About this Abstract |
Meeting |
2025 TMS Annual Meeting & Exhibition
|
Symposium
|
Novel Strategies for Rapid Acquisition and Processing of Large Datasets from Advanced Characterization Techniques
|
Presentation Title |
H-34: Classification of Defects in Tomographic Reconstructions of Hyperscale Advanced Packages |
Author(s) |
Orion L. Kafka, Jason Killgore, Newell Moser, Zachary Grey, Jake Rezac |
On-Site Speaker (Planned) |
Orion L. Kafka |
Abstract Scope |
The purpose of this project is to improve the accuracy, speed, and ease of use of x-ray computed tomography (xCT) as it pertains to the identification of critical defects in chiplets and advanced packages. Specifically, the proposal combines novel uses of advanced xCT instrumentation with sophisticated machine learning and shape classification to locate and classify submicron-scale defects when capturing millimeter scale (i.e. chiplet or package scale) fields of view. We will establish a workflow combining data acquisition, data processing, defect classification, and automation. This workflow will then be streamlined such that it can be readily transferred to the semiconductor industry. |
Proceedings Inclusion? |
Planned: |