Presentation
13 June 2022 Flexible rapid x-ray simulation for large-scale synthetic data generation across a variety of x-ray modalities
Author Affiliations +
Abstract
Computer simulations have been successfully employed in the development, assessment, and refinement of X-ray systems. However, many of these simulation tools have either been too slow, too complicated, or not capable of addressing the problem at hand. To address these issues, we introduce a fast flexible simulation tool, QSimRT that inherently interfaces with existing three-dimensional (3D) model creation tools (such as Blender) and utilizes state-of-the-art raytracing hardware and algorithms. The implementation of realistic X-ray simulation is multifold: our solution faithfully models the 3D, volumetric shapes, X-ray material properties, and packing structures of virtual items, accurately and efficiently simulates the X-ray physics, and can generalize to a breadth of X-ray measurement architectures, including AT, CT, air cargo, and cargo portal. In this talk, we demonstrate how our method yields realistic simulations, comparable to real-world data.
Conference Presentation
© (2022) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
David Coccarelli, Michael E. Gehm, and Joel A. Greenberg "Flexible rapid x-ray simulation for large-scale synthetic data generation across a variety of x-ray modalities", Proc. SPIE PC12104, Anomaly Detection and Imaging with X-Rays (ADIX) VII, PC1210403 (13 June 2022); https://doi.org/10.1117/12.2618945
Advertisement
Advertisement
KEYWORDS
X-rays

3D modeling

Computer simulations

Computer architecture

Computing systems

Interfaces

Particle physics

Back to Top