The nature of visual properties used for object perception in mid- and high-level vision areas of the brain is poorly understood. Past studies have employed simplistic stimuli probing models limited in descriptive power and mathematical under-pinnings. Unfortunately, pursuit of more complex stimuli and properties requires searching through a wide, unknown space of models and of images. The difficulty of this pursuit is exacerbated in brain research by the limited number of stimulus responses that can be collected for a given human subject over the course of an experiment. To more quickly identify complex visual features underlying cortical object perception, I develop, test, and use a novel method in which stimuli for use in the ongoing study are selected in realtime based on fMRI-measured cortical responses to recently-selected and displayed stimuli. A variation of the simplex method controls this ongoing selection as part of a search in visual space for images producing maximal activity — measured in realtime — in a pre-determined 1 cm3 brain region. I probe cortical selectivities during this search using photographs of real-world objects and synthetic “Fribble” objects. Real-world objects are used to understand perception of naturally-occurring visual properties. These objects are characterized based on feature descriptors computed from the scale invariant feature transform (SIFT), a popular computer vision method that is well established in its utility for aiding in computer object recognition and that I recently found to account for intermediate-level representations in the visual object processing pathway in the brain. Fribble objects are used to study object perception in an arena in which visual properties are well defined a priori. They are constructed from multiple well-defined shapes, and variation of each of these component shapes produces a clear space of visual stimuli. I study the behavior of my novel realtime fMRI search method, to assess its value in the investigation of cortical visual perception, and I study the complex visual properties my method identifies as highly-activating selected brain regions in the visual object processing pathway. While there remain further technical and biological challenges to overcome, my method uncovers reliable and interesting cortical properties for most subjects — though only for selected searches performed for each subject. I identify brain regions selective for holistic and component object shapes and for varying surface properties, providing examples of more precise selectivities within classes of visual properties previously associated with cortical object representation. I also find examples of “surround suppression,” in which cortical activity is inhibited upon viewing stimuli slightly deviation from the visual properties preferred by a brain region, expanding on similar observations at lower levels of vision.

We trained a combined deep convolutional neural network to predict seals’ age (3 categories) and gender (2 categories). The entire dataset contains 110 seals with around 489 thousand location records. Most records are continuous and measured in a certain step. We created five convolutional layers for feature representation and established two fully connected structure as age’s and gender’s classifier, respectively. Each classifier consists of three fully connected layers. Treating seals’ latitude and longitude as input, entire deep learning network, which includes 780,000 neurons and 2,097,000 parameters, can reach to 70.72% accuracy rate for predicting seals’ age and simultaneously achieve 79.95% for gender estimation.

Cette thèse porte sur la reconnaissance visuelle d'objets, un domaine qui reste un défi majeur en vision par ordinateur. En effet, malgré plus de vingt années de recherche, de nombreuses facettes du problème restent a ce jour irrésolues. La conception d'un système de reconnaissance d'objets repose essentiellement sur trois aspects: la représentation, la détection et l'apprentissage automatique. La principale contribution de cette thèse est de proposer un système générique pour la représentation statistique des caractéristiques visuelles et leur détection dans les images. Le modèle proposé combine différents concepts récemment proposés en vision par ordinateur, machine learning et neurosciences: a savoir les relations spatiales entre des caractéristiques visuelles, les modèles graphiques ainsi que les hiérarchies de cellules complexes. Le résultat de cette association prend la forme d'une hiérarchie de classes de caractéristiques visuelles. Son principal intérêt est de fournir un modèle représentant, à la fois, les aspects visuels locaux et globaux, en utilisant la structure géométrique et l'apparence des objets. L'exploitation des modèles graphiques offre un cadre probabiliste pour la représentation des hiérarchies et leur utilisation pour l'inférence. Un algorithme d'échange de messages récemment proposé (NBP) est utilisé pour inférer la position des caractéristiques dans les images. Lors de l'apprentissage, les hiérarchies sont construites de manière incrémentale en partant des caractéristiques de bas-niveaux. L'algorithme est basé sur l'analyse des co-occurrences. Il permet d'estimer la structure et les paramètres des hiérarchies. Les performances offertes par ce nouveau système sont évaluées sur différentes bases de données d'objets de difficulté croissante. Par ailleurs, un survol de l'état de l'art concernant les méthodes de reconnaissances d'objets et les détecteurs de caractéristiques offre une vue globale du domaine.

