Event Detection

The management of the vast amount of media assets captured at every day events such as meetings, birthday parties, vacation, and conferences has become an increasingly challenging problem. Today, most media management applications are media-centric. This means, they put the captured media assets into the center of the management. However, in recent years it has been proposed that events are a much better abstraction of human experience and thus provide a more appropriate means for managing media assets. Consequently, approaches that include events into their media management solution have been explored. However, they typically consider events only as some more metadata that can be extracted from the media assets. In addition, today's applications and approaches concentrate on particular problems such as event detection, tagging, sharing, classification, or clustering and are often focused on a single media type. In this paper, we argue for the benefits of an event-centric media management (EMMa) approach that looks at the problem of media management holistically. Based on a generic event model, we specify a media event model for the EMMa approach. The single phases and processes of the EMMa approach are defined in a general process chain for an event-centric media management, the EMMa cycle. This cycle follows the event concept throughout all phases and processes of the chain and puts the concept of events in the center of the media management. Based on the media event model and EMMa cycle, we design a component-based architecture for the EMMa approach and conduct an implementation of the approach.

The management of the vast amount of media assets captured at every day events such as meetings, birthday parties, vacation, and conferences has become an increasingly challenging problem. Today, most media management applications are media-centric. This means, they put the captured media assets into the center of the management. However, in recent years it has been proposed that events are a much better abstraction of human experience and thus provide a more appropriate means for managing media assets. Consequently, approaches that include events into their media management solution have been explored. However, they typically consider events only as some more metadata that can be extracted from the media assets. In addition, today's applications and approaches concentrate on particular problems such as event detection, tagging, sharing, classification, or clustering and are often focused on a single media type. In this paper, we argue for the benefits of an event-centric media management (EMMa) approach that looks at the problem of media management holistically. Based on a generic event model, we specify a media event model for the EMMa approach. The single phases and processes of the EMMa approach are defined in a general process chain for an event-centric media management, the EMMa cycle. This cycle follows the event concept throughout all phases and processes of the chain and puts the concept of events in the center of the media management. Based on the media event model and EMMa cycle, we design a component-based architecture for the EMMa approach and conduct an implementation of the approach.

Objective: The aim of this work is to assess how adding a driving-related task affects the detection of objects in peripheral vision, under mesopic conditions. Background: The main index used to assess the quality of road lighting installations refers to simple detection tasks in foveal vision, which raises methodological and practical questions. Method: The experimental design consisted of a three-phase experiment. In the first phase, two groups (control and experimental) performed a peripheral detection task (simple task). Based on these results an individual detection threshold was computed for each participant and eccentricity. A tracking task was performed in Phase 2 for both groups (steering a tracking target along a circuit, on a screen). In the third phase, the control group performed the same task as in Phase 2. The experimental group performed a double task, with a tracking (primary) task and a peripheral detection (secondary) task. Results: The data show an effect of the ...

Le traitement d'image est un outil de plus en plus utilisé dans l'étude des avalanches de neige, ceci dans le but de les prévenir. On présente ici un algorithme de contour actif spécifique à notre application : l'analyse du front des avalanches. Le but de cette étude consiste en l'extraction de paramètres dynamiques de l'avalanche à partir d'un simple film vidéo. La principale difficulté réside dans le manque évident de contraste de nos images, puisque nous avons à poursuivre un objet blanc sur un fond blanc. L'algorithme présenté repose sur le principe des contours actifs, où on recherche à minimiser une forme énergétique modélisant le contour de l'avalanche. Le résultat obtenu est une séquence de contour d'avalanche.

Excessive amounts of image spam cause many problems to e-mail users. Since image spam is difficult to detect using conventional text-based spam approach, various image processing techniques have been proposed. In this paper, we present an ensemble method using frequent itemset mining (FIM) for filtering image spam. Despite the fact that FIM techniques are well established in data mining, it is not commonly used in the ensemble method. In order to obtain a good filtering performance, a SIFT descriptor is used since it is widely known as effective image descriptors. K-mean clustering is applied to the SIFT keypoints which produce a visual codebook. The bag-of-word (BOW) feature vectors for each image is generated using a hard bag-of-features (HBOF) approach. FIM descriptors are obtained from the frequent itemsets of the BOW feature vectors. We combine BOW, FIM with another three different feature selections, namely Information Gain (IG), Symmetrical Uncertainty (SU) and Chi Square (CS...

