M. Tech. Mathematical Technology. / Aims of this research project is to mathematically analyse froth patterns and build a database of the images at different stages of the fermentation process, so that a decision-making procedure can be developed, which enables a computer to react according to what has been observed. This would allow around-the-clock observation which is not possible with humans. In addition, mechanised decision-making would minimize errors usually associated with human actions. Different mathematical algorithms for image processing will be considered and compared. These algorithms have been designed for different image processing situations. In this dissertation the algorithms will be applied to froth images in particular and will be used to simulate the human eye for decision-making in the fermentation process. The preamble of the study will be to consider algorithms for the detection of edges and then analyse these edges. MATLAB will be used to do the pre-processing of the images and to write and test any new algorithms designed for this project.

Image classification is a fundamental information extraction procedure in remote sensing that is used in land-cover and land-use mapping. Despite being considered as a replacement for manual mapping, it still requires some degree of analyst intervention. This makes the process of image classification time consuming, subjective, and error prone. For example, in unsupervised classification, pixels are automatically grouped into classes, but the user has to manually label the classes as one land-cover type or another. As a general rule, the larger the number of classes, the more difficult it is to assign meaningful class labels. A fully automated post-classification procedure for class labeling was developed in an attempt to alleviate this problem. It labels spectral classes by matching their spectral characteristics with reference spectra. A Landsat TM image of an agricultural area was used for performance assessment. The algorithm was used to label a 20- and 100-class image generated by the ISODATA classifier. The 20-class image was used to compare the technique with the traditional manual labeling of classes, and the 100-class image was used to compare it with the Spectral Angle Mapper and Maximum Likelihood classifiers. The proposed technique produced a map that had an overall accuracy of 51%, outperforming the manual labeling (40% to 45% accuracy, depending on the analyst performing the labeling) and the Spectral Angle Mapper classifier (39%), but underperformed compared to the Maximum Likelihood technique (53% to 63%). The newly developed class-labeling algorithm provided better results for alfalfa, beans, corn, grass and sugar beet, whereas canola, corn, fallow, flax, potato, and wheat were identified with similar or lower accuracy, depending on the classifier it was compared with. / vii, 93 leaves : ill., maps (some col.) ; 29 cm

To reduce cost, many digital cameras use a single sensor array instead of using three arrays for the red, green and blue. Thus at each pixel location only the red, green or blue intensity value is available. And to generate a complete color image, the camera must estimate the missing two values at each pixel location .Color filter arrays are used to capture only one portion of the spectrum (Red, Green or Blue) at each location. Various arrangements of the Color Filter Array (CFA) are possible, but the Bayer array is the most commonly used arrangement and we will deal exclusively with the Bayer array in this thesis. Since each of the three colors channels are effectively downsampled, it leads to aliasing artifacts. This thesis will analyze the effects of aliasing in the frequency- domain and present a method to reduce the deterioration in image quality due to aliasing artifacts. Two reference algorithms, AH-POCS (Adams and Hamilton - Projection Onto Convex Sets) and Adaptive Homogeneity-Directed interpolation, are discussed in de- tail. Both algorithms use the assumption that there is high correlation in the high- frequency regions to reduce aliasing. AH-POCS uses alias cancellation technique to reduce aliasing in the red and blue images, while the Adaptive Homogeneity-Directed interpolation algorithm is an edge-directed algorithm. We present here an algorithm that combines these two techniques and provides a better result on average when compared to the reference algorithms.

