More Accurate And Detailed Analysis of Semiconductor Defects In SEM Images Using SEMI-PointRend


A technical paper titled “SEMI-PointRend: Improved Semiconductor Wafer Defect Classification and Segmentation as Rendering” was published (preprint) by researchers at imec, University of Ulsan, and KU Leuven.

“In this study, we applied the PointRend (Point-based Rendering) method to semiconductor defect segmentation. PointRend is an iterative segmentation algorithm inspired by image rendering in computer graphics, a new image segmentation method that can generate high-resolution segmentation masks. It can also be flexibly integrated into common instance segmentation meta-architecture such as Mask-RCNN and semantic meta-architecture such as FCN. We implemented a model, termed as SEMI-PointRend, to generate precise segmentation masks by applying the PointRend neural network module. In this paper, we focus on comparing the defect segmentation predictions of SEMI-PointRend and Mask-RCNN for various defect types (line-collapse, single bridge, thin bridge, multi bridge non-horizontal). We show that SEMI-PointRend can outperforms Mask R-CNN by up to 18.8% in terms of segmentation mean average precision.”

Find the technical paper here. Published Feb. 2023.

Hwang, MinJin, Bappaditya Dey, Enrique Dehaerne, Sandip Halder, and Young-han Shin. “SEMI-PointRend: Improved Semiconductor Wafer Defect Classification and Segmentation as Rendering.” arXiv preprint arXiv:2302.09569 (2023). To be published by SPIE in the proceedings of Metrology, Inspection, and Process Control XXXVII. https://doi.org/10.48550/arXiv.2302.09569.

Leave a Reply

(Note: This name will be displayed publicly)