Welcome to ELS 2011, the 4 th European Lisp Symposium. In the recent years, all major academic events have suffered from a decreasing level of attendance and contribution, Lisp being no exception to the rule. Organizing ELS 2011 in this context was hence a challenge of its own, and I'm particularly happy that we have succeeded again. For the first time this year, we had a "special focus" on parallelism and efficiency, all very important matters for our community with the advent of multi-core programming. It appears that this calling has been largely heard, as half of the submissions were along these lines. Another notable aspect of this year's occurrence is the fact that four dialects of Lisp are represented: Common Lisp, Scheme, Racket and Clojure. This indicates that ELS is successful in attempting to gather all crowds around Lisp "the idea" rather than around Lisp "one particular language". The European Lisp Symposium is also more European than ever, and in fact, more international than ever, with people coming not only from western Europe and the U.S.A., but also from such countries as Croatia and Bulgaria. While attending the symposium is just seeing the tip of the iceberg, a lot have happened underwater. First of all, ELS 2011 would not have been possible without the submissions we got from the authors and the careful reviews provided by the programme committee members; I wish to thank them for that. I am also indebted to the keynote speakers who have agreed to come and spread the good word. I wish to express my utmost gratitude to our sponsors who contributed to making the event quite affordable this year again. Ralf Möller was our local chair, the "Grey Eminence" of the symposium, and we owe him a lot. Finally, my thanks go to Edgar Gonçalves for taking care of the website with such reactivity and attentiveness. I wish you all a great symposium! Welcome to Hamburg University of Technology (TUHH). We hope you will enjoy your stay at our university for the 2011 European Lisp Symposium. Not only interesting presentations will be part of the programme, but also social events such as the the social dinner at the Feuerschiff () and the Welcome Reception at Freiheit (). We would like to thank all sponsors for making the event possible. For those of you staying over the weekend, a tour to Miniatur-Wunderland (. miniatur-wunderland.de) will be offered. ELS 2011 iii TUHH () is a competitive entrepreneurial university focussing on highlevel performance and high quality standards. TUHH is dedicated to the principles of Humboldt (unity of research and education). TUHH has a strong international orientation and also focusses on its local environment. It contributes to the development of the technological and scientific competence of society, aiming at excellency at the national and international level in its strategic research fields, and educating young scientists and engineers within demanding programmes using advanced teaching methods. Let's not forget Hamburg, for why are we all here? People say Hamburg is Germany's most attractive city combining urbane sophistication, maritime flair, the open-mindedness of a metropolis, and high recreational value. The second-largest city in Germany, it has traditionally been seen as a gateway to the world. Its port is not only the largest seaport in Germany and the second-largest in Europe, but a residential neighborhood with leisure, recreational and educational facilities. Hamburg is a very special place.

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

The COVID-19 outbreak prompts the need for new ways to detect and prevent epidemics. Since cough is one of the COVID-19 symptoms, our work proposes a sound recognition system based on our previous works which are able to detect different type of cough through a continuously home sound analysis. It is a part of EpiSemioWatch AMI COVID UTC funded project which aims to develop a detection system of epidemic risk in retirement home using sound and movement sensors. This paper presents the sound part of the global system and more explicitly the differentiation of cough from other environmental sound. The proposed system detects automatically the sound activity through a wavelet based system, realize an acoustic segmentation speech/sound and classify the cough sounds. A first evaluation of the system in our lab is presented.

The study of protein dynamics is essential for understanding the multi-molecular complexes at subcellular levels. Fluorescent Protein (XFP)-tagging and time-lapse fluorescence microscopy enable to observe molecular dynamics and interactions in live cells, unraveling the live states of the matter. Original image analysis methods are then required to process challenging 2D or 3D image sequences. Recently, tracking methods that estimate the whole trajectories of moving objects have been successfully developed. In this paper, we address rather the detection of meaningful events in spatio-temporal fluorescence image sequences, such as apparent stable "stocking areas" involved in membrane transport. We propose an original patch-based Markov modeling to detect spatial irregularities in fluorescence images with low false alarm rates. This approach has been developed for real image sequences of cells expressing XFP-tagged Rab proteins, known to regulate membrane trafficking.

