Temporal textures are textures with motion. Examples include wavy water, rising steam and fire. We model image sequences of temporal textures using the spatio-temporal autoregressive model (STAR). This model expresses each pixel as a linear combination of surrounding pixels lagged both in space and in time. The model provides a base for both recognition and synthesis. We show how the least squares method can accurately estimate model parameters for large, causal neighborhoods with more than 1000 parameters. Synthesis results show that the model can adequately capture the spatial and temporal characteristics of many temporal textures. A 95% recognition rate is achieved for a 135 element database with 15 texture classes. 1.

This thesis is organized in three parts. In the first part, the Self-Organizing Map algorithm is introduced. The discussion focuses on the analysis of the Self-Organizing Map algorithm. It is shown that the nonlinear nature of the algorithm makes it difficult to analyze the algorithm except in some trivial cases. In the second part the Self-Organizing Map algorithm is applied to several patterns sequence analysis tasks. The first application is a voice quality analysis system. It is shown that the Self-Organizing Map algorithm can be applied to voice analysis by providing the visualization of certain deviations. The key point in the applicability of Self-Organizing Map algorithm is the topological nature of the mapping; similar voice samples are mapped to nearby locations in the map. The second application is a speech recognition system. Through several experiments it is demonstrated that by collecting some time dependent features and using them in conjunction with the basic Self-Organ...

The forces of contact during manipulation convey substantial information about the state of the manipulation.

Automatic Speech Recognition (ASR) still poses a problem to researchers. In particular, most ASR systems have not been able to fully handle adverse acoustic environments. Although a large number of modifications have resulted in increased levels of performance robustness, ASR systems still fall short of human recognition ability in a large number of environments. A possible shortcoming of the typical ASR system is the reliance on a single stream of front-end acoustic features and acoustic modeling feature probabilities. A single front-end feature extraction algorithm may not be capable of maintaining robustness to arbitrary acoustic environments. Acoustic modeling will also degrade due to distributional changes caused by the acoustic environment. This thesis explores the parallel use of multiple front-end and acoustic modeling elements to improve upon this shortcoming. Each ASR acoustic modeling component is trained to estimate class posterior probabilities in a particular acoustic environment. In addition to discriminative training of the probability estimator, existing feature extraction algorithms are modi#ed in suchaway as to improve class discrimination in the training environment. More specifically, Linear Discriminant Analysis provides a mechanism for obtaining discriminant temporal basis functions that can replace components of the existing algorithms that were designed in either an empirical or intuitive manner. Probability streams are generate...

The bore profile and input impedance of a musical wind instrument provide valuable information about its acoustical properties. The time domain technique of acoustic pulse reflectometry can be used to measure the input impulse response of a tubular object, such as a wind instrument, from which both its bore profile and input impedance can be calculated. In this thesis, after a discussion of the theory of acoustic pulse reflectometry, the operation of a practical reflectometer is described and measurements of input impulse response, bore profile and input impedance are investigated. In general, the experimentally measured input impulse response of a tubular object contains a DC offset which must be removed for accurate bore reconstruction. A new, faster method of determining the DC offset is introduced which doesn't require prior knowledge of the object's dimensions. The bore profile of a test object, calculated by applying a lossy reconstruction algorithm to its input impulse respon...

Gerstacker et al.: On Prefilter Computation for Reduced-State Equalization 2 it is shown that high performance can be obtained for TDMA mobile communications systems, if the LP scheme is employed for prefiltering.

This report includes a very short survey of the research done by me, A. R. Kian Abolfazlian, under Part A of my Ph.D study, at Computer Science Department, Aarhus University, Denmark. The organisation of this report is as this: In chapter 2, I shall give a short survey on the inconsistency results I have achieved for AI. These results are taken from [Abolfazlian 94A]. The Lemata used for obtaing the proofs for the Theorems along with the proofs themselves are omitted here. They can be found in [Abolfazlian 94A]. In chapter 3, I shall give a short introduction to the field of Computational Intelligence. In chapters 4 to 6, I shall give my results obtained in the three main areas of research within the field of computational Intelligence. The more detailed descriptions of the results mentioned in these chapters should be sought in the references given in the respective chapters. In chapter 7, I shall give a short description of other researchs initiated or participated by me during this period. And finaly In chapter 8, I shall give a description of the work planned for Part B of my Ph.D study. At last I want to thank God for having given me the strength for going through Part A of my Ph.D study. I enjoyed my work, and would gladly do it again if the chance were offered me. A. R. Kian Abolfazlian Chapter 2 Artificial Intelligence and Inconsistency There are various psychological and philosophical understandings of how we, human beings, operate, and what our intelligence is about. These understandings have given rise to different research programmes within the field of AI. The most dominant of all these ideas has formed what John Haugeland has called "Good Old Fashioned Artificial Intelligence (GOFAI)", which has based its understanding of human beings on the idea of Turi...

The analysis of a speech segment is conventionally performed through linear prediction and the subsequent minimisation of a data error term in the least squares sense. The parameters derived as such maximise the likelihood of the data. In a learning problem, the addition of penalty terms, or regularisers, to the data term facilitates the estimation of the Maximum a Posteriori , or MAP, parameters. A direct equivalence can be drawn between the type of regulariser used and the prior assumptions regarding the solution. The Bayesian evidence procedure provides a framework for MAP parameter estimation and model order selection. In this paper, the use of suitable quadratic regularisers for the determination of linear prediction MAP parameters is addressed. The application of continuity constraints across successive speech segments will be demonstrated to enhance the tracking of formants for speech embedded in gaussian noise. The use of variable order models for speech analysis-synthesis is a...

From the TIMIT database labelled speech waveform segments from 33 speakers were extracted. There were 8 categories of speech data, each representing a vowel sound. In each category were 80 - 130 utterances. The waveform segments were processed by taking a FFT, on 32 msec frames and binning the result into 12 frequency bands. This way each frame will be represented by 12 numbers/values. They become points in IR 12 and each utterance is a short trajectory in IR 12 . The 8 vowel categories become 8 clusters of such trajectories. By projecting the clusters onto the screen of a SUN workstation, it was observed that each vowel cluster appeared to be substantially gaussian. The covariance matrix and the centers were computed. Dimension estimates of each vowel by Principal Components Analysis show that the eight clusters lie close to a plane in the filterbankspace, and that the principal axes of the vowel clusters make a small and consistent angle with respect to this plane. This confirms ...

The solutions of the partial realization problem have to satisfy a finite number of interpolation conditions at 1. The minimal degree of an interpolating deterministic system is called the algebraic degree or McMillan degree of the partial covariance sequence and is easy to compute. The solutions of the partial stochastic realization problem have to satisfy the same interpolation conditions and have to fulfill a positive realness constraint. The minimal degree of a stochastic realization is called the positive degree. The interpolating deterministic solutions can be parameterized by the KimuraGeorgiou parameterization. In the literature, the solutions of the partial stochastic realization problem are then described by checking the positive realness constraints for each interpolating deterministic system. In this paper, an alternative parameterization for the deterministic solutions of the interpolation problem is presented. Both the solutions of the partial and partial stochastic reali...