imaging algorithms have shown to be able to benefit from this hardware taking advantage of the extremely high floating-point processing performance, compact size, huge memory bandwidth, and relatively low cost of these units, which make them appealing for onboard data processing. In this paper, we propose a parallel implementation of an augmented Lagragian based method for unsupervised hyperspectral linear unmixing on GPUs using CUDA. The method called simplex identification via split augmented Lagrangian (SISAL) aims to identify the endmembers of a scene, i.e., is able to unmix hyperspectral data sets in which the pure pixel assumption is violated. The efficient implementation of SISAL method presented in this work exploits the GPU architecture at low level, using shared memory and coalesced accesses to memory.

The visual image in Epiphanic Short History

The visual image is a fundamental component of epiphany, stressing its immediacy and vividness, corresponding to the enargeia of the traditional ekphrasis and also playing with cultural and social meanings. Morris Beja in his seminal book Epiphany in the Modern Novel, draws our attention to the distinction made by Joyce between the epiphany originated in a common object, in a discourse or gesture and the one arising in “a memorable phase of the mind itself”. This type materializes in the “dream-epiphany” and in the epiphany based in memory. On the other hand, Robert Langbaum in his study of the epiphanic mode, suggests that the category of “visionary epiphany” could account for the modern effect of an internally glowing vision like Blake’s “The Tyger”, which projects the vitality of a real tyger. The short story, whose length renders it a fitting genre for the use of different types of epiphany, has dealt with the impact of the visual image in this technique, to convey different effects and different aesthetic aims. This paper will present some examples of this occurrence in short stories of authors in whose work epiphany is a fundamental concept and literary technique: Walter Pater, Joseph Conrad, K. Mansfield, Clarice Lispector. Pater’s “imaginary portraits” concentrate on “priviledged moments” of the lives of the characters depicting their impressions through pictorial language; Conrad tries to show “moments of awakening” that can be remembered by the eye; Mansfield suggests that epiphany, the “glimpse”, should replace plot as an internal ordering principle of her impressionist short-stories; in C. Lispector the visualization of some situations is so aggressive that it causes nausea and a radical revelation on the protagonist’s.

Study of behavioural and neural bases of visuo-spatial working memory with an fMRI paradigm based on an n-back task

The goal of this study was to propose a new functional magnetic resonance imaging (fMRI) paradigm using a language-free adaptation of a 2-back working memory task to avoid cultural and educational bias. We additionally provide an index of the validity of the proposed paradigm and test whether the experimental task discriminates the behavioural performances of healthy participants from those of individuals with working memory deficits. Ten healthy participants and nine patients presenting working memory (WM) deficits due to acquired brain injury (ABI) performed the developed task. To inspect whether the paradigm activates brain areas typically involved in visual working memory (VWM), brain activation of the healthy participants was assessed with fMRIs. To examine the task's capacity to discriminate behavioural data, performances of the healthy participants in the task were compared with those of ABI patients. Data were analysed with GLM-based random effects procedures and t-tests. We found an increase of the BOLD signal in the specialized areas of VWM. Concerning behavioural performances, healthy participants showed the predicted pattern of more hits, less omissions and a tendency for fewer false alarms, more self-corrected responses, and faster reaction times, when compared with subjects presenting WM impairments. The results suggest that this task activates brain areas involved in VWM and discriminates behavioural performances of clinical and non-clinical groups. It can thus be used as a research methodology for behavioural and neuroimaging studies of VWM in block-design paradigms.

Real-Time Visual Ground-Truth System for Indoor Robotic Applications

The robotics community is concerned with the ability to infer and compare the results from researchers in areas such as vision perception and multi-robot cooperative behavior. To accomplish that task, this paper proposes a real-time indoor visual ground truth system capable of providing accuracy with at least more magnitude than the precision of the algorithm to be evaluated. A multi-camera architecture is proposed under the ROS (Robot Operating System) framework to estimate the 3D position of objects and the implementation and results were contextualized to the Robocup Middle Size League scenario.