Social Force Model (SFM) is commonly used in crowd analysis. In this paper, modified SFM is proposed to detect and localize the abnormality in crowd scene. This task is done by estimating the interaction forces in image frames based on SFM theory. The algorithm is jointly used with optical flow, which provides flow vector to be used in particle advection. The moving particles are treated as a main cue instead of particle tracking. Some modifications of SFM algorithm has been proposed here in order to capture the particles which carry significant information of the crowd. The interaction forces are being selected based on Fisher’s equation. The computed interaction forces determine the synergy between the advected particles, whereby high magnitude of interaction force has high possibility of abnormal behaviour happened.

Streaming media provides a number of unique challenges for computational linguistics. This paper studies the temporal variation in word co-occurrence statistics, with application to event detection. We develop a spectral clustering approach to find groups of mutually informative terms occurring in discrete time frames. Experiments on large datasets of tweets show that these groups identify key real world events as they occur in time, despite no explicit supervision. The performance of our method rivals state-of-the-art methods for event detection on F-score, obtaining higher recall at the expense of precision.

Sur base d'un des thèmes de l'appel à projets Fonds Européen de la Défense 2023 qui couvrait l'automatisation, avec de l'Inte lligence Artificielle, des Tests d'Intrusion militaires, nous avons réalisé un état de l'art sur les associations IHM-Cybersécurité et IHM-Agent Cognitif. Nous avons étudié HCISec qui améliore l'utilisabilité des dispositifs de Cybersécurité. Nous avons effectué l'Analyse Socio-Technique d'un Agent Cognitif destiné à automatiser ces Tests d'Intrusion militaires . Nous avons énoncé un processus de Conception Centré Utilisateur adéquat et l'avons inséré dans un Cycle de Vie de Développement Logiciel AgileUX. Nous avons ensuite étudié les Interactions Humain-Machine entre l'acteur humain et cet Agent Cognitif sur base d'un Cas d'Utilisation. Nous avons conçu l'architecture logiciel le basée sur l'Architecture Cognitive Soar et un Module de Contrôle de tests d'intrusion exposant des Interfaces Humain-Machine. Nous avons enfin décrit ces interfaces consistant en 3 écrans principaux et la navigation entre ces écrans. CCS CONCEPTS • Security and privacy→Intrusion/anomaly detection and malware mitigation • Human-centered computing → Human computer interaction (HCI)• Computing methodologies→Artificial intelligence→Distributed artificial intelligence Based on a theme of the 2023 European Defence Fund (EDF) addressing the automation, with Artificial Intelligence, of military Penetration Testing, we have carried a state-of-the-art upon the HMI-Cybersecurity et  HMI-Cognitive Agent associations. We have studied HCISec which aims at improving the usability of Cybersecurity mechanisms and tools. We have performed the Socio-Technical Analysis of a Cognitive Agent aimed at automating military Pen Tests. We have defined a User Centered Design process and embedded it within an AgileUx Software Development Life Cycle. We then studied the Human-Machine Interactions between the Human Actor and this Cognitive Agent based on a Use Case. We have then designed the software architecture based on the Soar Cognitive Architecture and a Pen Tests control module exposing Human-Machine Interfaces. At last, we have described those interfaces as 3 screen layouts and the navigation flow between those screens.

Foreground segmentation in video sequences is an important area of the image processing that attracts great interest among the scientist community, since it makes possible the detection of the objects that appear in the sequences under analysis, and allows us to achieve a correct performance of high level applications which use foreground segmentation as an initial step. The current Ph.D. thesis entitled Parametric Region-Based Foreground Segmentation in Planar and Multi-View Sequences details, in the following pages, the research work carried out within this eld. In this investigation, we propose to use parametric probabilistic models at pixel-wise and region level in order to model the di erent classes that are involved in the classi cation process of the di erent regions of the image: foreground, background and, in some sequences, shadow. The development is presented in the following chapters as a generalization of the techniques proposed for objects segmentation in 2D planar seq...

