Measurement of visual quality is crucial for various image and video processing applications. It is widely applied in image acquisition, media transmission, video compression, image/video restoration, etc. The goal of image quality assessment (QA) is to develop a computable quality metric which is able to properly evaluate image quality. The primary criterion is better QA consistency with human judgment. Computational complexity and resource limitations are also concerns in a successful QA design. Many methods have been proposed up to now. At the beginning, quality measurements were directly taken from simple distance measurements, which refer to mathematically signal fidelity, such as mean squared error or Minkowsky distance. Lately, QA was extended to color space and the Fourier domain in which images are better represented. Some existing methods also consider the adaptive ability of human vision. Unfortunately, the Video Quality Experts Group indicated that none of the more sophisticated metrics showed any great advantage over other existing metrics. This thesis proposes a general approach to the QA problem by evaluating image information entropy. An information theoretic model for the human visual system is proposed and an information theoretic solution is presented to derive the proper settings. The quality metric is validated by five subjective databases from different research labs. The key points for a successful quality metric are investigated. During the testing, our quality metric exhibits excellent consistency with the human judgments and compatibility with different databases. Other than full reference quality assessment metric, blind quality assessment metrics are also proposed. In order to predict quality without a reference image, two concepts are introduced which quantitatively describe the inter-scale dependency under a multi-resolution framework. Based on the success of the full reference quality metric, several blind quality metrics are proposed for five different types of distortions in the subjective databases. Our blind metrics outperform all existing blind metrics and also are able to deal with some distortions which have not been investigated.

Image noise reduction, or denoising, is an active area of research, although many of the techniques cited in the literature mainly target additive white noise. With an emphasis on signal-dependent noise, this thesis presents the General Adaptive Monte Carlo Bayesian Image Denoising (GAMBID) algorithm, a model-free approach based on random sampling. Testing is conducted on synthetic images with two different signal-dependent noise types as well as on real synthetic aperture radar and ultrasound images. Results show that GAMBID can achieve state-of-the-art performance, but suffers from some limitations in dealing with textures and fine low-contrast features. These aspects can by addressed in future iterations when GAMBID is expanded to become a versatile denoising framework.

Digital image acquisition can be a time consuming process for situations where high spatial resolution is required. As such, optimizing the acquisition mechanism is of high importance for many measurement applications. Acquiring such data through a dynamically small subset of measurement locations can address this problem. In such a case, the measured information can be regarded as incomplete, which necessitates the application of special reconstruction tools to recover the original data set. The reconstruction can be performed based on the concept of sparse signal representation. Recovering signals and images from their sub-Nyquist measurements forms the core idea of compressive sensing (CS). In this work, a CS-based data-guided statistical sparse measurements method is presented, implemented and evaluated. This method significantly improves image reconstruction from sparse measurements. In the data-guided statistical sparse measurements approach, signal sampling distribution is optimized for improving image reconstruction performance. The sampling distribution is based on underlying data rather than the commonly used uniform random distribution. The optimal sampling pattern probability is accomplished by learning process through two methods - direct and indirect. The direct method is implemented for learning a nonparametric probability density function directly from the dataset. The indirect learning method is implemented for cases where a mapping between extracted features and the probability density function is required. The unified model is implemented for different representation domains, including frequency domain and spatial domain. Experiments were performed for multiple applications such as optical coherence tomography, bridge structure vibration, robotic vision, 3D laser range measurements and fluorescence microscopy. Results show that the data-guided statistical sparse measurements method significantly outperforms the conventional CS reconstruction performance. Data-guided statistical sparse measurements method achieves much higher reconstruction signal-to-noise ratio for the same compression rate as the conventional CS. Alternatively, Data-guided statistical sparse measurements method achieves similar reconstruction signal-to-noise ratio as the conventional CS with significantly fewer samples.

Evolutionary algorithms (EAs) are well-established techniques to approach those problems which for the classical optimization methods are difficult to solve. Tackling problems with mixed-type of variables, many local optima, undifferentiable or non-analytical functions are some examples to highlight the outstanding capabilities of the evolutionary algorithms. Among the various kinds of evolutionary algorithms, differential evolution (DE) is well known for its effectiveness and robustness. Many comparative studies confirm that the DE outperforms many other optimizers. Finding more accurate solution(s), in a shorter period of time for complex black-box problems, is still the main goal of all evolutionary algorithms. The opposition concept, on the other hand, has a very old history in philosophy, set theory, politics, sociology, and physics. But, there has not been any opposition-based contribution to optimization. In this thesis, firstly, the opposition-based optimization (OBO) is constituted. Secondly, its advantages are formally supported by establishing mathematical proofs. Thirdly, the opposition-based acceleration schemes, including opposition-based population initialization and generation jumping, are proposed. Fourthly, DE is selected as a parent algorithm to verify the acceleration effects of proposed schemes. Finally, a comprehensive set of well-known complex benchmark functions is employed to experimentally compare and analyze the algorithms. Results confirm that opposition-based DE (ODE) performs better than its parent (DE), in terms of both convergence speed and solution quality. The main claim of this thesis is not defeating DE, its numerous versions, or other optimizers, but to introduce a new notion into nonlinear continuous optimization via innovative metaheuristics, namely the notion of opposition. Although, ODE has been compared with six other optimizers and outperforms them overall. Furthermore, both presented experimental and mathematical results conform with each other and demonstrate that opposite points are more beneficial than pure random points for black-box problems; this fundamental knowledge can serve to accelerate other machine learning approaches as well (such as reinforcement learning and neural networks). And perhaps in future, it could replace the pure randomness with random-opposition model when there is no a priori knowledge about the solution/problem. Although, all conducted experiments utilize DE as a parent algorithm, the proposed schemes are defined at the population level and, hence, have an inherent potential to be utilized for acceleration of other DE extensions or even other population-based algorithms, such as genetic algorithms (GAs). Like many other newly introduced concepts, ODE and the proposed opposition-based schemes still require further studies to fully unravel their benefits, weaknesses, and limitations.