Função visual e desempenho na leitura em crianças do 1º ciclo do ensino básico do concelho de Lisboa

RESUMO - Esta tese pretende ser um contributo para o estudo das anomalias da função visual e da sua influência no desempenho da leitura. Apresentava como objetivos: (1) Identificar a prevalência de anomalias da função visual, (2) Caracterizar o desempenho da leitura em crianças com e sem anomalias da função visual, (3) Identificar de que modo as anomalias da função visual influenciam o desempenho da leitura e (4) Identificar o impacto das variáveis que determinam o desempenho da leitura. Foi recolhida uma amostra de conveniência com 672 crianças do 1º ciclo do ensino básico de 11 Escolas do Concelho de Lisboa com idades compreendidas entre os 6 e os 11 anos (7,69±1,19), 670 encarregados de educação e 34 Professores. Para recolha de dados, foram utilizados três instrumentos: 2 questionários de perguntas fechadas, avaliação da função visual e prova de avaliação da leitura com 34 palavras. Após observadas, as crianças foram classificadas em dois grupos: função visual normal (FVN=562) e função visual alterada (FVA=110). Identificou-se uma prevalência de 16,4% de crianças com FVA. No teste de leitura, estas crianças apresentaram um menor número de palavras lidas corretamente (FVA=31,00; FVN=33,00; p<0,001) e menor precisão (FVA=91,18%; FVN=97,06%; p<0,001). Esta tendência também foi observada na comparação entre os 4 anos de escolaridade. As crianças com função visual alterada mostraram uma tendência para a omissão de letras e a confusão de grafema. Quanto à fluência (FVA=24,71; FVN=27,39; p=0,007) esta foi inferior nas crianças com FVA para todos os anos de escolaridade, exceto o 3º ano. As crianças com hipermetropia (p=0,003) e astigmatismo (p=0,019) não corrigido leram menos palavras corretamente (30,00; 31,00) e com menor precisão (88,24%; 91,18%) que as crianças sem erro refrativo significativo (32,00; 94,12%). A performance escolar classificada pelos professores foi inferior nas crianças com FVA e mais de ¼ necessitavam de medidas de apoio especial na escola. Não se verificaram diferenças significativas na performance da leitura das crianças com FVA por grupos de habilitações dos encarregados de educação. Verificou-se que o risco de ter um desempenho na leitura alterado é superior [OR=4,29; I.C.95%(2,49;7,38)] nas crianças que apresentam FVA. Relativamente ao 1º ano de escolaridade, o 2º, 3º e 4º anos apresentam um menor risco de ter um desempenho na leitura alterado. As variáveis método de ensino, habilitações dos encarregados de educação, tipo de escola (pública/privada), idade do Professor e número de anos de experiência do Professor, não foram fatores estatisticamente significativos para explicar a alteração do desempenho na leitura, quando o efeito da função visual se encontra contemplado no modelo. Um mau desempenho na leitura foi considerado nas crianças que apresentaram uma precisão inferior a 90%. Este indicador pode ser utilizado para identificar crianças em risco, que necessitam de uma observação Ortóptica/Oftalmológica para confirmação ou exclusão da existência de alterações da função visual. Este trabalho constitui um contributo para a identificação de crianças em desvantagem educacional devido a anomalias da função visual tratáveis, propondo um modelo que pretende orientar os professores na identificação de crianças que apresentem um baixo desempenho na leitura.

Minimal computation structures for visual information applications based on printed electronics

