This paper deal with the problem of automatic detection of epileptic events in EEGs from depth electrodes using multiresolution wavelet analysis. The basic problems in events detection are considered: the time localization and characterization of epileptiform events, and the computational efficiency. The algorithm presented is based on a polynomial spline wavelet transform. The multiresolution representation obtained from this wavelet transform and the digital filters derived allow us an automatic detection, efficient and fast, of epileptiform activity. The detector proposed is based on the multiresolution energy function. This paper shows that it is possible to use a multiresolution wavelet scheme for detecting events in a nonstationary signal. EEG records from depth electrodes were analysed and the results obtained are shown.

Pouvoir lutter efficacement contre certains fléaux comme les incendies de forêt, les feux de brousse ou les catastrophes naturelles constitue un enjeu majeur dans plusieurs villes du monde.Avec l'avènement de la technologie de pointe représentée par les réseaux de capteurs, la détection de ces phénomènes devient plus aisée.En effet, des capteurs peuvent être déployés dans des zones difficiles d'accès et s'ils sont suffisamment nombreux pour couvrir la totalité de l'environnement à surveiller, une alerte peut être directement donnée par le capteur ayant détecté un certain type d'évènement (feu, secousse sismique...).Le centre de contrôle ayant reçu l'alerte peut ensuite décider d'intervenir sur la zone en cause.Nos travaux se situent dans ce cadre de la détection de phénomènes par un réseau de capteurs, en supposant que l'environnement est connu et que les capteurs sont mobiles, sans fil et en nombre insuffisant pour couvrir la totalité de l'environnement à surveiller.Parler de surveillance par un nombre faible d'entités mobiles nécessite de parcourir régulièrement certaines zones critiques de l'environnement, ce qui peut s'apparenter à une tâche de patrouille.Dans le cadre de cette thèse, nous nous sommes focalisés sur la détermination de stratégies de patrouille multi-capteurs appliquée à la détection d'évènements.Un problème similaire au nôtre est celui de la patrouille multi-agents dans un environnement connu.Ce problème consiste à faire visiter régulièrement les noeuds d'un graphe (représentant l'environnement) par des agents.Les capteurs peuvent être considérés comme des agents ayant des ressources limitées, en terme d'énergie en particulier.Le cadre de la patrouille multi-agents et les techniques proposées pour le résoudre ne peuvent pas être utilisés ici.Après avoir formulé mathématiquement le problème de la patrouille multi-capteurs appliquée à la détection d'évènements, nous proposons une technique de résolution approchée basée sur des colonies de fourmis.Des simulations ont été réalisées en considérant différents scenarii (topologies d'environnement, populations de capteurs, apparitions des événements) afin d'évaluer la pertinence de notre approche.Les résultats expérimentaux montrent que notre approche permet de déterminer des stratégies de patrouille satisfaisantes dans la majorité des scenarii. / To fight effectively against scourges like forest fires , brush fires or natural disasters is a major issue in many cities worldwide.With the advent of technology represented by sensor networks , detection of these phenomena becomes easier .Indeed , sensors can be deployed in remote areas and they are enough to cover the entire environment to monitor, an alert can be given directly by the sensor has detected a certain type of event (fire, earthquake ... ) .The control center has received the alert may then decide to intervene in the area in question .Our work takes place in the context of the detection of phenomena by a sensor network , assuming that the environment is known and that the sensors are mobile, wireless and insufficient to cover the entire environment to be monitored.Speaking of monitoring a small number of mobile entities requires regularly browse some critical environmental areas, which can be likened to a patrol task .In this thesis , we focused on identifying strategies patrol multi-sensor applied to the detection of events.A similar problem to ours is the multi-agent patrolling in a known environment .This problem is to regularly visit the nodes of a graph (representing the environment) by agents.The sensors can be considered as agents with limited resources , in terms of energy in particular.The framework of multi- agents and techniques proposed to solve patrol can not be used here .After mathematically formulated the problem of multi-sensor patrol applied to the detection of events, we propose an approximate solution technique based on ant colonies .Simulations were made ​​considering different scenarios ( environmental topologies populations sensors appearances events ) to assess the relevance of our approach.The experimental results show that our approach identifies strategies patrol satisfactory in the majority of scenarios.