Investigations into past accidents at nuclear power generating facilities such as that of Three Mile Island have identified human factors as one of the foremost critical aspects in plant safety. Errors resulting from limitations in human information processing are of particular concern for human-machine interfaces (HMI) in plant control rooms. This project examines the application of Ecological Interface Design (EID) in HMI information displays and the effects on operator situation awareness (SA) for turbine secondary systems based on the Swedish Forsmark 3 boiling-water reactor nuclear power plant. A work domain analysis was performed on the turbine secondary systems yielding part-whole decomposition and abstraction hierarchy models. Information display requirements were subsequently extracted from the models. The resulting EID information displays were implemented in a full-scope simulator and evaluated with six licensed operating crews from the Forsmark 3 plant. Three measures were used to examine SA: self-rated bias, Halden Open Probe Elicitation (HOPE), and Situation Awareness Control Room Inventory (SACRI). The data analysis revealed that operators achieved moderate to good SA; operators unfamiliar with EID information displays were able to develop and maintain comparable levels of SA to operators using traditional forms of single sensor-single indicator (SS-SI) information displays. With sufficient training and experience, operator SA is expected to benefit from the knowledge-based visual elements in the EID information displays. This project was researched in conjunction with the Cognitive Engineering Laboratory at the University of Toronto and the Institute for Energy Technology (IFE) in Halden, Norway.

Multiple classifier system has shown to be an effective technique for classification. The success of multiple classifiers does not entirely depend on the base classifiers and/or the aggregation technique. Other parameters, such as training data, feature attributes, and correlation among the base classifiers may also contribute to the success of multiple classifiers. In addition, interaction of these parameters with each other may have an impact on multiple classifiers performance. In the present study, we intended to examine some of these interactions and investigate further the effects of these interactions on the performance of classifier ensembles. The proposed research introduces a different direction in the field of multiple classifiers systems. We attempt to understand and compare ensemble methods from the cooperation perspective. In this thesis, we narrowed down our focus on cooperation at training level. We first developed measures to estimate the degree and type of cooperation among training data partitions. These evaluation measures enabled us to evaluate the diversity and correlation among a set of disjoint and overlapped partitions. With the aid of properly selected measures and training information, we proposed two new data partitioning approaches: Cluster, De-cluster, and Selection (CDS) and Cooperative Cluster, De-cluster, and Selection (CO-CDS). In the end, a comprehensive comparative study was conducted where we compared our proposed training approaches with several other approaches in terms of robustness of their usage, resultant classification accuracy and classification stability. Experimental assessment of CDS and CO-CDS training approaches validates their robustness as compared to other training approaches. In addition, this study suggests that: 1) cooperation is generally beneficial and 2) classifier ensembles that cooperate through sharing information have higher generalization ability compared to the ones that do not share training information.

Flexible approaches for eliciting preferences of decision makers involved in a conflict are developed along with applications to real-world disputes. More specifically, two multiple criteria decision making approaches are proposed for capturing the relative preferences of a decision maker participating in a conflict situation. A case study in logistics concerned with the conflict arising over the expansion of port facilities on the west coast of North America as well as a transportation negotiation dispute are used to illustrate how these approaches can be integrated with the Graph Model for Conflict Resolution, a practical conflict analysis methodology. Ascertaining the preferences of the decision makers taking part in a conflict constitutes a key element in the construction of a formal conflict model. In practice, the relative preferences, which reflect each decision maker’s objectives or goals in a given situation, are rather difficult to obtain. The first method for preference elicitation is to integrate an Analytic Hierarchy Process (AHP) preference ranking method with the Graph Model for Conflict Resolution. The AHP approach is used to elicit relative preferences of decision makers, and this preference information is then fed into a graph model for further stability analyses. The case study of the Canadian west coast port congestion conflict is investigated using this integrated model. Another approach is based on a fuzzy multiple criteria out-ranking technique called ELECTRE III. It is also employed for ranking states or possible scenarios in a conflict from most to least preferred, with ties allowed, by the decision maker according to his or her own value system. The model is applied to a transportation negotiation dispute between the two key parties consisting of shippers and carriers.

