Is this project an undergraduate, graduate, or faculty project?
Undergraduate
Project Type
group
Campus
Daytona Beach
Authors' Class Standing
Thomas Pasfield, Junior, Ioannis Paraschos, Senior, Kian Greene, Senior, Ryan Reynolds, Senior, Dylan Pereira, Senior, Jacob Koscinski, Senior
Lead Presenter's Name
Thomas Pasfield
Lead Presenter's College
DB College of Engineering
Faculty Mentor Name
Dr. Mihhail Berezovski
Abstract
Additive manufacturing (3D printing) is increasingly utilized within modern industry. 3D-printed components introduce new structural behavior and problems to account for. Computed Tomography (CT) provides a method to analyze this structure without compromising the material, making it a valuable tool moving forward. This work aims to consolidate analysis methods to improve their usefulness within the additive manufacturing industrial process. Existing methods and modified approaches are tested on a 1789-image dataset provided by the Pacific Northwest National Laboratory. Our analysis focuses on segmentation, dimensioning, and 3D-printing-specific needs such as layer separation, in-fill analysis, and path accuracy. We approach segmentation via region-growing methods, where the user provides seed values. Our various dimensioning implementations utilize optimization and voting methods to fit prior knowledge about shape to the observed article. 3D-printing-specific features build from our other methods to observe in-fill density and layer separation. Evaluation methods are employed to provide quantitative analysis of scanned samples. This work hopes to aid in standardizing and certifying CT analysis within the additive manufacturing industrial field.
Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?
No
Quantitative Analysis for Industrial Computed Tomography
Additive manufacturing (3D printing) is increasingly utilized within modern industry. 3D-printed components introduce new structural behavior and problems to account for. Computed Tomography (CT) provides a method to analyze this structure without compromising the material, making it a valuable tool moving forward. This work aims to consolidate analysis methods to improve their usefulness within the additive manufacturing industrial process. Existing methods and modified approaches are tested on a 1789-image dataset provided by the Pacific Northwest National Laboratory. Our analysis focuses on segmentation, dimensioning, and 3D-printing-specific needs such as layer separation, in-fill analysis, and path accuracy. We approach segmentation via region-growing methods, where the user provides seed values. Our various dimensioning implementations utilize optimization and voting methods to fit prior knowledge about shape to the observed article. 3D-printing-specific features build from our other methods to observe in-fill density and layer separation. Evaluation methods are employed to provide quantitative analysis of scanned samples. This work hopes to aid in standardizing and certifying CT analysis within the additive manufacturing industrial field.