Neural networks based deep learning methods have gained significant success in several real world tasks: from machine translation to web recommendation, and it is also greatly improving the computer vision and the natural language processing. Compared with conventional machine learning techniques, neural network based deep learning do not require careful engineering and consideration domain expertise to design a feature extractor that transformed the raw data to a suitable internal representation. Its extreme efficacy on multiple levels of representation and feature learning ensures this type of approaches can process high dimensional data. It integrates the feature representation, learning and recognition into a systematical framework, which allows the learning starts at one level (i.e., being with raw input) and end at a higher slightly more abstract level. By simply stacking enough such transformations, very complex functions can be obtained. In general, high level feature representation facilitate the discrimination of patterns, and additionally can reduce the impact of irrelevant variations. However, previous studies indicate that deep composition of the networks make the training errors become vanished. To overcome this weakness, several techniques have been developed, for instance, dropout, stochastic gradient decent and residual network structures. In this study, we incorporates latent information into different network structures (e.g., restricted Boltzmann machine, recursive neural networks, and long short term memory). The conditional latent information reflects the high dimensional correlation existed in the data structure, and the typical network structure may not learn this kind of features due to limitation of the initial design (i.e., the network size the parameters). Similarly to residual nets, the conditional neural networks jointly learns the global features and local features, and the specifically designed network structure helps to incorporate the modulation derived from the probability distribution. The proposed models have been widely tested in different datasets, for instance, the conditional RBM has been applied to detect the speech components, and a language model based gated RBM has been used to recognize speech related EEG patterns. The conditional RNN has been tested in both general natural language modeling and medical notes prediction tasks. The results indicate that by introducing conditional branches in the conventional network structures, the latent features can be globally and locally learned.

The problem of location or movement prediction can be described as the task of predicting the future location of an item using the past locations of that item. It is a problem of increasing interest with the arrival of location-based services and autonomous vehicles. Even if short term prediction is more commonly studied, especially in the case of vehicles, long-term prediction can be useful in many applications like scheduling, resource managing or traffic prediction. In this master thesis project, I present a feature representation of movement that can be used for learning of long-term movement patterns and for long-term movement prediction both in space and time. The representation relies on periodicity in data and is based on weighted n-grams of windowed trajectories. The algorithm is evaluated on heavy transport vehicles movement data to assess its ability to from a search index retrieve vehicles that with high probability will move along a route that matches a desired transport mission. Experimental results show the algorithm is able to achieve a consistent low prediction distance error rate across different transport lengths in a limited geographical area under business operation conditions. The results also indicate that the total population of vehicles in the index is a critical factor in the algorithm performance and therefore in its real-world applicability. / Lokaliserings- eller rörelseprognosering kan beskrivas som uppgiften att förutsäga ett objekts framtida placering med hjälp av de tidigare platserna för objektet. Intresset för problemet ökar i och med införandet av platsbaserade tjänster och autonoma fordon. Även om det är vanligare att studera kortsiktiga förutsägelser, särskilt när det gäller fordon, kan långsiktiga förutsägelser vara användbara i många applikationer som schemaläggning, resurshantering eller trafikprognoser. I detta masterprojekt presenterar jag en feature-representation av rörelse som kan användas för att lära in långsiktiga rörelsemönster och för långsiktig rörelseprediktion både i rymden och tiden. Representationen bygger på periodicitet i data och är baserad på att dela upp banan i fönster och sedan beräkna viktade n-grams av banorna från de olika fönstren. Algoritmen utvärderas på transportdata för tunga transportfordon för att bedöma dess förmåga att från ett sökindex hämta fordon som med stor sannolikhet kommer att röra sig längs en rutt som matchar ett önskat transportuppdrag. Experimentella resultat visar att algoritmen kan uppnå ett konsekvent lågt fel i relativt predikterat avstånd över olika transportlängder i ett begränsat geografiskt område under verkliga förhållanden. Resultaten indikerar även att den totala populationen av fordon i indexet är en kritisk faktor för algoritmens prestanda och därmed även för dess applicerbarhet för verklig användning.