The main goal of this thesis is to develop real-time computer vision algorithms in order to detect and to track targets in uncertain complex environments purely based on a visual sensor. Two major subjects addressed by this work are: 1. The development of fast and robust image segmentation algorithms that are able to search and automatically detect targets in a given image. 2. The development of sound filtering algorithms to reduce the effects of noise in signals from the image processing. The main constraint of this research is that the algorithms should work in real-time with limited computing power on an onboard computer in an aircraft. In particular, we focus on contour tracking which tracks the outline of the target represented by contours in the image plane. This thesis is concerned with three specific categories, namely image segmentation, shape modeling, and signal filtering. We have designed image segmentation algorithms based on geometric active contours implemented via level set methods. Geometric active contours are deformable contours that automatically track the outlines of objects in images. In this approach, the contour in the image plane is represented as the zero-level set of a higher dimensional function. (One example of the higher dimensional function is a three-dimensional surface for a two-dimensional contour.) This approach handles the topological changes (e.g., merging, splitting) of the contour naturally. Although geometric active contours prevail in many fields of computer vision, they suffer from the high computational costs associated with level set methods. Therefore, simplified versions of level set methods such as fast marching methods are often used in problems of real-time visual tracking. This thesis presents the development of a fast and robust segmentation algorithm based on up-to-date extensions of level set methods and geometric active contours, namely a fast implementation of Chan-Vese's (active contour) model (FICVM). The shape prior is a useful cue in the recognition of the true target. For the contour tracker, the outline of the target can be easily disrupted by noise. In geometric active contours, to cope with deviations from the true outline of the target, a higher dimensional function is constructed based on the shape prior, and the contour tracks the outline of an object by considering the difference between the higher dimensional functions obtained from the shape prior and from a measurement in a given image. The higher dimensional function is often a distance map which requires high computational costs for construction. This thesis focuses on the extraction of shape information from only the zero-level set of the higher dimensional function. This strategy compensates for inaccuracies in the calculation of the shape difference that occur when a simplified higher dimensional function is used. This is named as contour-based shape modeling. Filtering is an essential element in tracking problems because of the presence of noise in system models and measurements. The well-known Kalman filter provides an exact solution only for problems which have linear models and Gaussian distributions (linear/Gaussian problems). For nonlinear/non-Gaussian problems, particle filters have received much attention in recent years. Particle filtering is useful in the approximation of complicated posterior probability distribution functions. However, the computational burden of particle filtering prevents it from performing at full capacity in real-time applications. This thesis concentrates on improving the processing time of particle filtering for real-time applications. In principle, we follow the particle filter in the geometric active contour framework. This thesis proposes an advanced blob tracking scheme in which a blob contains shape prior information of the target. This scheme simplifies the sampling process and quickly suggests the samples which have a high probability of being the target. Only for these samples is the contour tracking algorithm applied to obtain a more detailed state estimate. Curve evolution in the contour tracking is realized by the FICVM. The dissimilarity measure is calculated by the contour based shape modeling method and the shape prior is updated when it satisfies certain conditions. The new particle filter is applied to the problems of low contrast and severe daylight conditions, to cluttered environments, and to the appearing/disappearing target tracking. We have also demonstrated the utility of the filtering algorithm for multiple target tracking in the presence of occlusions. This thesis presents several test results from simulations and flight tests. In these tests, the proposed algorithms demonstrated promising results in varied situations of tracking.

Transmission of medical images, that mandate lossless transmission of content over WLANs, presents a great challenge. The large size of these images coupled with the low acceptance of traditional image compression techniques within the medical community compounds the problem even more. These factors are of enormous significance in a hospital setting in the context of real-time image collaboration. However, recent advances in medical image compression techniques such as diagnostically lossless compression methodology, has made the solution to this difficult problem feasible. The growing popularity of high speed wireless LAN in enterprise applications and the introduction of the new 802.11n draft standard have made this problem pertinent. The thesis makes recommendations on the degree of compression to be performed for specific instances of image communication applications based on the image size and the underlying network devices and their topology. During our analysis, it was found that for most cases, only a portion of the image; typically the region of interest of the image will be able to meet the time deadline requirement. This dictates a need for adaptive method for maximizing the percentage of the image delivered to the receiver within the deadline. The problem of maximizing delivery of regions of interest of image data within the deadline has been effectively modeled as a multi-commodity flow problem in this work. Though this model provides an optimal solution to the problem, it is NP hard in computational complexity and hence cannot be implemented in dynamic networks. An approximation algorithm that uses greedy approach to flow allocation is proposed to cater to the connection requests in real time. While implementing integer programming model is not feasible due to time constraints, the heuristic can be used to provide a near-optimal solution for the problem of maximizing the reliable delivery of regions of interest of medical images within delay deadlines. This scenario may typically be expected when new connection requests are placed after the initial flow allocations have been made.

