Robust Detection, Segmentation, and Metrology of High Bandwidth Memory 3D Scans Using an Improved Semi-Supervised Deep Learning Approach
Recent advancements in 3D deep learning have led to significant progress in improving accuracy and reducing processing time, with applications spanning various domains such as medical imaging, robotics, and autonomous vehicle navigation for identifying and segmenting different structures. In this st...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/12/5470 |
| _version_ | 1797592741180342272 |
|---|---|
| author | Jie Wang Richard Chang Ziyuan Zhao Ramanpreet Singh Pahwa |
| author_facet | Jie Wang Richard Chang Ziyuan Zhao Ramanpreet Singh Pahwa |
| author_sort | Jie Wang |
| collection | DOAJ |
| description | Recent advancements in 3D deep learning have led to significant progress in improving accuracy and reducing processing time, with applications spanning various domains such as medical imaging, robotics, and autonomous vehicle navigation for identifying and segmenting different structures. In this study, we employ the latest developments in 3D semi-supervised learning to create cutting-edge models for the 3D object detection and segmentation of buried structures in high-resolution X-ray semiconductors scans. We illustrate our approach to locating the region of interest of the structures, their individual components, and their void defects. We showcase how semi-supervised learning is utilized to capitalize on the vast amounts of available unlabeled data to enhance both detection and segmentation performance. Additionally, we explore the benefit of contrastive learning in the data pre-selection step for our detection model and multi-scale Mean Teacher training paradigm in 3D semantic segmentation to achieve better performance compared with the state of the art. Our extensive experiments have shown that our method achieves competitive performance and is able to outperform by up to 16% on object detection and 7.8% on semantic segmentation. Additionally, our automated metrology package shows a mean error of less than 2 <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi mathvariant="sans-serif">μ</mi></semantics></math></inline-formula>m for key features such as Bond Line Thickness and pad misalignment. |
| first_indexed | 2024-03-11T01:57:26Z |
| format | Article |
| id | doaj.art-da8383402e794d15b687107ace38dd7e |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T01:57:26Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-da8383402e794d15b687107ace38dd7e2023-11-18T12:31:30ZengMDPI AGSensors1424-82202023-06-012312547010.3390/s23125470Robust Detection, Segmentation, and Metrology of High Bandwidth Memory 3D Scans Using an Improved Semi-Supervised Deep Learning ApproachJie Wang0Richard Chang1Ziyuan Zhao2Ramanpreet Singh Pahwa3Institute for Infocomm Research (I2R), Agency for Science, Technology and Research (A*STAR), 1 Fusionopolis Way, #21-01, Connexis South Tower, Singapore 138632, SingaporeInstitute for Infocomm Research (I2R), Agency for Science, Technology and Research (A*STAR), 1 Fusionopolis Way, #21-01, Connexis South Tower, Singapore 138632, SingaporeInstitute for Infocomm Research (I2R), Agency for Science, Technology and Research (A*STAR), 1 Fusionopolis Way, #21-01, Connexis South Tower, Singapore 138632, SingaporeInstitute for Infocomm Research (I2R), Agency for Science, Technology and Research (A*STAR), 1 Fusionopolis Way, #21-01, Connexis South Tower, Singapore 138632, SingaporeRecent advancements in 3D deep learning have led to significant progress in improving accuracy and reducing processing time, with applications spanning various domains such as medical imaging, robotics, and autonomous vehicle navigation for identifying and segmenting different structures. In this study, we employ the latest developments in 3D semi-supervised learning to create cutting-edge models for the 3D object detection and segmentation of buried structures in high-resolution X-ray semiconductors scans. We illustrate our approach to locating the region of interest of the structures, their individual components, and their void defects. We showcase how semi-supervised learning is utilized to capitalize on the vast amounts of available unlabeled data to enhance both detection and segmentation performance. Additionally, we explore the benefit of contrastive learning in the data pre-selection step for our detection model and multi-scale Mean Teacher training paradigm in 3D semantic segmentation to achieve better performance compared with the state of the art. Our extensive experiments have shown that our method achieves competitive performance and is able to outperform by up to 16% on object detection and 7.8% on semantic segmentation. Additionally, our automated metrology package shows a mean error of less than 2 <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi mathvariant="sans-serif">μ</mi></semantics></math></inline-formula>m for key features such as Bond Line Thickness and pad misalignment.https://www.mdpi.com/1424-8220/23/12/54703D semi-supervised Learning3D object detection3D semantic segmentationcontrastive learning3D metrology |
| spellingShingle | Jie Wang Richard Chang Ziyuan Zhao Ramanpreet Singh Pahwa Robust Detection, Segmentation, and Metrology of High Bandwidth Memory 3D Scans Using an Improved Semi-Supervised Deep Learning Approach Sensors 3D semi-supervised Learning 3D object detection 3D semantic segmentation contrastive learning 3D metrology |
| title | Robust Detection, Segmentation, and Metrology of High Bandwidth Memory 3D Scans Using an Improved Semi-Supervised Deep Learning Approach |
| title_full | Robust Detection, Segmentation, and Metrology of High Bandwidth Memory 3D Scans Using an Improved Semi-Supervised Deep Learning Approach |
| title_fullStr | Robust Detection, Segmentation, and Metrology of High Bandwidth Memory 3D Scans Using an Improved Semi-Supervised Deep Learning Approach |
| title_full_unstemmed | Robust Detection, Segmentation, and Metrology of High Bandwidth Memory 3D Scans Using an Improved Semi-Supervised Deep Learning Approach |
| title_short | Robust Detection, Segmentation, and Metrology of High Bandwidth Memory 3D Scans Using an Improved Semi-Supervised Deep Learning Approach |
| title_sort | robust detection segmentation and metrology of high bandwidth memory 3d scans using an improved semi supervised deep learning approach |
| topic | 3D semi-supervised Learning 3D object detection 3D semantic segmentation contrastive learning 3D metrology |
| url | https://www.mdpi.com/1424-8220/23/12/5470 |
| work_keys_str_mv | AT jiewang robustdetectionsegmentationandmetrologyofhighbandwidthmemory3dscansusinganimprovedsemisuperviseddeeplearningapproach AT richardchang robustdetectionsegmentationandmetrologyofhighbandwidthmemory3dscansusinganimprovedsemisuperviseddeeplearningapproach AT ziyuanzhao robustdetectionsegmentationandmetrologyofhighbandwidthmemory3dscansusinganimprovedsemisuperviseddeeplearningapproach AT ramanpreetsinghpahwa robustdetectionsegmentationandmetrologyofhighbandwidthmemory3dscansusinganimprovedsemisuperviseddeeplearningapproach |