This master thesis deals with the detection and recognition of faces in the image. The content of this thesis is a description of methods that are used for the face detection and recognition. Method described in detail is the principal component analysis (PCA). This method is subsequently used in the implementation of face recognition in video sequence. In conjunction with the implementation work describes the OpenCV library package, which was used for implementation, specifically the C ++ API. Finally described application tests were done on two different video sequences.

Multidimenzionální indexování je účinným nástrojem pro zachycení podobností mezi objekty bez nutnosti jejich explicitní kategorizace. V posledních letech byla tato metoda hojně využívána pro anotaci objektů a tvořila významnou část publikací spojených s projektem Visipedia. Tato práce analyzuje možnosti strojového učení z multidimenzionálně indexovaných obrázků na základě jejich obrazových příznaků a přestavuje metody predikce multidimenzionálních souřadnic pro předem neznámé obrázky. Práce studuje příslušené algoritmy pro extrakci příznaků, analyzuje relevantní metody strojového účení a popisuje celý proces vývoje takového systému. Výsledný systém je pak otestován na dvou různých datasetech a provedené experimenty prezentují první výsledky pro úlohu svého druhu.

Development of Internet and its massive spread resulted in increased volume of multimedia data. The increase in the amount of multimedia data raises the need for efficient similarity detection between multimedia files for the purpose of preventing and detecting violations of copyright licenses or for detection of similar or duplicate files. This thesis discusses the current options in the field of the content-based image and video comparison and focuses on the feature extraction techniques, distance metrics, design and implementation of the mediaDiff application module for the content-based comparison of video files.

This thesis deals with extraction of data from cell images and their visualisation. Cell images are processed by FISH method. It discusses theory of diagnosis evaluation automation and cell features visualization. That concerns image processing, cell nuclei segmentation, feature extraction and data visualization.

This thesis deals with optical devices and methods image processing for recognizing hand gestures. The types of gestures, possible applications, contact based devices and vision based devices are described in thesis. Next, a review of hand detection, features extraction and gesture classification is provided. Proposed gesture recognition system consists of infrared camera FLIR A655sc, infrared FLIR Lepton module, webcam Logitech S7500, method for hand gesture analysis and a database of gestures for classification. For each of the devices, gesture recognition is evaluated in terms of speed and accuracy in different environments. The proposed method was implemented in MATLAB.

From biometrical point of view, main features of retina are fovea, optic nerve and blood vessel tree. Blood vessel tree is unique for each person and this biological feature is used in biometric systems for person-recognition by retinal images. This document describes methods for optic disc and fovea localization, method for vessel tree segmentation, which is based on well-known \emph{Matched filters} method and also describes method for localization of blood vessel bifurcations. Main goal of this thesis is creation of program which can automatically preprocess input image, segment blood vessels and localize vessel bifircations. The program is implemented in Java with OpenCV library.

Sleep is an important part of life as it affects the performance of one's activities during all awake hours. The study of sleep and wakefulness is therefore of great interest, particularly to the clinical and medical fields where sleep disorders are diagnosed. When studying sleep, it is common to talk about different types, or stages, of sleep. A common task in sleep research is to determine the sleep stage of the sleeping subject as a function of time. This process is known as sleep stage scoring. In this study, I seek to determine whether there is any benefit to using unsupervised feature learning in the context of electroencephalogram-based (EEG) sleep scoring. More specifically, the effect of generating and making use of new feature representations for hand-crafted features of sleep data – meta-features – is studied. For this purpose, two scoring algorithms have been implemented and compared. Both scoring algorithms involve segmentation of the EEG signal, feature extraction, feature selection and classification using a support vector machine (SVM). Unsupervised feature learning was implemented in the form of a dimensionality-reducing deep-belief network (DBN) which the feature space was processed through. Both scorers were shown to have a classification accuracy of about 76 %. The application of unsupervised feature learning did not affect the accuracy significantly. It is speculated that with a better choice of parameters for the DBN in a possible future work, the accuracy may improve significantly. / Sömnen är en viktig del av livet eftersom den påverkar ens prestation under alla vakna timmar. Forskning om sömn and vakenhet är därför av stort intresse, i synnerhet för de kliniska och medicinska områdena där sömnbesvär diagnostiseras. I forskning om sömn är det är vanligt att tala om olika typer av sömn, eller sömnstadium. En vanlig uppgift i sömnforskning är att avgöra sömnstadiet av den sovande exemplaret som en funktion av tiden. Den här processen kallas sömnmätning. I den här studien försöker jag avgöra om det finns någon fördel med att använda oövervakad inlärning av särdrag för att utföra elektroencephalogram-baserad (EEG) sömnmätning. Mer specifikt undersöker jag effekten av att generera och använda nya särdragsrepresentationer som härstammar från handgjorda särdrag av sömndata – meta-särdrag. Två sömnmätningsalgoritmer har implementerats och jämförts för det här syftet. Sömnmätningsalgoritmerna involverar segmentering av EEG-signalen, extraktion av särdragen, urval av särdrag och klassificering genom användning av en stödvektormaskin (SVM). Oövervakad inlärning av särdrag implementerades i form av ett dimensionskrympande djuptrosnätverk (DBN) som användes för att bearbetasärdragsrymden. Båda sömnmätarna visades ha en klassificeringsprecision av omkring 76 %. Användningen av oövervakad inlärning av särdrag hade ingen signifikant inverkan på precisionen. Det spekuleras att precisionen skulle kunna höjas med ett mer lämpligt val av parametrar för djuptrosnätverket.