Je tiens à remercier Monsieur Pierre BAZEX, Professeur à l'Université de Toulouse III, pour l'honneur qu'il me fait en participant à mon jury de thèse. Ma gratitude va également à Monsieur Bernard CAUSSE, Professeur au département Informatique de l'Institut Universitaire de Technologie de Bayonne, pour l'intérêt qu'il a porté à mes travaux et pour avoir accepté d'être président de mon jury de thèse. J'adresse mes sincères remerciements à Monsieur Guy GOUARDERES, Professeur au département

Data from the Pierre Auger Observatory are analyzed to search for anisotropies near the direction of the Galactic Centre at EeV energies. The exposure of the surface array in this part of the sky is already significantly larger than that of the fore-runner experiments. Our results do not support previous findings of localized excesses in the AGASA and SUGAR data. We set an upper bound on a pointlike flux of cosmic rays arriving from the Galactic Centre which excludes several scenarios predicting sources of EeV neutrons from Sagittarius A. Also the events detected simultaneously by the surface and fluorescence detectors (the 'hybrid' data set), which have better pointing accuracy but are less numerous than those of the surface array alone, do not show any significant localized excess from this direction.

An increasing number of road accidents, a major global issue, are linked to growing demands for faster vehicle speeds. Analytical research highlights the significance of selfvisual distraction events for drivers in these occurrences. In this study, we present an IoT-Fog-based framework that is intended to automatically identify and notify drivers of selfvisual distraction as it occurs. With a conditional evaluation dependent on the status of the driver, the proposed framework uses Convolutional Neural Networks (CNNs) to extract characteristics and expressions indicative of a distracted state of the driver. The framework shows encouraging results, proving the effectiveness of the model with a high accuracy rate of 90% and precision value of 91%. Notably, in terms of detection accuracy, the suggested framework outperforms other state-of-the-art techniques. Real-time alerts that involve prompt recognition and persistent beeping until normality is restored can reduce traffic accidents and are an essential intervention for saving lives.

In this work, we present an event detection method in Twitter based on clustering of hashtags and introduce an enhancement technique by using the semantic similarities between the hashtags. To this aim, we devised two methods for tweet vector generation and evaluated their effect on clustering and event detection performance in comparison to word-based vector generation methods. By analyzing the contexts of hashtags and their co-occurrence statistics with other words, we identify their paradigmatic relationships and similarities. We make use of this information while applying a lexico-semantic expansion on tweet contents before clustering the tweets based on their similarities. Our aim is to tolerate spelling errors and capture statements which actually refer to the same concepts. We evaluate our enhancement solution on a three-day dataset of tweets with Turkish content. In our evaluations, we observe clearer clusters, improvements in accuracy, and earlier event detection times.

This is the first time that our team participate TRECVID. This paper summarizes our approach submitted to Semantic Indexing (SIN) task in TRECVID 2011. Our approach adopts bag-of-features method to transform original visual and audio features into histogram features, using pre-trained codebook. After feature transformation, one-versus-others SVMs with Chi-square kernel are trained. In decision step, averaged probability is calculated as a final score to rank shots. Under this framework, we tested 4 visual features including dense grid SIFT, color SIFT, OLBPC and DAISY together with 1 audio feature consisting of MFCC with delta and acceleration. Our audio visual combination model achieves best results in terms of mean xinfAP. Besides, considering the huge amount of data this year, we employed several speedup strategies such as k-means clustering with GPU acceleration and homogeneous kernel map. All these efforts rank us at the 12 th out of 19 teams in full run and the 13 th out of 27 teams in the light run test.

Two paleosol sequences of the Late (South France) and Middle Pleistocene (South China) were investigated using a high resolution approach associated with SEM-EDS and XRD analysis on selected grains. This approach has enabled us to distinguish the ...

