Fusion of multiple stereo algorithms is performed in order to obtain accurate disparity segmentation. Reliable disparity map of real-time stereo images is estimated and disparity segmentation is performed for object detection purpose. First, stereo algorithms which have high performance in real-time applications are chosen among the algorithms in the literature and three of them are implemented. Then, the results of these algorithms are fused to gain better performance in disparity estimation. In fusion process, if a pixel has the same disparity value in all algorithms, that disparity value is assigned to the pixel. Other pixels are labelled as unknown disparity. Then, unknown disparity values are estimated by a refinement procedure where neighbourhood disparity information is used. Finally, the resultant disparity map is segmented by using mean shift segmentation. The proposed method is tested in three different stereo data sets and several real stereo pairs. The experimental results indicate an improvement for the stereo analysis performance by the usage of fusion process and refinement procedure. Furthermore, disparity segmentation is realized successfully by using mean shift segmentation for detecting objects at different depth levels.

No description available.

No description available.

The master thesis focuses on disparity map estimation using convolutional neural network. It discusses the problem of using convolutional neural networks for image comparison and disparity computation from stereo image as well as existing approaches of solutions for given problem. It also proposes and implements system that consists of convolutional neural network that measures the similarity between two image patches, and filtering and smoothing methods to improve the result disparity map. Experiments and results show, that the most quality disparity maps are computed using CNN on input patches with the size of 9x9 pixels combined with matching cost agregation and correction algorithm and bilateral filter.

This dissertation investigates the determinants of economic growth in China since 1978, with a focus on the determinants of spatial growth. A study of the theories of economic growth shows that both proximate and fundamental factors can contribute to economic growth. In the case of China, institutional changes are the keys to the Chinese transitional economy. Given the special nature of China's economy, the main institutional reforms since 1978 are examined, together with the gradual transition process. Furthermore, from the overview of empirical literature, it is found that the proximate determinants such as initial gross domestic product (GDP), investment, population growth, human capital and openness are determinants of economic growth in China based on the findings in cross-country growth literature. From growth accounting exercises, capital formation and total factor productivity (TFP) growth can be seen to play important roles in the rapid economic growth in China. However, while the nationwide economic growth is impressive, the pace of reform and economic development has been uneven across provinces. In the existing literature, geography and preferential policy are emphasised as particular factors that affect coastal-interior disparity. This study incorporates the economic variables identified as important stimulants to growth, drawing on major findings in the study of convergence and economic growth to estimate the determinants of regional economic growth in China. To address the weaknesses of using ordinary least squares (OLS) for cross-country regression analyses, fixed-effects ordinary least squares (OLS) and random-effects generalised least squares (GLS) panel data estimators are applied to provincial data from 1994 to 2003. It is concluded that the convergence hypothesis does not hold in China, and that export, investment, education, foreign direct investment (FDI) growth and coastal dummy have a positive effect on regional GDP per capita growth in China while population growth affects the annual growth rate negatively. / Thesis (M.Com. (Economics))--North-West University, Potchefstroom Campus, 2006.

No description available.

Walking can be a safe, healthy, and affordable mode of school transportation. However, most students today do not use walking for their school travel. More research is needed to understand the correlates of walking to or from school and to identify effective interventions. This is a cross-sectional study of 73 public elementary schools in the Austin Independent School District of Texas. The first phase used geographic information systems and field audits to examine school-level disparities in the environmental support for walking in schools’ attendance areas. The second phase involved surveys of students’ parents or guardians to identify the multi-level correlates of using walking as their children’s typical school travel mode. In the first phase, results from analyses of variance and linear regressions indicated the existence of disparities. Lower economic status of student population was associated with poorer street conditions (e.g., maintenance, visual quality, amenities, and perceived safety), shorter distances to school, and lower traffic volumes. Higher percentage of Hispanic students within a school was associated with increased danger from traffic and crime and more sidewalks, greater population density, and mixed land uses. The second phase used binary logistic regressions to predict walking to or from school. Among the personal and social factors, parents’ education, car ownership, personal barriers, and school bus availability were negative correlates, while parents’ and children’s positive attitude and regular walking habit and supportive peer influences were positive correlates. Of the physical environmental factors, long distance and safety concerns were the strongest negative correlates, followed by the presence of highways or freeways, convenience stores, office buildings, and bus stops en route. In conclusion, environmental interventions are needed to develop centrallylocated neighborhood schools, barrier-free attendance areas, and well-maintained pedestrian infrastructure. Disparities and fine-grained differences are found in the environmental support for walking. A high priority for low-income, Hispanic children and interventions tailored for specific contexts and populations appear necessary. Safety improvement is indispensible in terms of both traffic and crime and should be supplemented with educational programs that target both parents and children. Finally, multi-agency collaborations are needed at the policy level to support and facilitate these multi-level interventions.

