Aung, A. P. P., Lin, Z., Pahwa, R. S., Made, R. I., & Jayavelu, S. (2025). Defect detection and localization using 2D slicing method on 3D X-ray microscopy. In AI4X Conference.
Abstract:
Defects in semiconductor devices remain a persistent challenge. Certain defects can significantly impair device performance, ultimately affecting functionality. Their rarity often complicates detection, making non-destructive methods essential. These techniques preserve the defect’s context, pro-viding valuable insights into its root cause and guiding appropriate corrective actions. However, analyzing data from non-destructive methods, such as3D X-ray microscopy (XRM), poses difficulties due to large data sizes and limited availability. To overcome these challenges, we propose a two-step machine learning model that performs both defect detection of the sample as either defective or non-defective, and defect localization, pinpointing the failure region within the sample.
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the A*STAR - Industry Alignment Fund - Pre-Positioning: Machine Learning Guided Failure Analysis & Diagnostic Capability Development for Next Generation 3D-IC Packaging
Grant Reference no. : M23K8a0050