Hybrid K -means clustering and support vector machine method for via and metal line detections in delayered IC images

This brief proposes a hybrid K-means clustering and support vector machine (HKCSVM) method to detect the positions of vias and metal lines from delayered IC images for subsequent netlist extraction. The main contributions of the proposed method include: 1) fully automated detection of via and metal...

Full description

Saved in:
Bibliographic Details
Main Authors: Cheng, Deruo, Shi, Yiqiong, Lin, Tong, Gwee, Bah-Hwee, Toh, Kar-Ann
Other Authors: School of Electrical and Electronic Engineering
Format: Article
Language:English
Published: 2019
Subjects:
Online Access:https://hdl.handle.net/10356/81439
http://hdl.handle.net/10220/50385
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-81439
record_format dspace
spelling sg-ntu-dr.10356-814392020-09-26T22:19:53Z Hybrid K -means clustering and support vector machine method for via and metal line detections in delayered IC images Cheng, Deruo Shi, Yiqiong Lin, Tong Gwee, Bah-Hwee Toh, Kar-Ann School of Electrical and Electronic Engineering Temasek Laboratories IC Image Analysis Support Vector Machine Engineering::Electrical and electronic engineering This brief proposes a hybrid K-means clustering and support vector machine (HKCSVM) method to detect the positions of vias and metal lines from delayered IC images for subsequent netlist extraction. The main contributions of the proposed method include: 1) fully automated detection of via and metal line positions without any need of human interventions and 2) novel hybrid methodology to embody K-means clustering and support vector machine for retrieving precise positions of vias and metal lines in contrast to the individual techniques, which can only provide a region for the detected elements. From experiments on 50 SEM images of 1536 × 2048 pixels taken from fabricated IC chips (@TSMC 130-nm CMOS process), our proposed HKCSVM method achieves an F-score of 99.82% for via detection, and mean intersection over union of 94.44% and mean pixel accuracy of 95.83% for metal line detection, which are both superior to the reported approaches. Accepted version 2019-11-11T06:29:30Z 2019-12-06T14:31:00Z 2019-11-11T06:29:30Z 2019-12-06T14:31:00Z 2018 Journal Article Cheng, D., Shi, Y., Lin, T., Gwee, B.-H., & Toh, K.-A. (2018). Hybrid K -means clustering and support vector machine method for via and metal line detections in delayered IC images. IEEE Transactions on Circuits and Systems II: Express Briefs, 65(12), 1849-1853. doi:10.1109/TCSII.2018.2827044 1549-7747 https://hdl.handle.net/10356/81439 http://hdl.handle.net/10220/50385 10.1109/TCSII.2018.2827044 en IEEE Transactions on Circuits and Systems II: Express Briefs © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: https://doi.org/10.1109/TCSII.2018.2827044 5 p. application/pdf
institution Nanyang Technological University
building NTU Library
country Singapore
collection DR-NTU
language English
topic IC Image Analysis
Support Vector Machine
Engineering::Electrical and electronic engineering
spellingShingle IC Image Analysis
Support Vector Machine
Engineering::Electrical and electronic engineering
Cheng, Deruo
Shi, Yiqiong
Lin, Tong
Gwee, Bah-Hwee
Toh, Kar-Ann
Hybrid K -means clustering and support vector machine method for via and metal line detections in delayered IC images
description This brief proposes a hybrid K-means clustering and support vector machine (HKCSVM) method to detect the positions of vias and metal lines from delayered IC images for subsequent netlist extraction. The main contributions of the proposed method include: 1) fully automated detection of via and metal line positions without any need of human interventions and 2) novel hybrid methodology to embody K-means clustering and support vector machine for retrieving precise positions of vias and metal lines in contrast to the individual techniques, which can only provide a region for the detected elements. From experiments on 50 SEM images of 1536 × 2048 pixels taken from fabricated IC chips (@TSMC 130-nm CMOS process), our proposed HKCSVM method achieves an F-score of 99.82% for via detection, and mean intersection over union of 94.44% and mean pixel accuracy of 95.83% for metal line detection, which are both superior to the reported approaches.
author2 School of Electrical and Electronic Engineering
author_facet School of Electrical and Electronic Engineering
Cheng, Deruo
Shi, Yiqiong
Lin, Tong
Gwee, Bah-Hwee
Toh, Kar-Ann
format Article
author Cheng, Deruo
Shi, Yiqiong
Lin, Tong
Gwee, Bah-Hwee
Toh, Kar-Ann
author_sort Cheng, Deruo
title Hybrid K -means clustering and support vector machine method for via and metal line detections in delayered IC images
title_short Hybrid K -means clustering and support vector machine method for via and metal line detections in delayered IC images
title_full Hybrid K -means clustering and support vector machine method for via and metal line detections in delayered IC images
title_fullStr Hybrid K -means clustering and support vector machine method for via and metal line detections in delayered IC images
title_full_unstemmed Hybrid K -means clustering and support vector machine method for via and metal line detections in delayered IC images
title_sort hybrid k -means clustering and support vector machine method for via and metal line detections in delayered ic images
publishDate 2019
url https://hdl.handle.net/10356/81439
http://hdl.handle.net/10220/50385
_version_ 1681059696766615552