Doctor's Theses (authored and supervised):
"Visual Analysis of Methods for Processing 3D X-ray Computed Tomography Data of Advanced Composites";
Supervisor, Reviewer: E. Gröller, Ch. Heinzl;
Visual Computing and Human-Centered Technology,
oral examination: 2019-12-05.
Advanced composites have excellent mechanical properties at low weight and can be realized as complex components that can be manufactured quickly and cost-effectively. Due to these outstanding characteristics, these materials are used in many di˙erent areas of industry, such as aviation and automotive. Industrial 3D X-ray computed tomography (XCT) is used as a non-destructive testing (NDT) method to inspect the quality of components and to develop new advanced composite materials. XCT has the ability to determine the inner and outer geometries of a specimen non-destructively. For example, interesting features in fiber-reinforced polymers (FRPs) such as fibers, pores, and higher-density inclusions can be detected. The high resolutions of modern XCT devices generate large volume datasets, which reveal very fine structures. However, this high information content makes the exploration and analysis of the datasets with conventional methods very diÿcult and time-consuming. In this doctoral thesis, typical NDT application scenarios of advanced composites using XCT are addressed and visual analysis methods and visualization techniques are designed to provide material experts with tools to improve their workflow and to eÿciently analyze the XCT data, so that domain-specific questions can be answered easily and quickly. This work describes a novel visualization system for the interactive exploration and detailed analysis of FRPs, a tool for the visual analysis and evaluation of segmentation filters to accurately determine porosity in FRPs, and a more general system for the visual comparison of interesting features in an ensemble of XCT datasets are presented. The results of the individual visualization systems are presented using real-world and simulated XCT data. The proposed visual analysis methods support the experts in their workflows by enabling improved data analysis processes that are simple, fast, and well-founded, and provide new insights into material characterization with XCT.
Electronic version of the publication:
Created from the Publication Database of the Vienna University of Technology.