This research focuses on developing new techniques and designs for highly con- trollable microactuating systems with large force-stroke outputs. A fixed-fixed mi- crobeam is the actuating element in the introduced techniques. Either buckling of a microbridge by thermal stress, lateral deflection of a microbridge by electro- magnetic force, or combined effects of both can be employed for microactuation. The proposed method here is MicroElectroThermoMagnetic Actuation (METMA), which uses the combined techniques of electrical or electro-thermal driving of a mi- crobridge in the presence of a magnetic field. The electrically controllable magnetic field actuates and controls the electrically or electrothermally driven microstruc- tures. METMA provides control with two electrical inputs, the currents driving the microbridge and the current driving the external magnetic field. This method enables a more controllable actuating system. Different designs of microactuators have been implemented by using MEMS Pro as the design software and MUMPs as the standard MEMS fabrication technology. In these designs, a variety of out-of- plane buckling or displacement of fixed-fixed microbeams have been developed and employed as the actuating elements. This paper also introduces a novel actuating technique for larger displacements that uses a two-layer buckling microbridge actu- ated by METMA. Heat transfer principles are applied to investigate temperature distribution in a microbeam, electrothermal heating, and the resulting thermoelas- tic effects. Furthermore, a method for driving microactuators by applying powerful electrical pulses is proposed. The integrated electromagnetic and electrothermal microactuation technique is also studied. A clamped-clamped microbeam carry- ing electrical current has been modeled and simulated in ANSYS. The simulations include electrothermal, thermoelastic, electromagnetic, and electrothermomagnetic effects. The contributions are highlighted, the results are discussed, the research and design limitations are reported, and future works are proposed.

In recent years, carbon nanotubes (CNTs) have emerged as one of the best field emitters for a variety of technological applications. The field emitting cathodes have several advantages over the conventional thermionic cathodes: (i) current density from field emission would be orders of magnitude greater than in the thermionic case, (ii) a cold cathode would minimize the need for cooling, and (iii) a field emitting cathode can be miniaturized. In spite of good performance of such cathodes, the procedure to estimate the device current is not straight forward and the required insight towards design optimization is not well understood. In addition, the current in CNT-based thin film devices shows fluctuation. Such fluctuation in field emission current is not desirable for many biomedical applications such as x-ray devices. The CNTs in a thin film undergo complex dynamics during field emission, which includes processes such as (i) evolution, (ii) electromechanical interaction, (iii) thermoelectric heating, (iv) ballistic transport, and (v) electron gas flow. These processes are coupled and nonlinear. Therefore, they must be analyzed accurately from the stability and long-term performance point of view. In this research, we develop detailed physics-based models of CNTs considering the aspects mentioned above. The models are integrated in a systematic manner to calculate the device current by using the Fowler-Nordheim equation. Using the models, we were able to capture the fluctuations in the field emission current, which have been observed in actual experiments. A detailed analysis of the results reveals the deflected shapes of the CNTs in an ensemble and the extent to which the initial state of geometry and orientation angles affect the device current. In addtion, investigations on the influence of defects and impurities in CNTs on their field emission properties have been carried out. By inclusion of defects and impurities, the field emission properties of CNTs can be tailored for specific device applications in future. For stable performance of CNT-based field emission devices, such as x-ray generators, design optimization studies have been presented. It has been found that the proposed design minimizes transience in field emission current. In this thesis, it has been demonstrated that phonon-assisted control of field emission current in CNT based thin film is possible. Finally, experimental studies pertaining to crosstalk phenomenon in a multi-pixel CNT array are presented.

There has been a growth in demand for advancing algorithms in surveillance applications concerning moving vehicles where analysis of traffic has a potential application to security, traffic management (congestion and accident detection), speed measurement, car counting and statistics, as well as turning movement at intersections. This research focuses on multiple-vehicle detection, recognition, and tracking in urban environments based on video sequences obtained from a single CCD camera mounted on a pole at urban highways and crossroads. The proposed system integrates several modules including segmentation, object detection, object recognition and classification, and tracking. Background segmentation, based on Gaussian Mixture models, is used to extract moving objects from images using the respective foreground object information such as location, size, and color distribution. To recognize vehicles, a 3D polyhedral car model described by a set of parameters is built and mapped to the 2D edge information attained from the video sequence. The matching process is then used to classify the foreground object obtained into vehicles and non-vehicles. The output from the recognition model is used in tracking multiple cars based on a deterministic data association method that takes place between consecutive frame information. The multiple-vehicle surveillance system developed in this thesis, based on integrating different modules, provides a novel approach for vehicle monitoring. Furthermore, the system makes use of minimal a priori knowledge about vehicle location, size, type, numbers, and pathways. The system implemented in this work functions well under various camera perspectives, background clutter, vehicle viewpoints, road types, scale changes, image noise, image resolutions, and lighting conditions.