新聞事件可描述為「一個時間區間內、同一主題的相似新聞之集合」，而新聞大多僅是一完整事件的零碎片段，其內容也易受到媒體立場或撰寫角度不同有所差異；除此之外，龐大的新聞量亦使得想要瞭解事件全貌的困難度大增。因此，本研究將利用文字探勘技術群聚相關新聞為事件，以增進新聞所帶來的價值。 分類分群為文字探勘中很常見的步驟，亦是本研究將新聞群聚成事件所運用到的主要方法。最近鄰 (k-nearest neighbor, kNN)搜尋法可視為分類法中最常見的演算法之一，但由於kNN在分類上必須要每篇新聞兩兩比較並排序才得以選出最近鄰，這也產生了kNN在實作上的效能瓶頸。本研究提出了一個「建立距離參考基準點」的方法RTD-based kNN (Relative Text-Distance-based kNN)，透過在向量空間中建立一個基準點，讓所有文件利用與基準點的相對距離建立起遠近的關係，使得在選取前k個最近鄰之前，直接以相對關係篩選出較可能的候選文件，進而選出前k個最近鄰，透過相對距離的概念減少比較次數以改善效率。 本研究於Google News中抽取62個事件(共742篇新聞)，並依其分群結果作為測試與評估依據，以比較RTD-based kNN與kNN新聞事件分群時的績效。實驗結果呈現出RTD-based kNN的基準點以常用字字彙建立較佳，分群後的再合併則有助於改善結果，而在RTD-based kNN與kNN的F-measure並無顯著差距(α=0.05)的情況下，RTD-based kNN的運算時間低於kNN達28.13%。顯示RTD-based kNN能提供新聞事件分群時一個更好的方法。最後，本研究提供一些未來研究之方向。 / News Events can be described as "the aggregation of many similar news that describe the particular incident within a specific timeframe". Most of news article portraits only a part of a passage, and many of the content are bias because of different media standpoint or different viewpoint of reporters; in addition, the massive news source increases complexity of the incident. Therefore, this research paper employs Text Mining Technique to cluster similar news to a events that can value added a news contributed. Classification and Clustering technique is a frequently used in Text Mining, and K-nearest neighbor(kNN) is one of most common algorithms apply in classification. However, kNN requires massive comparison on each individual article, and it becomes the performance bottlenecks of kNN. This research proposed Relative Text-Distance-based kNN(RTD-based kNN), the core concept of this method is establish a Base, a distance reference point, through a Vector Space, all documents can create the distance relationship through the relative distance between itself and base. Through the concept of relative distance, it can decrease the number of comparison and improve the efficiency. This research chooses a sample of 62 events (with total of 742 news articles) from Google News for the test and evaluation. Under the condition of RTD-based kNN and kNN with a no significant difference in F-measure (α=0.05), RTD-based kNN out perform kNN in time decreased by 28.13%. This confirms RTD-based kNN is a better method in clustering news event. At last, this research provides some of the research aspect for the future.

The aim of this diploma thesis is to obtain information on the motion of people in a scene from the record of the stationary camera. The procedure to detect exceptional events in the scene was designed. Exceptional events can be fast-moving persons, or persons moving in di erent places than everyone else in the scene. To trace the motion of persons, two algorithms were applied and tested - Optical flow and CAMSHIFT. The analysis of the resulting motions is performed by monitoring the progress of motion, and its comparison with the other motions in the scene. The analysis result is represented by detected exceptional motions that can be found in the video. The areas where the motion occurs in the scene, and where the motion is the most common are also described together with the motion direction analysis. The exceptional motion parts extracted from the video represent the main result of the work.

Tracab’s optical tracking system allows to track the 2-dimensional trajectories of players and ball during a football game. Using this data it is possible to train machine learning models to identify events that happen during the match. In this thesis, we explore the detection of corners, free kicks, and throw-in events by means of neural networks. Training a model to solve this task is not easy; the neural network needs to model the spatio-temporal interactions between different agents moving in a 2-dimensional space. We decided to address this problem using graph neural networks in combination with recurrent neural networks, which allow us to model respectively the spatial and temporal components of the data. Tracking the position of the ball is difficult, which makes the dataset noisy. In this thesis, we mainly work with a version of the dataset where the position of the ball has been manually corrected. However, to study how the noisy position of the ball affects the results we also train the models on the original data. The results show that detecting the corner and the throw-in is much easier than detecting the free kick. Moreover, the noisy position of the ball affects significantly the performance of the model. We conclude that to train the model on the original data it is necessary to use a much larger training set. Since the amount of training data for these events is limited, we also train the model on the more generic ball-dead-to-alive event, for which much more data is available, and we observe that by increasing the amount of training data the results can improve significantly. In this report, we also provide an in-depth discussion about all the challenges faced during the project and how different hyperparameters and design choices can affect the results. / Tracabs optiska spårningssystem gör det möjligt att spåra de 2-dimensionella banorna för spelare och boll under en fotbollsmatch. Med hjälp av dessa data är det möjligt att träna maskininlärningsmodeller för att identifiera händelser som inträffar under matchen. I denna avhandling utforskar vi upptäckten av hörnor, frisparkar och inkastningshändelser med hjälp av neurala nätverk. Att träna en modell för att lösa denna uppgift är inte lätt; det neurala nätverket behöver modellera de rums-temporala interaktionerna mellan olika agenter som rör sig i ett 2-dimensionellt rum. Vi bestämde oss för att ta itu med detta problem med hjälp av grafiska neurala nätverk i kombination med återkommande neurala nätverk, vilket gör att vi kan modellera de rumsliga respektive temporala komponenterna i datan. Det är svårt att spåra bollens position, vilket gör datauppsättningen bullrig. I detta examensarbete arbetar vi främst med en version av datamängden där bollens position har korrigerats manuellt. Men för att studera hur bollens bullriga position påverkar resultaten tränar vi också modellerna på originaldata. Resultaten visar att det är mycket lättare att upptäcka hörna och inkastet än att upptäcka frisparken. Dessutom påverkar bollens bullriga position avsevärt modellens prestanda. Vi drar slutsatsen att för att träna modellen på originaldata är det nödvändigt att använda en mycket större träningsuppsättning. Eftersom mängden träningsdata för dessa evenemang är begränsad, tränar vi också modellen på den mer generiska bollen död-till-levande-händelsen, för vilken mycket mer data finns tillgänglig, och vi observerar att genom att öka mängden träningsdata resultaten kan förbättras avsevärt. I denna rapport ger vi också en fördjupad diskussion om alla utmaningar som ställs inför under projektet och hur olika hyperparametrar och designval kan påverka resultaten.

This work examines sleep apnea syndrome, sleep physiology and self control of respiration during sleep. There is a review of respiration disorders during sleep and methods of monitoring sleep apnea syndrome. In another part the data of monitoration are processed and method of flow, saturation and snoring signal events detection is described, program algorithm is described and results are presented.