In the early nineties, Mark Weiser wrote a series of seminal papers that introduced the concept of Ubiquitous Computing. According to Weiser, computers require too much attention from the user, drawing his focus from the tasks at hand. Instead of being the centre of attention, computers should be so natural that they would vanish into the human environment. Computers become not only truly pervasive but also effectively invisible and unobtrusive to the user. This requires not only for smaller, cheaper and low power consumption computers, but also for equally convenient display solutions that can be harmoniously integrated into our surroundings. With the advent of Printed Electronics, new ways to link the physical and the digital worlds became available. By combining common printing techniques such as inkjet printing with electro-optical functional inks, it is starting to be possible not only to mass-produce extremely thin, flexible and cost effective electronic circuits but also to introduce electronic functionalities into products where it was previously unavailable. Indeed, Printed Electronics is enabling the creation of novel sensing and display elements for interactive devices, free of form factor. At the same time, the rise in the availability and affordability of digital fabrication technologies, namely of 3D printers, to the average consumer is fostering a new industrial (digital) revolution and the democratisation of innovation. Nowadays, end-users are already able to custom design and manufacture on demand their own physical products, according to their own needs. In the future, they will be able to fabricate interactive digital devices with user-specific form and functionality from the comfort of their homes. This thesis explores how task-specific, low computation, interactive devices capable of presenting dynamic visual information can be created using Printed Electronics technologies, whilst following an approach based on the ideals behind Personal Fabrication. Focus is given on the use of printed electrochromic displays as a medium for delivering dynamic digital information. According to the architecture of the displays, several approaches are highlighted and categorised. Furthermore, a pictorial computation model based on extended cellular automata principles is used to programme dynamic simulation models into matrix-based electrochromic displays. Envisaged applications include the modelling of physical, chemical, biological, and environmental phenomena.

A pictorial field guide for the rapid identification of sandflies (DIPTERA: PSYCHODIDAE), commonly encountered in Pará state, Brazil.

A pictorial field guide to the 30 species of sandfly most commonly encountered in Pará State is presented, based on the easily recognised external characters of the length of the 5th palpal segment, thoracic infuscation, abdominal colour and head and body size. In most cases this allows identification to the species. In others, especially with females, it gives an indication of the species, which may then be confirmed with data from more detailed taxanomix studies. This type of field guide helps in teaching, rapid sorting of flies prior to dissection and in acquainting visitors with the variety of species present in a given area.A rapid technique for the taxonomic sorting of unmounted, freshly killed female sandflies is required, prior to the dissection of large numbers of a particular species. Such a method is useful in areas where numerous species occur in studies on natural flagellate infections, age determination and for ecological studies. With the above points in mind a pictorial field guide has been designed that enables the identification of unmounted, unmacerated specimens of the 30 more commonly encountered species of phleboto-mine sandflies (***) in Pará State, North Brazil. It is based on the easily recognised external characters of the length of the 5th palpal segment, thoracic infuscation, ad-dominal colour and proboscis and body size.Taxonomy of male phlebotomine sandflies is based on the structure of the genitalia and, as most of this is external, a wholly external character key is readily made. Female taxonomy, however, is based on the internal character of the cibarium, pharynx and sperma thecae. In order to produce an external character key we therefore return to an unso phisticated "phlebotometry" (see Martins et al., 1978 p. 3 for review), using relative lengths of the proboscis, palpal segments and body, along with the degree of infuscation. Ihis idea is not new; indeed many sandfly specialists presently use external characters to separate certain species (H. Fraiha, R. P. Lane, P. D. Ready, D. G. Young and R. D. Ward personal communications 1983 & 1984).A key used to separate five anthropophillic sandflies by Biagi (1966), in Mexico, was based mainly on palpal segment length and infuscation. Floch and Abonnenc (1952) stressed the use of relative lengths of palpal segments in their keys to the sandflies of French Guiana, and four members of the shannoni group have been similarly separated according to the degree of infuscation by Morales et al. (1982). The use of thoracic infuscation as a reliable character seems to be gaining favour, having been used by young & Fairchild (1974) and Ready & Fraiha (1981). Indeed Chariotis 1974) showed the usefulness of thoracic infuscation to sepenate 7 anthropophillic species, during studies onvesicular stomatitis in Panama. Identification using external characters is essential for work on viral isolations from sandflies, where bulk samples of whole sandflies are used.Perhaps the major advantage of a simple visual guide is for teaching purposes. Technical staff in this lnstitute are able to identify most of the species they encounter without having to use the standard, more unwieldly (and in many cases unavailable) internal character keys, and the guides presented below have allowed rapid species sorting prior to the dissection of sandflies in our leismaniasis study areas (Ryan et at. ,1985).

