VisGFX - A Visual Guidance Framework for XCT Simulation

X-ray computed tomography requires balancing image quality with constraints such as acquisition time and cost. Although X-ray simulation can provide valuable insights into different configuration options, manually analyzing large sets of simulation results to determine the optimal parameter combination for a given use case is cognitively demanding and time-consuming. We introduce a visual guidance system addressing this challenge by the combination of a visualization framework with multiobjective optimization. The system supports users in finding optimal parameter combinations through three key components: an overview visualization showing the hierarchical structure of the dataset and subsequent operations acting on it, detailed visualization of high-dimensional simulation data for exploration and selection, and a visual guidance view that assists with navigation and interpretation. Together, these components help users to efficiently filter options and identify the most suitable parameter settings for their goals. We demonstrate our visualization framework through a case study that focuses on determining XCT scan parameters to optimize reconstruction image quality for multi-material industrial samples. Specifically, we determine the optimal trade-off between scan parameters for achieving high quality of the final reconstruction at minimum acquisition requirements.