Submitted by Maria Aparecida (cidazen@gmail.com) on 2017-04-27T13:46:48Z No. of bitstreams: 1 Caio Eduardo Falcão Matos.pdf: 1884390 bytes, checksum: a5489f8f52a87e6c5958458ed5470488 (MD5) / Made available in DSpace on 2017-04-27T13:46:48Z (GMT). No. of bitstreams: 1 Caio Eduardo Falcão Matos.pdf: 1884390 bytes, checksum: a5489f8f52a87e6c5958458ed5470488 (MD5) Previous issue date: 2017-02-08 / Breast cancer is one of the leading causes of death among women over the world. The high mortality rates that cancers achieves across the world highlight the importance of developing and investigating the means for the early detection and diagnosis of this disease. Computer Detection and Diagnosis Systems (Computer Assisted Detection / Diagnosis) have been used and proposed as a way to help health professionals. This work proposes a new methodology for discriminating patterns of malignancy and benignity of masses in mammography images by analysis of local characteristics. To do so, it is proposed a combined methodology of feature detectors and descriptors with a model of data representation for an analysis. The goal is to capture both texture and geometry in areas of mammograms. We use the SIFT, SURF and ORB detectors, and the descriptors HOG, LBP, BRIEF and Haar Wavelet. The generated characteristics are coded by a bag of features model to provide a new representation of the data and therefore decrease a dimensionality of the space of characteristics. Finally, this new representation is classified using three approaches: Support Vector Machine, Random Forest, and Adaptive Boosting to differentiate as malignant and benign masses. The methodology provides promising results for the diagnosis of malignant and benign mass encouraging that as local characteristics generated by descriptors and detectors produce a satisfactory a discriminating set. / O câncer de mama é apontado como uma das principais causas de morte entre as mulheres. As altas taxas de mortalidade e registros de ocorrência desse câncer em todo o mundo evidenciam a importância do desenvolvimento e investigação de meios para a detecção e diagnóstico precoce dessa doença. Sistemas de Detecção e Diagnóstico auxiliados por computador (Computer Aided Detection/Diagnosis) vêm sendo usados e propostos como forma de auxílio aos profissionais de saúde. Este trabalho propõe uma metodologia para discriminação de padrões de malignidade e benignidade de massas em imagens de mamografia através da análise de características locais. Para tanto, a metodologia combina detectores e descritores de características locais com um modelo de representação de dados para a análise, tanto de textura quanto de geometria em regiões extraídas das mamografias. São utilizados os detectores SIFT, SURF e ORB, e descritores HOG, LBP, BRIEF e Haar Wavelet. Com as características geradas é aplicado o modelo Bag of Features em uma etapa de representação que objetiva prover nova representação dos dados e por conseguinte diminuir a dimensionalidade do espaço de características. Por fim, esta nova representação é classificada utilizando três abordagens: Máquina de Vetores de Suporte, Random Forests e Adaptive Boosting visando diferenciar as massas malignas e benignas. A metodologia contém resultados promissores para o diagnóstico de massas malignas e benignas fomentando que as características locais geradas pelos descritores e detectores produzem um conjunto descriminate satisfatório.

When classifying texts using a linear classifier, the texts are commonly represented as feature vectors. Previous methods to represent features as vectors have been unable to capture the context of individual words in the texts, in theory leading to a poor representation of natural language. Bidirectional Encoder Representations from Transformers (BERT), uses a multi-headed self-attention mechanism to create deep bidirectional feature representations, able to model the whole context of all words in a sequence. A BERT model uses a transfer learning approach, where it is pre-trained on a large amount of data and can be further fine-tuned for several down-stream tasks. This thesis uses one multilingual, and two dedicated Swedish BERT models, for the task of classifying Swedish texts as of either easy-to-read or standard complexity in their respective domains. The performance on the text classification task using the different models is then compared both with feature representation methods used in earlier studies, as well as with the other BERT models. The results show that all models performed better on the classification task than the previous methods of feature representation. Furthermore, the dedicated Swedish models show better performance than the multilingual model, with the Swedish model pre-trained on more diverse data outperforming the other.

The main aim of this master's thesis is to describe the subject of the implementation of decision boundaries with the help of artificial neural networks. The objective is to present theoretical knowledge concerning this field and on practical examples prove these statements. The work contains basic theoretical description of the field of pattern recognition and the field of feature based representation of objects. A classificator working on the basis of Bayes decision is presented in this part, and other types of classificators are named as well. The work then deals with artificial neural networks in more detail; it contains a theoretical description of their function and their abilities in the creation of decision boundaries in the feature plane. Examples are shown from literature for the use of neural networks in corresponding problems. As part of this work, the program ANN-DeBC was created using Matlab, for the generation of practical results about the usage of feed-forward neural networks for the implementation of decision boundaries. The work contains a detailed description of this program, and the achieved results are presented and analyzed. It is shown as well, how artificial neural networks are creating decision boundaries in the form of geometrical shapes. The effects of the chosen topology of the neural network and the number of training samples on the success of the classification are observed, and the minimal values of these parameters are determined for the successful creation of decision boundaries at the individual examples. Furthermore, it's presented how the neural networks behave at the classification of realistically distributed training samples, and what methods can affect the shape of the created decision boundaries.

