Método de visualização e extração de atributos de equimoses multifacetadas nos membros inferiores

• Ano de defesa: 2014
• Autor(a) principal: Thomaz, Ricardo de Lima
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; BR; Programa de Pós-graduação em Engenharia Elétrica; Engenharias; UFU
• Palavras-chave em Português: Equimoses multifacetadas; Reconstrução tridimensional; Extração de superfície; Método de visualização; Extração de atributos; Quantificação e caracterização; Multifaceted ecchymosis; Three-dimensional reconstruction; Surface extraction; Visualization method; Feature extraction; Quantification and characterization; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/14559; https://doi.org/10.14393/ufu.di.2014.26

This dissertation presents a computational system for three-dimensional reconstruction and surface extraction of the human lower limb as a new methodology of visualizing images of multifaceted ecchymosis on the lower limbs. Also, it presents algorithms for quantitative feature extraction of images to aid on the objective characterization of wounds. Through standardization of image acquisition by a mechanical system, an algorithm was developed for three-dimensional reconstruction and surface extraction based on the extraction of depth from silhouettes. To extract quantitative data from images of ecchymosis, algorithms were implemented to extract texture features based on the Discrete Cosine Transform, Two-dimensional Haar Wavelet and Haralick s descriptors, and for extracting color features based on the HSV (hue, saturation and value) color space. In order to validate this work, it was designed a model simulating a lower limb with multifaceted ecchymosis, which was later submitted to all algorithms developed. It was observed that the systems for three-dimensional reconstruction and surface extraction of the object were able to generate a new visualization method of the lesion. However, due to procedural flaws during image acquisition, there was a displacement error of 11.84%. The algorithms for quantitative feature extraction succeeded in separating intentionally modified images of the lesion by calculating the Hausdorff distance between their features. The results allow us to conclude that the developed systems provided adequate three-dimensional and two-dimensional visualization of the surface of the simulated model, as well as are capable of defining the features to differentiate the modifications applied onto the simulated images. Despite the lack of experiments with real ecchymoses, the systems developed in this work show great potential to be included in the standard methods for the visualization of ecchymoses and the definition of attributes capable of differentiating the different stages of absorption.