The tectonic stress eld was investigated in and around the aftershock area of the Hokkaido Eastern Iburi earthquake (M JMA = 6.7) occurred on 6 September 2018. We deployed 26 temporary seismic stations in the aftershock area for approximately 2 months and located 1785 aftershocks precisely. Among these aftershocks 818 focal mechanism solutions were determined using the rst motion polarity of P wave from the temporary observation and the permanent seismic networks of Hokkaido University, Japan Meteorological Agency (JMA), and High Sensitivity Seismograph Network Japan (Hi-net). We found that (1) the reverse faulting and the strike-slip faulting are dominant in the aftershock area, (2) the average azimuths of P-and T-axes are N78° ± 33°E and N3° ± 52°W, respectively, and (3) the average dips of Pand T-axes are 25° ± 16° and 46° ± 20°, respectively: the P-axis is close to be horizontal and the T-axis is close to be vertical. We applied a stress inversion method to the focal mechanism solutions to estimate a stress eld in the aftershock area. As a result, we found that the reverse fault type stress eld is dominant in the aftershock area. An axis of the maximum principal stress (σ 1 ) has the azimuth of N73° ± 8°E and the dipping eastward of 17° ± 6° and an axis of the medium principal stress (σ 2 ) has the azimuth of N126° ± 91°E and the dipping southward of 16° ± 13°, indicating that both of σ 1 -and σ 2 -axes are close to be horizontal. An axis of the minimum principal stress (σ 3 ) has the dipping westward of 64° ± 9° that is close to be vertical. The results strongly suggest that the reverse-fault-type stress eld is predominant as an average over the aftershock area which is in the western boundary of the Hidaka Collision Zone. Although the average of the stress ratio is R = 0.6 ± 0.2 in the whole aftershock area, R decreases systematically as the depth is getting deep, which is modeled by a quadratic polynomial of depth.

There has been an almost explosive growth in digital video in recent years. The convention for enabling users to navigate digital video is the Video Cassette Recorder-like (VCR-like) control set, which is dictated by the proliferation of media players that embody it, including Windows Media Player and QuickTime. However, there is a dearth of research seeking to understand how users relate to this control set and how useful it actually is in practice. This paper details our empirical investigation of the issue. A digital video navigation system with a VCR-like control set was developed and subsequently used by a large sample of users (n ¼ 200), who were required to complete a number of goal-directed navigational tasks. Each user's navigational activity was tracked and recorded automatically by the system. Analysis of the navigational data revealed a range of results concerning how the VCR-like control set both enhanced and limited the user's ability to locate sequences of interest, including a number of searching and browsing strategies that were exploited by the users.

As sessile organisms, plants must cope with multiple and combined variations of signals in their environment. However, very few reports have studied the genome-wide effects of systematic signal combinations on gene expression. Here, we evaluate a high level of signal integration, by modeling genome-wide expression patterns under a factorial combination of carbon (C), light (L), and nitrogen (N) as binary factors in two organs (O), roots and leaves. Signal management is different between C, N, and L and in shoots and roots. For example, L is the major factor controlling gene expression in leaves. However, in roots there is no obvious prominent signal, and signal interaction is stronger. The major signal interaction events detected genome wide in Arabidopsis roots are deciphered and summarized in a comprehensive conceptual model. Surprisingly, global analysis of gene expression in response to C, N, L, and O revealed that the number of genes controlled by a signal is proportional to the magnitude of the gene expression changes elicited by the signal. These results uncovered a strong constraining structure in plant cell signaling pathways, which prompted us to propose the existence of a ''code'' of signal integration.

A mobile platform mounted with omnidirectional vision sensor (ODVS) can be used to monitor large areas and detect interesting events such as independently moving persons and vehicles. To avoid false alarms due to extraneous features, the image motion induced by the moving platform should be compensated. This paper describes a formulation and application of parametric egomotion compensation for an ODVS. Omni images give 360 • view of surroundings but undergo considerable image distortion. To account for these distortions, the parametric planar motion model is integrated with the transformations into omni image space. Prior knowledge of approximate camera calibration and camera speed is integrated with the estimation process using a Bayesian approach. Iterative, coarse-to-fine, gradient-based estimation is used to correct the motion parameters for vibrations and other inaccuracies in prior knowledge. Experiments with a camera mounted on various types of mobile platforms demonstrate successful detection of moving persons and vehicles.

