Esta tese apresenta um estudo sobre a implementação de sistema de detecção e reconhecimento de faces no processador CELL na plataforma CBE, utilizando um sistema Playstation 3. Inicialmente, diversas abordagens para reconhecimento e detecção de faces são estudadas, bem como arquiteturas de processador multi núcleos. São apresentadas três implementação, sendo a segunda implementação premiada com quarto colocado no IBM CELL UNIVERSITY CHALLENGE 2007 para desenvolvimento de programas para plataforma Cell BE. A terceira implementação apresenta os resultados interessantes relacionados a vetorização do processamento dos dados da detecção de objetos e os recursos adotados para obter o melhor desempenho. / This thesis presents a study of face detection implementation on CBE plataform and employ the system with Playstation 3 hardware. Several approaches for face detection and recognition are studied as well as multicore processor architetures. We implemented three versions of system. First implementation was a naive reference implementation with worst performance. Second implementation granted fourth prize in IBM CELL UNIVERSITY CHALLENGE 2007 that incentive development on CBE plataform. Third implementation had most interesting results with vectorized approaches on code of object detection.

A new algorithm is proposed which uses the Hough Transform to recognize two dimensional objects independent of their orientations, sizes and locations. The binary image of an object is represented by a set of straight lines. Features of the straight lines, namely the lengths and the angles of their normals, their lengths and the end point positions are extracted using the Hough Transform. A data structure for the extracted lines is constructed so that it is efficient to match the features of the lines of one object to those of another object, and determine if one object is a rotated and/or scaled version of the other. Finally a generalized Hough Transform is used to match the end points of the two sets of lines. The simulation experiments show good results for objects with significant linear features .

For the last several years, convolutional neural network (CNN) based object detection systems have used a regression technique to predict improved object bounding boxes based on an initial proposal using low-level image features extracted from the CNN. In spite of its prevalence, there is little critical analysis of bounding-box regression or in-depth performance evaluation. This thesis surveys an array of techniques and parameter settings in order to further optimize bounding-box regression and provide guidance for its implementation. I refute a claim regarding training procedure, and demonstrate the effectiveness of using principal component analysis to handle unwieldy numbers of features produced by very deep CNNs.

In the design of autonomous mobile robots, databases have been used mainly to store information on the environment in which the device is to operate. For most of the models and ready systems, the database when used, is not a stand alone component in the system, rather it is only intended to keep static information on the disposition and properties of objects on the map.

Abstract not available

Abstract not available

Speech signal is basically meant to carry the information about the linguistic message. But, it also contains the speaker-specific information. It is generated by acoustically exciting the cavities of the mouth and nose, and can be used to recognize (identify/verify) a person. This thesis deals with the speaker identification task; i.e., to find the identity of a person using his/her speech from a group of persons already enrolled during the training phase. Listeners use many audible cues in identifying speakers. These cues range from high level cues such as semantics and linguistics of the speech, to low level cues relating to the speaker's vocal tract and voice source characteristics. Generally, the vocal tract characteristics are modeled in modern day speaker identification systems by cepstral coefficients. Although, these coeficients are good at representing vocal tract information, they can be supplemented by using both pitch and voicing information. Pitch provides very important and useful information for identifying speakers. In the current speaker recognition systems, it is very rarely used as it cannot be reliably extracted, and is not always present in the speech signal. In this thesis, an attempt is made to utilize this pitch and voicing information for speaker identification. This thesis illustrates, through the use of a text-independent speaker identification system, the reasonable performance of the cepstral coefficients, achieving an identification error of 6%. Using pitch as a feature in a straight forward manner results in identification errors in the range of 86% to 94%, and this is not very helpful. The two main reasons why the direct use of pitch as a feature does not work for speaker recognition are listed below. First, the speech is not always periodic; only about half of the frames are voiced. Thus, pitch can not be estimated for half of the frames (i.e. for unvoiced frames). The problem is how to account for pitch information for the unvoiced frames during recognition phase. Second, the pitch estimation methods are not very reliable. They classify some of the frames unvoiced when they are really voiced. Also, they make pitch estimation errors (such as doubling or halving of pitch value depending on the method). In order to use pitch information for speaker recognition, we have to overcome these problems. We need a method which does not use the pitch value directly as feature and which should work for voiced as well as unvoiced frames in a reliable manner. We propose here a method which uses the autocorrelation function of the given frame to derive pitch-related features. We call these features the maximum autocorrelation value (MACV) features. These features can be extracted for voiced as well as unvoiced frames and do not suffer from the pitch doubling or halving type of pitch estimation errors. Using these MACV features along with the cepstral features, the speaker identification performance is improved by 45%.

