Sparse Representation based Anomaly Detection using HOMV in H.264 Compressed Videos

In this paper, we have proposed an anomaly detection algorithm based on Histogram of Oriented Motion Vectors (HOMV) 1] in sparse representation framework. Usual behavior is learned at each location by sparsely representing the HOMVs over learnt normal feature bases obtained using an online dictionary learning algorithm. In the end, anomaly is detected based on the likelihood of the occurrence of sparse coefficients at that location. The proposed approach is found to be robust compared to existing methods as demonstrated in the experiments on UCSD Ped1 and UCSD Ped2 datasets.

A Bacterial Foraging Optimization and Learning Automata Based Feature Selection for Motor Imagery EEG Classification

Selection of relevant features is an open problem in Brain-computer interfacing (BCI) research. Sometimes, features extracted from brain signals are high dimensional which in turn affects the accuracy of the classifier. Selection of the most relevant features improves the performance of the classifier and reduces the computational cost of the system. In this study, we have used a combination of Bacterial Foraging Optimization and Learning Automata to determine the best subset of features from a given motor imagery electroencephalography (EEG) based BCI dataset. Here, we have employed Discrete Wavelet Transform to obtain a high dimensional feature set and classified it by Distance Likelihood Ratio Test. Our proposed feature selector produced an accuracy of 80.291% in 216 seconds.

MULTI-INSTRUMENT DETECTION IN POLYPHONIC MUSIC USING GAUSSIAN MIXTURE BASED FACTORIAL HMM

We formulate the problem of detecting the constituent instruments in a polyphonic music piece as a joint decoding problem. From monophonic data, parametric Gaussian Mixture Hidden Markov Models (GM-HMM) are obtained for each instrument. We propose a method to use the above models in a factorial framework, termed as Factorial GM-HMM (F-GM-HMM). The states are jointly inferred to explain the evolution of each instrument in the mixture observation sequence. The dependencies are decoupled using variational inference technique. We show that the joint time evolution of all instruments' states can be captured using F-GM-HMM. We compare performance of proposed method with that of Student's-t mixture model (tMM) and GM-HMM in an existing latent variable framework. Experiments on two to five polyphony with 8 instrument models trained on the RWC dataset, tested on RWC and TRIOS datasets show that F-GM-HMM gives an advantage over the other considered models in segments containing co-occurring instruments.

Sparse representation-based human detection: a scale-embedded dictionary approach

Human detection is a complex problem owing to the variable pose that they can adopt. Here, we address this problem in sparse representation framework with an overcomplete scale-embedded dictionary. Histogram of oriented gradient features extracted from the candidate image patches are sparsely represented by the dictionary that contain positive bases along with negative and trivial bases. The object is detected based on the proposed likelihood measure obtained from the distribution of these sparse coefficients. The likelihood is obtained as the ratio of contribution of positive bases to negative and trivial bases. The positive bases of the dictionary represent the object (human) at various scales. This enables us to detect the object at any scale in one shot and avoids multiple scanning at different scales. This significantly reduces the computational complexity of detection task. In addition to human detection, it also finds the scale at which the human is detected due to the scale-embedded structure of the dictionary.

Empirical Analysis of Sampling Based Estimators for Evaluating RBMs

The Restricted Boltzmann Machines (RBM) can be used either as classifiers or as generative models. The quality of the generative RBM is measured through the average log-likelihood on test data. Due to the high computational complexity of evaluating the partition function, exact calculation of test log-likelihood is very difficult. In recent years some estimation methods are suggested for approximate computation of test log-likelihood. In this paper we present an empirical comparison of the main estimation methods, namely, the AIS algorithm for estimating the partition function, the CSL method for directly estimating the log-likelihood, and the RAISE algorithm that combines these two ideas.

Distributed maximum likelihood for self-localization in sensor networks

We show that the sensor localization problem can be cast as a static parameter estimation problem for Hidden Markov Models and we develop fully decentralized versions of the Recursive Maximum Likelihood and the Expectation-Maximization algorithms to localize the network. For linear Gaussian models, our algorithms can be implemented exactly using a distributed version of the Kalman filter and a message passing algorithm to propagate the derivatives of the likelihood. In the non-linear case, a solution based on local linearization in the spirit of the Extended Kalman Filter is proposed. In numerical examples we show that the developed algorithms are able to learn the localization parameters well.

