Skip navigation


  • DSpace logo
  • Home
  • Collections
  • Researcher Profile
  • Explore by
    • Researcher Profile
  • VGU library
  • Help
  • User Guide
  • Sign on to:
    • My DSpace
    • Receive email
      updates
    • Edit Account details

VGU RESEARCH REPOSITORY


Please use this identifier to cite or link to this item: https://epub.vgu.edu.vn/handle/dlibvgu/1878
DC FieldValueLanguage
dc.contributor.authorNguyen Ngoc Vinhen_US
dc.date.accessioned2024-11-15T07:42:54Z-
dc.date.accessioned2024-11-15T07:42:57Z-
dc.date.available2024-11-15T07:42:54Z-
dc.date.available2024-11-15T07:42:57Z-
dc.date.issued2024-
dc.identifier.urihttps://epub.vgu.edu.vn/handle/dlibvgu/1878-
dc.description.abstractThe integrated circuit (IC) is a compact electronic device comprising interconnected components such as transistors, resistors, and capacitors. These elements are etched onto a small piece of semiconductor material, typically silicon, through photolithography. ICs are essential to a vast array of electronic devices, including computers, smartphones, and televisions, where they perform critical functions like data processing and storage. However, ICs are vulnerable to external factors such as scratches, dust particles, and electrostatic discharges, which can degrade their performance or lead to complete failure with potentially severe consequences. To address this issue, I propose a machine learning anomaly detection method using the Reconstruction-Based Neural Network approach, which utilizes multiple Convolutional Neural Network Autoencoders. This approach identifies anomalies by analyzing the pattern features found in the microscopic images of the integrated circuits. It utilizes numerous CNN-Autoencoders trained to recognize and differentiate between typical and atypical patterns in the circuitry. The method involves a detailed image mapping process, where masks are used to locate specific anomaly spots on the images. By mapping these masks onto the input images, the system can determine the presence and exact location of an anomaly. Experimental evaluations using real microscopic images showed that the method achieved a high precision rate, successfully identifying and locating anomalies on the ICs with a 91% detection rateen_US
dc.language.isoenen_US
dc.rightsAttribution-NonCommercial 4.0 International*
dc.rights.urihttps://creativecommons.org/licenses/by-nc/4.0/*
dc.subjectIntegrated circuitsen_US
dc.subjectAnomaly detectionen_US
dc.subjectReconstruction-based neural networksen_US
dc.subjectMachine learningen_US
dc.subjectDeep learningen_US
dc.titleReconstruction-based neural networks for anomaly detection in microscopic integrated circuit imagesen_US
dc.typeThesisen_US
item.fulltextWith Fulltext-
item.languageiso639-1other-
item.grantfulltextrestricted-
Appears in Collections:Computer Science (CS)
Computer Science (CS)
Files in This Item:
File Description SizeFormat Existing users please Login
Reconstruction-based neural networks for anomaly detection in microscopic integrated circuit images.pdf18.67 MBAdobe PDF
Show simple item record

Page view(s)

91
checked on Aug 24, 2025

Download(s)

37
checked on Aug 24, 2025

Google ScholarTM

Check


This item is licensed under a Creative Commons License Creative Commons

© Copyright 2020 by Vietnamese - German University Library.
Add: Ring road 4, Quarter 4, Thoi Hoa Ward, Ben Cat City, Binh Duong Province
Tel.:(0274) 222 0990. Ext.: 70206