This paper presents a survey of trajectory-based activity analysis for visual surveillance. It describes techniques that use trajectory data to define a general set of activities that are applicable to a wide range of scenes and environments. Events of interest are detected by building a generic topographical scene description from underlying motion structure as observed over time. The scene topology is automatically learned and is distinguished by points of interest and motion characterized by activity paths. The methods we review are intended for real-time surveillance through definition of a diverse set of events for further analysis triggering, including virtual fencing, speed profiling, behavior classification, anomaly detection, and object interaction.

This paper presents a seismic signal analysis framework on data sensed by a wireless sensor network deployed on Cotopaxi volcano. The results obtained by measuring environmental signals in the volcano and applying wavelet analysis, it is possible to find a signal classification to interpret the behaviour of the volcano. Sixteen sensors have been used and deployed in a strategy area, the information was obtained during three days of continuous monitoring. This information was sent to a surveillance laboratory located 45 km away from the station placed in the volcano, a WiFi-based long distance technology was used to this purpose. Volcanic information was processed using wavelet transform, a spectral pattern of seismic events determined four kinds of events, corresponding to a volcanic tremor, hybrid seisms, long period seisms and tectonic seisms.

In this paper is presented an approach in development of measurement system (MS) for analyzing the power quality (PQ). The main aim of this paper is to present procedures of PQ measurements according to the strictest norms and standards and to develop a system for measuring specific indicators of PQ. The developed MS is implemented on several examples. In this paper is presented an example of implementing the MS in analyzing the loads mode of an apartment with appliances of new generation (energyefficient appliances). In the paper is also explored the problem of implementation of universal acquisition cards for their application in analyzing PQ. The main focus of conclusions in this paper is comparison of present and future systems for monitoring the PQ.

The ability to correctly classify sentences that describe events is an important task for many natural language applications such as Question Answering (QA) and Summarisation. In this paper, we treat event detection as a sentence level text classification problem. We compare the performance of two approaches to this task: a Support Vector Machine (SVM) classifier and a Language Modeling (LM) approach. We also investigate a rule based method that uses hand crafted lists of terms derived from WordNet. These terms are strongly associated with a given event type, and can be used to identify sentences describing instances of that type. We use two datasets in our experiments, and evaluate each technique on six distinct event types. Our results indicate that the SVM consistently outperform the LM technique for this task. More interestingly, we discover that the manual rule based classification system is a very powerful baseline that outperforms the SVM on three of the six event types.

This paper presents new approach for time series data classification using Fuzzy Expert System (FES). In the proposed study, the power disturbance signals are considered as time series data for testing the designed FES. Initially the time series data are pre-processed through the advanced signal processing tool such as S-transform and various statistical features are extracted, which are used as inputs to the FES. The FES output is optimized i1sing Particle Swann Optimization (PSO) to bring the output to distinct classification level. Both Gaussian and trapezoidal membership functions are selected for designing the proposed FES arid the performance measure is derived by comparing the classification rates for the time series data without noise and with noise up to SNR 20 db. The proposed algorithm provides accurate classification rates even under noisy conditions compared to the existing techniques, which shows the efficacy and robustness of the proposed algorithm for time series data classification.

This paper explores the real-time summarization of scheduled events such as soccer games from torrential flows of Twitter streams. We propose and evaluate an approach that substantially shrinks the stream of tweets in real-time, and consists of two steps: (i) sub-event detection, which determines if something new has occurred, and (ii) tweet selection, which picks a representative tweet to describe each sub-event. We compare the summaries generated in three languages for all the soccer games in Copa America 2011 to reference live reports offered by Yahoo! Sports journalists. We show that simple text analysis methods which do not involve external knowledge lead to summaries that cover 84% of the sub-events on average, and 100% of key types of sub-events (such as goals in soccer). Our approach should be straightforwardly applicable to other kinds of scheduled events such as other sports, award ceremonies, keynote talks, TV shows, etc.

