Fetal Alcohol Spectrum Disorder (FASD) is the most common form of preventable mental retardation worldwide. This condition affects children whose mothers excessively consume alcohol whilst pregnant. FASD can be identied by physical and mental defects, such as stunted growth, facial deformities, cognitive impairment, and behavioural abnormalities. Magnetic Resonance Imaging provides a non-invasive means to study the neural correlates of FASD. One such approach aims to detect brain abnormalities through an assessment of volume and shape of sub-cortical structures on high-resolution MR images. Two brain structures of interest are the Caudate Nucleus and Hippocampus. Manual segmentation of these structures is time-consuming and subjective. We therefore present a method for automatically segmenting the Caudate Nucleus and Hippocampus from high-resolution MR images captured as part of an ongoing study into the neural correlates of FASD. Our method incorporates an Active Shape Model (ASM), which is used to learn shape variation from manually segmented training data. A discrete Geometrically Deformable Model (GDM) is rst deformed to t the relevant structure in each training set. The vertices belonging to each GDM are then used as 3D landmark points - effectively generating point correspondence between training models. An ASM is then created from the landmark points. This ASM is only able to deform to t structures with similar shape to those found in the training data. There are many variations of the standard ASM technique - each suited to the segmentation of data with particular characteristics. Experiments were conducted on the image search phase of ASM segmentation, in order to find the technique best suited to segmentation of the research data. Various popular image search techniques were tested, including an edge detection method and a method based on grey prole Mahalanobis distance measurement. A heuristic image search method, especially designed to target Caudate Nuclei and Hippocampi, was also developed and tested. This method was extended to include multisampling of voxel proles. ASM segmentation quality was evaluated according to various quantitative metrics, including: overlap, false positives, false negatives, mean squared distance and Hausdorff distance. Results show that ASMs that use the heuristic image search technique, without multisampling, produce the most accurate segmentations. Mean overlap for segmentation of the various target structures ranged from 0.76 to 0.82. Mean squared distance ranged from 0.72 to 0.76 - indicating sub-1mm accuracy, on average. Mean Hausdorff distance ranged from 2:7mm to 3:1mm. An ASM constructed using our heuristic technique will enable researchers to quickly, reliably, and automatically segment test data for use in the FASD study - thereby facilitating a better understanding of the eects of this unfortunate condition.

The 3D shape and position of objects inside the human body are commonly detected using Computed Tomography (CT) scanning. CT is an expensive diagnostic option in economically disadvantaged areas and the radiation dose experienced by the patient is significant. In this dissertation, we present a technique for reconstructing the 3D shape and position of bullets from multiple X-rays. This technique makes us of ubiquitous X-ray equipment and a small number of X-rays to reduce the radiation dose. Our work relies on Image Segmentation and Volume Reconstruction techniques. We present a method for segmenting bullets out of X-rays, based on their signature in intensity profiles. This signature takes the form of a distinct plateau which we model with a number of parameters. This model is used to identify horizontal and vertical line segments within an X-Ray corresponding to a bullet signature. Regions containing confluences of these line segments are selected as bullet candidates. The actual bullet is thresholded out of the region based on a range of intensities occupied by the intensity profiles that contributed to the region. A simple Volume Reconstruction algorithm is implemented that back-projects the silhouettes of bullets obtained from our segmentation technique. This algorithm operates on a 3D voxel volume represented as an octree. The reconstruction is reduced to the 2D case by reconstructing a slice of the voxel volume at a time. We achieve good results for our segmentation algorithm. When compared with a manual segmentation, our algorithm matches 90% of the bullet pixels in nine of the twelve test X-rays. Our reconstruction algorithm produces an acceptable results: It achieves a 70% match for a test case where we compare a simulated bullet with a reconstructed bullet.

World wide focus has over the years been shifting towards security issues, not in least due to recent world wide terrorist activities. Several researchers have proposed state of the art surveillance systems to help with some of the security issues with varying success. Recent studies have suggested that the ability of these surveillance systems to learn common environmental behaviour patterns as wells as to detect and predict unusual, or anomalous, activities based on those learnt patterns are possible improvements to those systems. In addition, some of these surveillance systems are still run by human operators, who are prone to mistakes and may need some help from the surveillance systems themselves in detection of anomalous activities. This dissertation attempts to address these suggestions by combining the fields of Image Understanding and Artificial Intelligence, specifically Bayesian Networks, to develop a prototype video surveillance system that can learn common environmental behaviour patterns, thus being able to detect and predict anomalous activity in the environment based on those learnt patterns. In addition, this dissertation aims to show how the prototype system can adapt to these anomalous behaviours and integrate them into its common patterns over a prolonged occurrence period. The prototype video surveillance system showed good performance and ability to detect, predict and integrate anomalous activity in the evaluation tests that were performed using a volunteer in an experimental indoor environment. In addition, the prototype system performed quite well on the PETS 2002 dataset 1, which it was not designed for. The evaluation procedure used some of the evaluation metrics commonly used on the PETS datasets. Hence, the prototype system provides a good approach to anomaly detection and prediction using Bayesian Networks trained on common environmental activities.