In this dissertation, we conducted a stereoscopic eye tracking experiment using naturalistic stereo images. We analyzed low level 2D and 3D scene features at binocular fixations and randomly selected places. The results reveal that humans tend to fixate on regions with higher luminance variations, but lower disparity variations. Because of the often observed co-occurrence of luminance and depth changes in natural environments, the dichotomy between luminance features and disparity features inspired us to study the accurate statistics of 2D and 3D scene properties. Using a range map database, we studied the distribution of disparity in natural scenes. The natural disparity distribution has a high peak at zero, and heavier tails that are similar to a Laplace distribution. The relevance of natural disparity distribution to other studies in neurobiology and visual psychophysics are discussed in detail. We also studied luminance, range and disparity statistics in natural scenes using a co-registered luminance-range database. The distributions of bandpass 2D and 3D scene features can be well modeled by generalized Gaussian models. There are positive correlations between bandpass luminance and depth, which can be captured by varying shape parameters in the probability density functions of the generalized Gaussians. In another study on suprathreshold luminance and depth discontinuities, we show that observing a significant luminance edge at a significant depth edge is much more likely than at homogeneous depth surfaces. It is also true that a significant depth edge happens at a significant luminance edge with a greater probability than at homogeneous luminance regions. Again, the dependency between luminance and depth discontinuities can be modeled successfully by generalized Gaussians. We applied our statistical models in 3D natural scenes to stereo correspondence. A Bayesian framework is proposed to incorporate the bandpass disparity prior, and the luminance-disparity dependency in the likelihood function. We compared our algorithm with a classical simulated annealing method based on heuristically defined energy functions. The computed disparity maps show great improvements both perceptually and objectively. / text

No description available.

This dissertation investigates the determinants of economic growth in China since 1978, with a focus on the determinants of spatial growth. A study of the theories of economic growth shows that both proximate and fundamental factors can contribute to economic growth. In the case of China, institutional changes are the keys to the Chinese transitional economy. Given the special nature of China's economy, the main institutional reforms since 1978 are examined, together with the gradual transition process. Furthermore, from the overview of empirical literature, it is found that the proximate determinants such as initial gross domestic product (GDP), investment, population growth, human capital and openness are determinants of economic growth in China based on the findings in cross-country growth literature. From growth accounting exercises, capital formation and total factor productivity (TFP) growth can be seen to play important roles in the rapid economic growth in China. However, while the nationwide economic growth is impressive, the pace of reform and economic development has been uneven across provinces. In the existing literature, geography and preferential policy are emphasised as particular factors that affect coastal-interior disparity. This study incorporates the economic variables identified as important stimulants to growth, drawing on major findings in the study of convergence and economic growth to estimate the determinants of regional economic growth in China. To address the weaknesses of using ordinary least squares (OLS) for cross-country regression analyses, fixed-effects ordinary least squares (OLS) and random-effects generalised least squares (GLS) panel data estimators are applied to provincial data from 1994 to 2003. It is concluded that the convergence hypothesis does not hold in China, and that export, investment, education, foreign direct investment (FDI) growth and coastal dummy have a positive effect on regional GDP per capita growth in China while population growth affects the annual growth rate negatively. / Thesis (M.Com. (Economics))--North-West University, Potchefstroom Campus, 2006.