Many studies over the past two decades have shown that people can use brain signals to convey their intent to a computer through brain-computer interfaces (BCIs). These devices operate by recording signals from the brain and translating these signals into device commands. They can be used by people who are severely paralyzed to communicate without any use of muscle activity. One of the major impediments in translating this novel technology into clinical applications is the current requirement for preliminary analyses to identify the brain signal features best suited for communication. This paper introduces and validates signal detection, which does not require such analysis procedures, as a new concept in BCI signal processing. This detection concept is realized with Gaussian mixture models (GMMs) that are used to model resting brain activity so that any change in relevant brain signals can be detected. It is implemented in a package called SIGFRIED (SIGnal modeling For Real-time Identification and Event Detection). The results indicate that SIGFRIED produces results that are within the range of those achieved using a common analysis strategy that requires preliminary identification of signal features. They indicate that such laborious analysis procedures could be replaced by merely recording brain signals during rest. In summary, this paper demonstrates how SIGFRIED could be used to overcome one of the present impediments to translation of laboratory BCI demonstrations into clinically practical applications.

Positron emission tomography (PET) systems employ mixed-signal front-end to carry out relatively simple, and ad hoc, processing of the charge pulses generated upon event detection. To obtain, and maintain over time, proper calibrations of the mixed-signal circuitry for generating accurate event information is a challenging task due to the simplicity of the event processing, and the huge number of channels and multiplexing of the input signals found in modern PET systems. It is also difficult to modify or extend the event-processing technologies when needs arise because it would involve making changes to the circuitry. These limitations can be circumvented by applying digital signal-processing technologies for analyzing event pulses generated in PET. With digital technologies, optimized event-processing algorithms can be implemented and they can be modified or extended with ease when needed. The resulting PET data-acquisition (DAQ) system is easier to calibrate and maintain, can generate more accurate event information, and has better extendibility. In this paper, we present our work toward developing a scalable all-digital DAQ system for PET, built upon a personal-computer platform for reducing cost. We will present the overall architecture of this digital DAQ system, and describe our implementations of several components of the system.

We report on the detection of an extreme flarelike event on DF Tauri during a spectroscopic monitoring program of some classical T Tauri stars known to show hot spots on their surfaces. These observations were performed on a nightly basis within roughly one rotational period of DF Tau between 2000 January 7 and 14. The estimated amplitude of variation during this time reached up to 6 mag in the B band. Significant inverse P Cygni profiles in high Balmer lines-evidence of mass accretion-dominated the spectrum obtained on January 8, when the brightness of DF Tau was approaching its maximum. The flux in the ultraviolet dropped dramatically on the following day. Ca ii H and K emission lines, the indicator of chromospheric activity, turned into deep, wide absorption, and all other emission lines in the blue end were also absent (filled in or weak absorption), suggesting severe absorption in the line of sight possibly caused by a transient envelope produced in the process of a violent accretion shock. The effect of veiling decreased rapidly from January 10, and normal P Cygni profiles, indicating the presence of strong wind, began to appear. Based on the unique features displayed, we propose that the spectroscopic record of this flarelike brightening on DF Tau, believed to be due to episodic mass accretion accompanied by the release of gravitational energy, might actually be taken as an isolated, snapshot view of the YY Orionis phenomenon.

Soccer is a dynamic sport where tracking the ball is crucial for players, coaches, and analysts. This research paper explores trajectory-based methods for soccer ball detection and tracking in video analysis. Using advanced computer vision techniques, we aim to improve the accuracy and efficiency of ball tracking in various game scenarios. We will examine current approaches, discuss the challenges, and propose a trajectory-based system that enhances performance in real-world applications.

A main requirement in clinical gait analysis is the ability to accurately identify gait events; especially, the initial contact of the heel with the floor and the toe off. The knowledge of the major events of the gait cycle is needed, for instance, in biomechanical data normalization ...