Over the last years researchers proposed numerous indoor localisation and navigation systems. However, solutions that use WiFi or Radio Frequency Identification require infrastructure to be deployed in the navigation area and infrastructureless techniques, e.g. the ones based on mobile cell ID or dead reckoning suffer from large accuracy errors. In this Thesis, we present a novel approach of infrastructure-less indoor navigation system based on computer vision Structure from Motion techniques. We implemented a prototype localisation and navigation system which can build a navigation map using area photos as input and accurately locate a user in the map. In our client-server architecture based system, a client is a mobile application, which allows a user to locate her or his position by simply taking a photo. The server handles map creation, localisation queries and pathnding. After the implementation, we evaluated the localisation accuracy and latency of the system by benchmarking navigation queries and the model creation algorithm. The system is capable of successfully navigating in Aalto University computer science department library. We were able to achieve an average error of 0.26 metres for successfully localised photos. In the Thesis, we also present challenges that we solved to adapt computer vision techniques for localisation purposes. Finally we observe the possible future work topics to adapt the system to a wide use. / Forskare har de senaste åren framfört olika inomhusnavigations- och lokaliseringssystem. Dock kräver lösningar som använder WiFi eller radiofrekvens identifikation en utbyggdnad av stödjande infrastruktur i navigationsområdena. Även teknikerna som används lider av precisionsfel. I det här examensarbetet redovisar vi en ny taktik för inomhusnavigation som använder sig av datorvisualiserings Structure from Motion-tekniker. Vi implementerade en navigationssystemsprototyp som använder bilder för att bygga en navigationskarta och kartlägga användarens position. I vårt klient-server baserat system är en klient en mobilapplikation som tillåter användaren att hitta sin position genom att ta en bild. På server-sidan hanteras kartor, lokaliseringsförfrågor och mättningar av förfrågorna och algoritmerna som används. Systemet har lyckats navigera genom Aalto Universitets datorvetenskapsbiblioteket. Vi lyckades uppnå en felmarginal pa 0.26 meter för lyckade lokaliseringsbilder. I arbetet redovisar vi utmaningar som vi har löst för att anpassa datorvisualiseringstekniker for lokalisering. Vi har även diskuterat potentialla framtida implementationer for att utvidga systemet.

Nowadays email is commonly used by citizens to establish communication with their government. On the received emails, governments deal with some common queries and subjects which some handling officers have to manually answer. Automatic email classification of the incoming emails allows to increase the communication efficiency by decreasing the delay between the query and its response. This thesis takes part within the IMAIL project, which aims to provide an automatic answering solution to the Swedish Social Insurance Agency (SSIA) (“Försäkringskassan” in Swedish). The goal of this thesis is to analyze and compare the classification performance of different sets of features extracted from SSIA emails on different automatic classifiers. The features extracted from the emails will depend on the previous preprocessing that is carried out as well. Compound splitting, lemmatization, stop words removal, Part-of-Speech tagging and Ngrams are the processes used in the data set. Moreover, classifications will be performed using Support Vector Machines, k- Nearest Neighbors and Naive Bayes. For the analysis and comparison of different results, precision, recall and F-measure are used. From the results obtained in this thesis, SVM provides the best classification with a F-measure value of 0.787. However, Naive Bayes provides a better classification for most of the email categories than SVM. Thus, it can not be concluded whether SVM classify better than Naive Bayes or not. Furthermore, a comparison to Dalianis et al. (2011) is made. The results obtained in this approach outperformed the results obtained before. SVM provided a F-measure value of 0.858 when using PoS-tagging on original emails. This result improves by almost 3% the 0.83 obtained in Dalianis et al. (2011). In this case, SVM was clearly better than Naive Bayes.

Práce pojednává o fonotaktickém a akustickém přístupu pro automatické rozpoznávání jazyka. První část práce pojednává o fonotaktickém přístupu založeném na výskytu fonémových sekvenci v řeči. Nejdříve je prezentován popis vývoje fonémového rozpoznávače jako techniky pro přepis řeči do sekvence smysluplných symbolů. Hlavní důraz je kladen na dobré natrénování fonémového rozpoznávače a kombinaci výsledků z několika fonémových rozpoznávačů trénovaných na různých jazycích (Paralelní fonémové rozpoznávání následované jazykovými modely (PPRLM)). Práce také pojednává o nové technice anti-modely v PPRLM a studuje použití fonémových grafů místo nejlepšího přepisu. Na závěr práce jsou porovnány dva přístupy modelování výstupu fonémového rozpoznávače -- standardní n-gramové jazykové modely a binární rozhodovací stromy. Hlavní přínos v akustickém přístupu je diskriminativní modelování cílových modelů jazyků a první experimenty s kombinací diskriminativního trénování a na příznacích, kde byl odstraněn vliv kanálu. Práce dále zkoumá různé druhy technik fúzi akustického a fonotaktického přístupu. Všechny experimenty jsou provedeny na standardních datech z NIST evaluaci konané v letech 2003, 2005 a 2007, takže jsou přímo porovnatelné s výsledky ostatních skupin zabývajících se automatickým rozpoznáváním jazyka. S fúzí uvedených technik jsme posunuli state-of-the-art výsledky a dosáhli vynikajících výsledků ve dvou NIST evaluacích.