Personal identification of individuals is becoming increasingly adopted in society today. Due to the large number of electronic systems that require human identification, faster and more secure identification systems are pursued. Biometrics is based upon the physical characteristics of individuals; of these the fingerprint is the most common as used within law enforcement. Fingerprint-based systems have been introduced into the society but have not been well received due to relatively high rejection rates and false acceptance rates. This limited acceptance of fingerprint identification systems requires new techniques to be investigated to improve this identification method and the acceptance of the technology within society. Electronic fingerprint identification provides a method of identifying an individual within seconds quickly and easily. The fingerprint must be captured instantly to allow the system to identify the individual without any technical user interaction to simplify system operation. The performance of the entire system relies heavily on the quality of the original fingerprint image that is captured digitally. A single fingerprint scan for verification makes it easier for users accessing the system as it replaces the need to remember passwords or authorisation codes. The identification system comprises of several components to perform this function, which includes a fingerprint sensor, processor, feature extraction and verification algorithms. A compact texture feature extraction method will be implemented within an embedded microprocessor-based system for security, performance and cost effective production over currently available commercial fingerprint identification systems. To perform these functions various software packages are available for developing programs for windows-based operating systems but must not constrain to a graphical user interface alone. MATLAB was the software package chosen for this thesis due to its strong mathematical library, data analysis and image analysis libraries and capability. MATLAB enables the complete fingerprint identification system to be developed and implemented within a PC environment and also to be exported at a later date directly to an embedded processing environment. The nucleus of the fingerprint identification system is the feature extraction approach presented in this thesis that uses global texture information unlike traditional local information in minutiae-based identification methods. Commercial solid-state sensors such as the type selected for use in this thesis have a limited contact area with the fingertip and therefore only sample a limited portion of the fingerprint. This limits the number of minutiae that can be extracted from the fingerprint and as such limits the number of common singular points between two impressions of the same fingerprint. The application of texture feature extraction will be tested using variety of fingerprint images to determine the most appropriate format for use within the embedded system. This thesis has focused on designing a fingerprint-based identification system that is highly expandable using the MATLAB environment. The main components that are defined within this thesis are the hardware design, image capture, image processing and feature extraction methods. Selection of the final system components for this electronic fingerprint identification system was determined by using specific criteria to yield the highest performance from an embedded processing environment. These platforms are very cost effective and will allow fingerprint-based identification technology to be implemented in more commercial products that can benefit from the security and simplicity of a fingerprint identification system.

Ecological modelling problems have characteristics both featured in other modelling fields and specific ones, hence, methods developed and tested in other research areas may not be suitable for modelling ecological problems or may perform poorly when used on ecological data. This thesis identifies issues associated with the techniques typically used for solving ecological problems and develops new generic methods for decision support, especially suitable for ecological data modelling, which are characterised by: (1) adaptive learning, (2) knowledge discovery and (3) accurate prediction. These new methods have been successfully applied to challenging real world ecological problems. Despite the fact that the number of possible applications of computational intelligence methods in ecology is vast, this thesis primarily concentrates on two problems: (1) species establishment prediction and (2) environmental monitoring. Our review of recent papers suggests that multi-layer perceptron networks trained using the backpropagation algorithm are most widely used of all artificial neural networks for forecasting pest insect invasions. While the multi-layer perceptron networks are appropriate for modelling complex nonlinear relationships, they have rather limited exploratory capabilities and are difficult to adapt to dynamically changing data. In this thesis an approach that addresses these limitations is proposed. We found that environmental monitoring applications could benefit from having an intelligent taste recognition system possibly embedded in an autonomous robot. Hence, this thesis reviews the current knowledge on taste recognition and proposes a biologically inspired artificial model of taste recognition based on biologically plausible spiking neurons. The model is dynamic and is capable of learning new tastants as they become available. Furthermore, the model builds a knowledge base that can be extracted during or after the learning process in form of IF-THEN fuzzy rules. It also comprises a layer that simulates the influence of taste receptor cells on the activity of their adjacent cells. These features increase the biological relevance of the model compared to other current taste recognition models. The proposed model was implemented in software on a single personal computer and in hardware on an Altera FPGA chip. Both implementations were applied to two real-world taste datasets.In addition, for the first time the applicability of transductive reasoning for forecasting the establishment potential of pest insects into new locations was investigated. For this purpose four types of predictive models, built using inductive and transductive reasoning, were used for predicting the distributions of three pest insects. The models were evaluated in terms of their predictive accuracy and their ability to discover patterns in the modelling data. The results obtained indicate that evolving connectionist systems can be successfully used for building predictive distribution models and environmental monitoring systems. The features available in the proposed dynamic systems, such as on-line learning and knowledge discovery, are needed to improve our knowledge of the species distributions. This work laid down the foundation for a number of interesting future projects in the field of ecological modelling, robotics, pervasive computing and pattern recognition that can be undertaken separately or in sequence.

The objective of this dissertation is to present a detection-based pattern recognition framework and demonstrate its applications in automatic speech recognition and broadcast news video story segmentation. Inspired by the studies of modern cognitive psychology and real-world pattern recognition systems, a detection-based pattern recognition framework is proposed to provide an alternative solution for some complicated pattern recognition problems. The primitive features are first detected and the task-specific knowledge hierarchy is constructed level by level; then a variety of heterogeneous information sources are combined together and the high-level context is incorporated as additional information at certain stages. A detection-based framework is a â divide-and-conquerâ design paradigm for pattern recognition problems, which will decompose a conceptually difficult problem into many elementary sub-problems that can be handled directly and reliably. Some information fusion strategies will be employed to integrate the evidence from a lower level to form the evidence at a higher level. Such a fusion procedure continues until reaching the top level. Generally, a detection-based framework has many advantages: (1) more flexibility in both detector design and fusion strategies, as these two parts can be optimized separately; (2) parallel and distributed computational components in primitive feature detection. In such a component-based framework, any primitive component can be replaced by a new one while other components remain unchanged; (3) incremental information integration; (4) high level context information as additional information sources, which can be combined with bottom-up processing at any stage. This dissertation presents the basic principles, criteria, and techniques for detector design and hypothesis verification based on the statistical detection and decision theory. In addition, evidence fusion strategies were investigated in this dissertation. Several novel detection algorithms and evidence fusion methods were proposed and their effectiveness was justified in automatic speech recognition and broadcast news video segmentation system. We believe such a detection-based framework can be employed in more applications in the future.