Sequential decision making in general state space models

Many problems in control and signal processing can be formulated as sequential decision problems for general state space models. However, except for some simple models one cannot obtain analytical solutions and has to resort to approximation. In this thesis, we have investigated problems where Sequential Monte Carlo (SMC) methods can be combined with a gradient based search to provide solutions to online optimisation problems. We summarise the main contributions of the thesis as follows. Chapter 4 focuses on solving the sensor scheduling problem when cast as a controlled Hidden Markov Model. We consider the case in which the state, observation and action spaces are continuous. This general case is important as it is the natural framework for many applications. In sensor scheduling, our aim is to minimise the variance of the estimation error of the hidden state with respect to the action sequence. We present a novel SMC method that uses a stochastic gradient algorithm to find optimal actions. This is in contrast to existing works in the literature that only solve approximations to the original problem. In Chapter 5 we presented how an SMC can be used to solve a risk sensitive control problem. We adopt the use of the Feynman-Kac representation of a controlled Markov chain flow and exploit the properties of the logarithmic Lyapunov exponent, which lead to a policy gradient solution for the parameterised problem. The resulting SMC algorithm follows a similar structure with the Recursive Maximum Likelihood(RML) algorithm for online parameter estimation. In Chapters 6, 7 and 8, dynamic Graphical models were combined with with state space models for the purpose of online decentralised inference. We have concentrated more on the distributed parameter estimation problem using two Maximum Likelihood techniques, namely Recursive Maximum Likelihood (RML) and Expectation Maximization (EM). The resulting algorithms can be interpreted as an extension of the Belief Propagation (BP) algorithm to compute likelihood gradients. In order to design an SMC algorithm, in Chapter 8 uses a nonparametric approximations for Belief Propagation. The algorithms were successfully applied to solve the sensor localisation problem for sensor networks of small and medium size.

Probabilistic inference for fast learning in control

A novel framework is provided for very fast model-based reinforcement learning in continuous state and action spaces. It requires probabilistic models that explicitly characterize their levels of condence. Within the framework, exible, non-parametric models are used to describe the world based on previously collected experience. It demonstrates learning on the cart-pole problem in a setting where very limited prior knowledge about the task has been provided. Learning progressed rapidly, and a good policy found after only a small number of iterations.

New methods for generating populations in Markov network based EDAs: Decimation strategies and model-based template recombination

Methods for generating a new population are a fundamental component of estimation of distribution algorithms (EDAs). They serve to transfer the information contained in the probabilistic model to the new generated population. In EDAs based on Markov networks, methods for generating new populations usually discard information contained in the model to gain in efficiency. Other methods like Gibbs sampling use information about all interactions in the model but are computationally very costly. In this paper we propose new methods for generating new solutions in EDAs based on Markov networks. We introduce approaches based on inference methods for computing the most probable configurations and model-based template recombination. We show that the application of different variants of inference methods can increase the EDAs’ convergence rate and reduce the number of function evaluations needed to find the optimum of binary and non-binary discrete functions.

Estimation and inference for grasping and manipulation tasks using vision and kinesthetic sensors

This thesis presents a novel framework for state estimation in the context of robotic grasping and manipulation. The overall estimation approach is based on fusing various visual cues for manipulator tracking, namely appearance and feature-based, shape-based, and silhouette-based visual cues. Similarly, a framework is developed to fuse the above visual cues, but also kinesthetic cues such as force-torque and tactile measurements, for in-hand object pose estimation. The cues are extracted from multiple sensor modalities and are fused in a variety of Kalman filters.

A hybrid estimator is developed to estimate both a continuous state (robot and object states) and discrete states, called contact modes, which specify how each finger contacts a particular object surface. A static multiple model estimator is used to compute and maintain this mode probability. The thesis also develops an estimation framework for estimating model parameters associated with object grasping. Dual and joint state-parameter estimation is explored for parameter estimation of a grasped object's mass and center of mass. Experimental results demonstrate simultaneous object localization and center of mass estimation.

Dual-arm estimation is developed for two arm robotic manipulation tasks. Two types of filters are explored; the first is an augmented filter that contains both arms in the state vector while the second runs two filters in parallel, one for each arm. These two frameworks and their performance is compared in a dual-arm task of removing a wheel from a hub.

This thesis also presents a new method for action selection involving touch. This next best touch method selects an available action for interacting with an object that will gain the most information. The algorithm employs information theory to compute an information gain metric that is based on a probabilistic belief suitable for the task. An estimation framework is used to maintain this belief over time. Kinesthetic measurements such as contact and tactile measurements are used to update the state belief after every interactive action. Simulation and experimental results are demonstrated using next best touch for object localization, specifically a door handle on a door. The next best touch theory is extended for model parameter determination. Since many objects within a particular object category share the same rough shape, principle component analysis may be used to parametrize the object mesh models. These parameters can be estimated using the action selection technique that selects the touching action which best both localizes and estimates these parameters. Simulation results are then presented involving localizing and determining a parameter of a screwdriver.