With the advance of the LCD technology, video quality is becoming increasingly important. In this thesis, we develop hardware-friendly low-complexity enhancement algorithms. Video quality enhancement methods can be classified into two main categories. Single frame methods are the first category. These methods have generally low computational complexity. Multi-frame methods combine information from more than one frame and require the motion information of objects in the scene to do so. We first concentrate on the contrast-enhancement problem by using both global (frame-wise) and local information derived from the image. We use the image histogram and present a regularization-based histogram modification method to avoid problems that are often created by histogram equalization. Next, we design a compression artifact reduction algorithm that reduces ringing artifacts, which is disturbing especially on large displays. Furthermore, to remove the blurriness in the original video we present a non-iterative diffusion-based sharpening algorithm, which enhances edges in a ringing-aware fashion. The diffusion-based technique works on gradient approximations in a neighborhood individually. This gives more freedom compared to modulating the high-pass filter output that is used to sharpen the edges. Motion estimation enables applications such as motion-compensated noise reduction, frame-rate conversion, de-interlacing, compression, and super-resolution. Motion estimation is an ill-posed problem and therefore requires the use of prior knowledge on motion of objects. Objects have inertia and are usually larger then pixels or a block of pixels in size, which creates spatio-temporal correlation. We design a method that uses temporal redundancy to improve motion-vector search by choosing bias vectors from the previous frame and adaptively penalizes deviations from the bias vectors. This increases the robustness of the motion-vector search. The spatial correlation is more reliable because temporal correlation is difficult to use when the objects move fast or accelerate in time, or have small sizes. Spatial smoothness is not valid across motion boundaries. We investigate using energy minimization for motion estimation and incorporate the spatial smoothness prior into the energy. By formulating the energy minimization iterations for each motion vector as the primal problem, we show that the dual problem is motion segmentation for that specific motion vector.

The electrocardiograph (ECG) is a common clinical and biomedical research tool used for both diagnostic and prognostic purposes. In recent years computer aided analysis of the ECG has enabled cardiographic patterns to be found which were hitherto not apparent. Many of these analyses rely upon the segmentation of the ECG into separate time delimited waveforms. The instants delimiting these segments are called the

Thesis (MSc (Mathematical Sciences)--University of Stellenbosch, 2005. / Image compression using weighted finite automata (WFA) is studied and implemented in Matlab. Other more prominent image compression techniques, namely JPEG, vector quantization, EZW wavelet image compression and fractal image compression are also presented. The performance of WFA image compression is then compared to those of some of the abovementioned techniques.

O padrão de codificação MPEG contém algoritmos que analisam a redundância temporal e espacial do vídeo. Devido a esta característica, grandes variações na taxa de dados são observadas numa sequência de vídeo. Deste modo, os sistemas multimídia apresentam dificuldades no planejamento, projeto e utilização dos recursos necessários para a reprodução adequada do vídeo. A proposta deste trabalho é avaliar o uso das abordagens "Network Calculus" (NC) e "Variability Characterization Curve" (VCC) na caracterização de vídeos MPEG, fornecendo informações úteis para a composição de sistemas multimídia. Usualmente, o NC é uma teoria utilizada para se obter limitantes de desempenho em redes de pacotes, enquanto o VCC apresenta técnicas para se obter limitantes inferior e superior de desempenho de um sistema computacional. Neste trabalho, porém, tanto o NC quanto o VCC são utilizados para definir limitantes de desempenho na forma de parâmetros a serem usados na caracterização de vídeo. Além disso, uma variação do VCC usando curvas aproximadas é utilizada neste trabalho para avaliação da economia de recursos. Adicionalmente aos métodos NC e VCC, é proposto um novo método para agregar informação ao conjunto de parâmetros utilizados na caracterização de vídeo. De forma a avaliar os métodos em diversas condições, são realizados testes com vídeos em diferentes configurações e conteúdos, usando os padrões MPEG-2, MPEG-4, H264/AVC. O resultado da aplicação destes métodos é um conjunto de parâmetros que podem ser utilizados para caracterizar os vídeos e indicar as particularidades e exigências que cada vídeo impõe aos sistemas multimídia. Além disso, uma ferramenta computacional para avaliação desta caracterização é proposta e implementada. / The MPEG video coding standard has algorithms for analyzing temporal and spatial video’s redundancies. Therefore, a great variation of data rate is observed in a video sequence. Consequently, planning, design and utilization of the necessary resources for playing video is not an easy task in multimedia systems. This work evaluates the application of Network Calculus (NC) and Variability Characterization Curve (VCC) approaches in MPEG video characterization by providing useful information to setup multimedia systems. Usually, NC theory is applied to obtain performance bounds for packet networks while VCC provides techniques to obtain lower and upper performance bounds for a computational system. However in this work, both NC and VCC are applied to define performance limits expressed as parameters to be used in this work to evaluate resource savings. In addition, a new method is proposed by adding information to the set of parameters used for video characterization. For evaluating these methods in several conditions, a set of tests in accomplished in different configurations with videos MPEG-2, MPEG-4 and H264/AVC. The result is a set of parameters that can be used to characterize videos and point out particularities and requirements imposed by each video to multimedia systems. Finally, a computational tool for evaluating this characterization is also proposed and implemented.