This thesis presents a solver to handle constrained-state-space ODEs. This solver locates the points where any of the states go outside of the constraining set, and then, transfers the control from the continuous time ODE to the function governing the behavior of the system on the boundaries of the constrained set. The main contribution of this solver is found simulating complex right-hand-sided ODEs for long periods of time. † December 2002. iii Runge-Kutta algorithms along with the most well-known implementations by Feldberg and Dormand-Prince. Later, we discuss continuous extensions for these algorithms and root-finding schemes. Finally, we present a novel approach for step size control, specially suited for Zeno-effect-exhibiting IVPs. Chapter 4 shows the outline of our solvers, and then presents some results obtained from applying the solvers to different problems. Finally, Chapter 5 presents the conclusions of this research and an outline for future work on the topic.

One of the objectives of the Field Exercise 2009 in Finland was to test the recently updated software for the Seismic Aftershock Monitoring System (SAMS). A tunnel structure being build-up by a series of explosions in depths ranging 300-400 m during the period of the exercise served as simulated inspection target; matching the conditions of one possible scenario during a

La reconnaissance de caractères ou de symboles s'appuie actuellement sur de nombreux descripteurs statistiques ou géométriques. Cependant, l'accroissement du nombre d'objets différents à traiter pose le problème de l'interaction avec l'utilisateur et la capacité de celui-ci à caractériser cette complexité. La représentation structurelle est une alternative intéressante car plus proche de l'appréhension visuelle humaine. Elle offre une approche plus intuitive de la reconnaissance permettant d'aborder des caractères difficiles à discriminer par des approches classiques. Nous nous intéressons donc, dans cet article, aux systèmes de reconnaissance de la structure des caractères et des symboles. Le travail étend la proposition de WONG ET YOU d'une reconnaissance basée sur une modélisation par graphe aléatoire à variables discrètes. L'utilisation d'attributs vectoriels plus complets impose le passage aux graphes aléatoires à variables continues.

La reconnaissance de caractères ou de symboles s'appuie actuellement sur de nombreux descripteurs statistiques ou géométriques. Cependant, l'accroissement du nombre d'objets différents à traiter pose le problème de l'interaction avec l'utilisateur et la capacité de celui-ci à caractériser cette complexité. La représentation structurelle est une alternative intéressante car plus proche de l'appréhension visuelle humaine. Elle offre une approche plus intuitive de la reconnaissance permettant d'aborder des caractères difficiles à discriminer par des approches classiques. Nous nous intéressons donc, dans cet article, aux systèmes de reconnaissance de la structure des caractères et des symboles. Le travail étend la proposition de WONG ET YOU d'une reconnaissance basée sur une modélisation par graphe aléatoire à variables discrètes. L'utilisation d'attributs vectoriels plus complets impose le passage aux graphes aléatoires à variables continues.

Cette connaissance a été matérialisée par l'administration d'un questionnaire aux informaticiens de l'Université Pédagogique Nationale sélectionnés tout en gardant l'esprit de sondage aléatoire simple à probabilité égale et d'en interpréter le résultat afin d'en tirer les conséquences qui en découlent, se basant sur les théories exactes existantes en la matière, tout en donnant des recommandations visant a améliorer les compétences en Programmation Orienté Aspect.

Given the deluge of multimedia content that is becoming available over the Internet, it is increasingly important to be able to effectively examine and organize these large stores of information in ways that go beyond browsing or collaborative filtering. In this paper we review previous work on audio and video processing, and define the task of Topic-Oriented Multimedia Summarization (TOMS) using Natural Language Generation: given a set of automatically extracted features from a video (such as visual concepts, ASR transcripts, etc.) a TOMS system will automatically generate a paragraph of natural language ("a recounting"), which summarizes the important information in a video belonging to a certain topic area, and provides explanations for why a video was matched, retrieved, etc. We see this as a first step towards systems that will be able todiscriminate visually similar, but semantically different videos, compare two videos and provide textual output or summarize a large number of videos at once. In this paper, we introduce our approach of solving the TOMS problem. We extract Visual Concept features and ASR transcription features from a given video, and develop a Template-Based Natural Language Generation (NLG) System to produce a textual recounting based on the extracted features. We also propose possible experimental designs for continuously evaluating and improving TOMS systems, and present results of a pilot evaluation of our initial system.