Lastly, the next best touch theory is further extended to model classes. Instead of estimating parameters, object class determination is incorporated into the information gain metric calculation. The best touching action is selected in order to best discern between the possible model classes. Simulation results are presented to validate the theory.

Wireless optical communication-based spatial pattern

We propose a novel communication technique which utilizes a set of mutually distinguishable optical patterns instead of convergent facula to transmit information. The communication capacity is increased by exploiting the optical spatial bandwidth resources. An optimum detector for this communication is proposed based on maximum-likelihood decision. The fundamental rule of designing signal spatial pattern is formulated from analysis of the probability of error decision. Finally, we present a typical electro-optical system scheme of the proposed communication. (c) 2006 Elsevier GmbH. All rights reserved.

Approximations for Binary Gaussian Process Classification

We provide a comprehensive overview of many recent algorithms for approximate inference in Gaussian process models for probabilistic binary classification. The relationships between several approaches are elucidated theoretically, and the properties of the different algorithms are corroborated by experimental results. We examine both 1) the quality of the predictive distributions and 2) the suitability of the different marginal likelihood approximations for model selection (selecting hyperparameters) and compare to a gold standard based on MCMC. Interestingly, some methods produce good predictive distributions although their marginal likelihood approximations are poor. Strong conclusions are drawn about the methods: The Expectation Propagation algorithm is almost always the method of choice unless the computational budget is very tight. We also extend existing methods in various ways, and provide unifying code implementing all approaches.

Defining plausible migration rates based on historical tagging data: a Bayesian mark-recapture model applied to English sole (Parophrys vetulus)

A generalized Bayesian population dynamics model was developed for analysis of historical mark-recapture studies. The Bayesian approach builds upon existing maximum likelihood methods and is useful when substantial uncertainties exist in the data or little information is available about auxiliary parameters such as tag loss and reporting rates. Movement rates are obtained through Markov-chain Monte-Carlo (MCMC) simulation, which are suitable for use as input in subsequent stock assessment analysis. The mark-recapture model was applied to English sole (Parophrys vetulus) off the west coast of the United States and Canada and migration rates were estimated to be 2% per month to the north and 4% per month to the south. These posterior parameter distributions and the Bayesian framework for comparing hypotheses can guide fishery scientists in structuring the spatial and temporal complexity of future analyses of this kind. This approach could be easily generalized for application to other species and more data-rich fishery analyses.

Identificação de danos estruturais via método de Monte Carlo com cadeias de Markov

O presente trabalho apresenta um estudo referente à aplicação da abordagem Bayesiana como técnica de solução do problema inverso de identificação de danos estruturais, onde a integridade da estrutura é continuamente descrita por um parâmetro estrutural denominado parâmetro de coesão. A estrutura escolhida para análise é uma viga simplesmente apoiada do tipo Euler-Bernoulli. A identificação de danos é baseada em alterações na resposta impulsiva da estrutura, provocadas pela presença dos mesmos. O problema direto é resolvido através do Método de Elementos Finitos (MEF), que, por sua vez, é parametrizado pelo parâmetro de coesão da estrutura. O problema de identificação de danos é formulado como um problema inverso, cuja solução, do ponto de vista Bayesiano, é uma distribuição de probabilidade a posteriori para cada parâmetro de coesão da estrutura, obtida utilizando-se a metodologia de amostragem de Monte Carlo com Cadeia de Markov. As incertezas inerentes aos dados medidos serão contempladas na função de verossimilhança. Três estratégias de solução são apresentadas. Na Estratégia 1, os parâmetros de coesão da estrutura são amostrados de funções densidade de probabilidade a posteriori que possuem o mesmo desvio padrão. Na Estratégia 2, após uma análise prévia do processo de identificação de danos, determina-se regiões da viga potencialmente danificadas e os parâmetros de coesão associados à essas regiões são amostrados a partir de funções de densidade de probabilidade a posteriori que possuem desvios diferenciados. Na Estratégia 3, após uma análise prévia do processo de identificação de danos, apenas os parâmetros associados às regiões identificadas como potencialmente danificadas são atualizados. Um conjunto de resultados numéricos é apresentado levando-se em consideração diferentes níveis de ruído para as três estratégias de solução apresentadas.