Ontology-based representation and generation of workflows for micro-task human-machine computation

Doctoral Program in Computer Science

Detection of road accident accumulation zones with a visual analytics approach

Nowadays, road accidents are a major public health problem, which increase is forecasted if road safety is not treated properly, dying about 1.2 million people every year around the globe. In 2012, Portugal recorded 573 fatalities in road accidents, on site, revealing the largest decreasing of the European Union for 2011, along with Denmark. Beyond the impact caused by fatalities, it was calculated that the economic and social costs of road accidents weighted about 1.17% of the Portuguese gross domestic product in 2010. Visual Analytics allows the combination of data analysis techniques with interactive visualizations, which facilitates the process of knowledge discovery in sets of large and complex data, while the Geovisual Analytics facilitates the exploration of space-time data through maps with different variables and parameters that are under analysis. In Portugal, the identification of road accident accumulation zones, in this work named black spots, has been restricted to annual fixed windows. In this work, it is presented a dynamic approach based on Visual Analytics techniques that is able to identify the displacement of black spots on sliding windows of 12 months. Moreover, with the use of different parameterizations in the formula usually used to detect black spots, it is possible to identify zones that are almost becoming black spots. Through the proposed visualizations, the study and identification of countermeasures to this social and economic problem can gain new grounds and thus the decision- making process is supported and improved.

Identification and characterization of deforestation hot spots in Venezuela using MODIS satellite images

Land cover changes over time as a result of human activity. Nowadays deforestation may be considered one of the main environmental problems. The objective of this study was to identify and characterize changes to forest cover in Venezuela between 2005-2010. Two maps of deforestation hot spots were generated on the basis of MODIS data, one using digital techniques and the other by means of direct visual interpretation by experts. These maps were validated against Landsat ETM+ images. The accuracy of the map obtained digitally was estimated by means of a confusion matrix. The overall accuracy of the maps obtained digitally was 92.5%. Expert opinions regarding the hot spots permitted the causes of deforestation to be identified. The main processes of deforestation were concentrated to the north of the Orinoco River, where 8.63% of the country's forests are located. In this region, some places registered an average annual forest change rate of between 0.72% and 2.95%, above the forest change rate for the country as a whole (0.61%). The main causes of deforestation for the period evaluated were agricultural and livestock activities (47.9%), particularly family subsistence farming and extensive farming which were carried out in 94% of the identified areas.

Promoção da motivação e da autonomia com recurso às tecnologias de informação e comunicação em alunos normovisuais e em alunos com deficiência visual

Relatório de estágio de mestrado em Ensino de Informática

Quando a luta pelos direitos sexuais se faz no feminino: um estudo de caso sobre o associativismo lésbico em Portugal

Dissertação de mestrado em Sociologia (área de especialização em Desenvolvimento e Políticas Sociais)

Reliability of the interpretation of coronary angiography by the simple visual method

OBJECTIVE: Evaluation of inter and intraobserver reproducibility of by the visual method interpretation of cineangiogram in a clinically based context. METHODS: Five interventional cardiologists analyzed 11 segments of 8 coronary cineangiograms at a two month apart sessions. The percent luminal reduction by the lesions were analyzed by two different classifications: in one (A) the lesions were graded in 0% = absent, 1-50% = mild, 51 - 69 = moderate, and > or = 70% = severe; the other classification (B) was a dichotomic one : <70% = nonsignificant and > or = 70%=significant lesions. The agreement were measured by the kappa (k) index. RESULTS: Interobserver agreement was moderate for classification A (1st measurement, k = 0.36 -- 0.63, k m = 0.49; 2nd measurement, k = 0.39-0.68, k m = 0.52) and good for classification B (1st measurement, k = 0.55-0.73, k m = 0.63; 2nd measurement, k = 0.37-0.82, k m = 0.61). Intraobserver levels of agreement were k = 0.57-0.95 for classification A and 0.62-1.0 for classification B. CONCLUSION: The higher level of reproducibility obtained by adopting the dichotomous criteria usually considered for ischemic limits demonstrates that in the present clinical context, the reliability of the simple visual method is adequate for the identification of patients with clinically significant lesions and candidates for myocardial revascularization procedures.