UOL; CAPES / Esta tese apresenta um método para a extração de regiões de interesse (ROIs) de imagens. No contexto deste trabalho, ROIs são definidas como os objetos semânticos que se destacam em uma imagem, podendo apresentar qualquer tamanho ou localização. O novo método baseia-se em modelos computacionais de atenção visual (VA), opera de forma completamente bottom-up, não supervisionada e não apresenta restrições com relação à categoria da imagem de entrada. Os elementos centrais da arquitetura são os modelos de VA propostos por Itti-Koch-Niebur e Stentiford. O modelo de Itti-Koch-Niebur considera as características de cor, intensidade e orientação da imagem e apresenta uma resposta na forma de coordenadas, correspondentes aos pontos de atenção (POAs) da imagem. O modelo Stentiford considera apenas as características de cor e apresenta a resposta na forma de áreas de atenção na imagem (AOAs). Na arquitetura proposta, a combinação de POAs e AOAs permite a obtenção dos contornos das ROIs. Duas implementações desta arquitetura, denominadas 'primeira versão' e 'versão melhorada' são apresentadas. A primeira versão utiliza principalmente operações tradicionais de morfologia matemática. Esta versão foi aplicada em dois sistemas de recuperação de imagens com base em regiões. No primeiro, as imagens são agrupadas de acordo com as ROIs, ao invés das características globais da imagem. O resultado são grupos de imagens mais significativos semanticamente, uma vez que o critério utilizado são os objetos da mesma categoria contidos nas imagens. No segundo sistema, á apresentada uma combinação da busca de imagens tradicional, baseada nas características globais da imagem, com a busca de imagens baseada em regiões. Ainda neste sistema, as buscas são especificadas através de mais de uma imagem exemplo. Na versão melhorada da arquitetura, os estágios principais são uma análise de coerência espacial entre as representações de ambos modelos de VA e uma representação multi-escala das AOAs. Se comparada à primeira versão, esta apresenta maior versatilidade, especialmente com relação aos tamanhos das ROIs presentes nas imagens. A versão melhorada foi avaliada diretamente, com uma ampla variedade de imagens diferentes bancos de imagens públicos, com padrões-ouro na forma de bounding boxes e de contornos reais dos objetos. As métricas utilizadas na avaliação foram presision, recall, F1 e area of overlap. Os resultados finais são excelentes, considerando-se a abordagem exclusivamente bottom-up e não-supervisionada do método. / This thesis presents a method for the extraction of regions of interest (ROIs) from images. By ROIs we mean the most prominent semantic objects in the images, of any size and located at any position in the image. The novel method is based on computational models of visual attention (VA), operates under a completely bottom-up and unsupervised way and does not present con-straints in the category of the input images. At the core of the architecture is de model VA proposed by Itti, Koch and Niebur and the one proposed by Stentiford. The first model takes into account color, intensity, and orientation features and provides coordinates corresponding to the points of attention (POAs) in the image. The second model considers color features and provides rough areas of attention (AOAs) in the image. In the proposed architecture, the POAs and AOAs are combined to establish the contours of the ROIs. Two implementations of this architecture are presented, namely 'first version' and 'improved version'. The first version mainly on traditional morphological operations and was applied in two novel region-based image retrieval systems. In the first one, images are clustered on the basis of the ROIs, instead of the global characteristics of the image. This provides a meaningful organization of the database images, since the output clusters tend to contain objects belonging to the same category. In the second system, we present a combination of the traditional global-based with region-based image retrieval under a multiple-example query scheme. In the improved version of the architecture, the main stages are a spatial coherence analysis between both VA models and a multiscale representation of the AOAs. Comparing to the first one, the improved version presents more versatility, mainly in terms of the size of the extracted ROIs. The improved version was directly evaluated for a wide variety of images from different publicly available databases, with ground truth in the form of bounding boxes and true object contours. The performance measures used were precision, recall, F1 and area overlap. Experimental results are of very high quality, particularly if one takes into account the bottom-up and unsupervised nature of the approach.