Browsing by Author "Corona, Enrique"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Pyramidal stereo matching and optimal surface recovery for 3-D visualization(Texas Tech University, 2002-08) Corona, EnriqueThree-dimensional surface recovery based on a pair of stereoscopic images is a very well-known ill-posed problem with solutions depending mainly on the correct measures of the shifts between corresponding points (disparities) in the images acquired by a known imaging system. Noise, occlusions, and distortion present in the pair of images make the task of finding precise disparities difficult and very time consuming. This work presents a three-dimensional surface restoration method based on the recovery of the optimum surface within a 3-D cross-correlation coefficient volume via a two-stage dynamic programming technique. This procedure is applied to a set of optic nerve head (ONH) images, which are used for finding clinical measures of progression of glaucoma. Registration of these types of images is performed through a two-step coarse-to- fine procedure using power cepstrum and cross-correlation operations, while a local registration based on the weighted mean of second-degree polynomials is used for image fitting. Variations in topography of the ONH can be measured through cup-to-disc ratios which are computed from the 3-D surface generated from longitudinal stereo disc photographs of glaucoma patients spanning several years. These computer-generated measures of cup-to-disc volume ratios correlate well with the traditional stereo cup-to-disc ratios manually computed from clinical interpretations. Such algorithmic approach to semi-automated computation of cup-to-disc volume ratios may potentially provide a more precise and repeatable measure of progression of glaucoma than the existing clinical measures. Moreover, the 3-D surface recovery technique developed in this thesis may provide a general technique for visualizing 3-D objects in a natural scene.Item Unsupervised learning methods: An efficient clustering framework with integrated model selection(2012-08) Corona, Enrique; Nutter, Brian; Mitra, Sunanda; Pal, Ranadip; López-Benitez, NoéClassification is one of the most important practices in data analysis. In the context of machine learning, this practice can be viewed as the problem of identifying representative data patterns in such a manner that coherent groups are formed. If the data structure is readily available (e.g. supervised learning), it is usually used to establish classification rules for discrimination. However, when the data is unlabeled, its underlying structure must be unveiled first. Consequently, unsupervised classification poses more challenges. Among them, the fundamental question of an appropriate number of groups or clusters in the data must be addressed. In this context, the "jump" method, an efficient but limited linear approach that finds plausible answers to the number of clusters in a dataset, is improved via the optimization of an appropriate objective function that quantifies the quality of particular cluster configurations. Recent developments showing interesting associations between spectral clustering (SC) and kernel principal component analysis (KPCA) are used to extend the improved method to the non-linear domain. This is achieved by mapping the input data to a new space where the original clusters appear as linear structures. The characteristics of this mapping depend to a large extent on the parameters of the kernel function selected. By projecting these linear structures to the unit sphere, the proposed method is able to measure the quality of the resulting cluster configurations. These quality scores aid in the simultaneous decision of the kernel parameters (i.e. model selection) and the number of clusters present in the dataset. Results of the enhanced jump method are compared to other relative validation criteria such as minimum description length (MDL), Akaike's information criterion (AIC) and consistent Akaike's information criterion (CAIC). The extension of the method is tested with other cluster validity indices, in similar settings, such as the adjusted Rand index (ARI) and the balanced line fit (BLF). Finally, image segmentation examples are shown as a real world application of the technique.