Ces travaux de recherche concernent la modélisation de l'information dynamique fonctionnelle fournie par les champs de déplacements apparents à l'aide de base de polynômes orthogonaux. Leur objectif est de modéliser le mouvement et la texture extraites afin de l'exploiter dans les domaines de l'analyse et de la reconnaissance automatique d'images et de vidéos. Nous nous intéressons aussi bien aux mouvements humains qu'aux textures dynamiques. Les bases de polynômes orthogonales ont été étudiées. Cette approche est particulièrement intéressante car elle offre une décomposition en multi-résolution et aussi en multi-échelle. La première contribution de cette thèse est la définition d'une méthode spatiale de décomposition d'image : l'image est projetée et reconstruite partiellement avec un choix approprié du degré d'anisotropie associé à l'équation de décomposition basée sur des transformations polynomiales. Cette approche spatiale est étendue en trois dimensions afin d'extraire la texture dynamique dans des vidéos. Notre deuxième contribution consiste à utiliser les séquences d'images qui représentent les parties géométriques comme images initiales pour extraire les flots optiques couleurs. Deux descripteurs d'action, spatial et spatio-temporel, fondés sur la combinaison des informations du mouvement/texture sont alors extraits. Il est ainsi possible de définir un système permettant de reconnaître une action complexe (composée d'une suite de champs de déplacement et de textures polynomiales) dans une vidéo. / The research relies on modeling the dynamic functional information from the fields of apparent movement using basic orthogonal polynomials. The goal is to model the movement and texture extracted for automatic analysis and recognition of images and videos. We are interested both in human movements as dynamic textures. Orthogonal polynomials bases were studied. This approach is particularly interesting because it offers a multi-resolution and a multi-scale decomposition. The first contribution of this thesis is the definition of method of image spatial decomposition: the image is projected and partially rebuilt with an appropriate choice of the degree of anisotropy associated with the decomposition equation based on polynomial transformations. This spatial approach is extended into three dimensions to retrieve the dynamic texture in videos. Our second contribution is to use image sequences that represent the geometric parts as initial images to extract color optical flow. Two descriptors of action, spatial and space-time, based on the combination of information of motion / texture are extracted. It is thus possible to define a system to recognize a complex action (composed of a series of fields of motion and polynomial texture) in a video.

This thesis proposes a general approach to solve the offline flight-maneuver identification problem using machine learning methods. The purpose of the study was to provide means for the aircraft professionals at the flight test and verification department of Saab Aeronautics to automate the procedure of analyzing flight test data. The suggested approach succeeded in generating binary classifiers and multiclass classifiers that identified six flight maneuvers of different complexity from real flight test data. The binary classifiers solved the problem of identifying one maneuver from flight test data at a time, while the multiclass classifiers solved the problem of identifying several maneuvers from flight test data simultaneously. To achieve these results, the difficulties that this time series classification problem entailed were simplified by using different strategies. One strategy was to develop a maneuver extraction algorithm that used handcrafted rules. Another strategy was to represent the time series data by statistical measures. There was also an issue of an imbalanced dataset, where one class far outweighed others in number of samples. This was solved by using a modified oversampling method on the dataset that was used for training. Logistic Regression, Support Vector Machines with both linear and nonlinear kernels, and Artifical Neural Networks were explored, where the hyperparameters for each machine learning algorithm were chosen during model estimation by 4-fold cross-validation and solving an optimization problem based on important performance metrics. A feature selection algorithm was also used during model estimation to evaluate how the performance changes depending on how many features were used. The machine learning models were then evaluated on test data consisting of 24 flight tests. The results given by the test data set showed that the simplifications done were reasonable, but the maneuver extraction algorithm could sometimes fail. Some maneuvers were easier to identify than others and the linear machine learning models resulted in a poor fit to the more complex classes. In conclusion, both binary classifiers and multiclass classifiers could be used to solve the flight maneuver identification problem, and solving a hyperparameter optimization problem boosted the performance of the finalized models. Nonlinear classifiers performed the best on average across all explored maneuvers.