High-Precision and Rapid Detection of Complex Defects in Transferred 2D Materials Enabled by Machine Learning Algorithms

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：19

、訪客IP：3.133.141.1

姓名

盧美珍(Than Su Su Htay) 查詢紙本館藏

畢業系所

能源工程研究所

論文名稱

(High-Precision and Rapid Detection of Complex Defects in Transferred 2D Materials Enabled by Machine Learning Algorithms)

相關論文

★ 捲對捲乾轉印方法於製作高效能石墨烯透明導電膜之研究	★ 利用氟素高分子摻雜於提升石墨烯導電膜的效能與穩定性之研究
★ 以石墨烯混成陶瓷粉末於製作高導熱及高電阻之聚亞醯胺薄膜的研究	★ 以奈米銅催化輔助控制多孔石墨烯之孔隙結構及其於超級電容之研究
★ 研究超潔淨石墨烯之場效電晶體於提升基因感測器之效能	★ 利用氟化自組裝膜輔助轉印石墨烯薄膜及其於場效電晶體特性之研究
★ 多孔石墨烯邊界態之氮改質於超級電容的效能研究	★ 石墨烯場效應電晶體應用於DNA生醫感測晶片之元件整合和效能評估的研究
★ 添加氟化石墨烯於奈米高分子複合材料以增強防腐性能	★ 石墨烯功能性改質於鋰離子電池負極材料之研究
★ 紫外光輻照於輔助轉印高品質石墨烯之研究	★ 氟化石墨烯複合結構於鋰離子電池的人工固態電解質界面膜之研究
★ 超高附著力之氟化石墨烯薄膜於固態磨潤之研究	★ 真空壓印於二維材料轉印製程之研究
★ 氟化石墨烯複合結構在鋰金屬電池中的雙功能陽極之機制探討	★ 氟化石墨烯複合材料塗層於多功能披覆之研究

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

至系統瀏覽論文 (2030-11-1以後開放)

摘要(中)

二維材料因其在光學和下一代半導體裝置中的廣泛應用而受到廣泛關注。然而，在整合到設備之前，典型的 2D 材料必須從生長基板轉移到目標基板，這個過程通常會引入複雜的缺陷，例如皺紋、破裂和殘留物。確保高品質的石墨烯生產對於這些應用至關重要。儘管成像技術取得了進步，但高效、準確地檢測轉移的二維材料中的缺陷仍然是一個挑戰。這主要是由於無法精確、快速地辨識這些材料的複雜形態，限制了它們的實際應用。本研究重點關注透過化學氣相沉積 (CVD) 合成的石墨烯，以及隨後使用卷對卷 (RTR)、濕法和乾法轉移技術等方法將其轉移到不同的基材上。引入了創新的自動分割工具，以 JSON（JavaScript 物件表示法）格式產生缺陷註釋，以提高缺陷偵測的一致性和效率。光學顯微鏡 (OM) 影像與 YOLOv7 深度學習模型一起用於準確識別和量化缺陷，特別是那些形狀不規則的缺陷。透過實施精細的資料集分割策略並優化模型的損失函數，偵測精度提高了 10%。此外，自動分割將手動註釋時間減少了 75%，提高了資料一致性，並將預測率提高了 18%。這些發現表明，自動化系統顯著改善了石墨烯的缺陷檢測，簡化了品質控制流程，並提高了生產效率。這些進步預計將支持高性能石墨烯技術的開發，為傳統檢測方法的局限性提供可靠的解決方案。

摘要(英)

2D materials have gained significant attention due to their wide applications in optical and next-generation semiconductor devices. However, before being integrated into devices, typical 2D materials must be transferred from the growth substrate to a target substrate, a process that often introduces complex defects such as wrinkles, ruptures, and residues. Ensuring high-quality graphene production is critical for these applications. Despite advancements in imaging technologies, achieving high-efficiency and accurate detection of defects in transferred 2D materials remains a challenge. This is primarily due to the inability to precisely and rapidly recognize the intricate morphology of these materials, limiting their practical application. In this study focus on graphene synthesized via Chemical Vapor Deposition (CVD) and its subsequent transfer to different substrates using methods such as Roll-to-Roll (RTR), wet, and dry transfer techniques. An innovative automated segmentation tool has been introduced, generating defect annotations in JSON (JavaScript Object Notation) format to enhance consistency and efficiency in defect detection. Optical microscopy (OM) images, along with the YOLOv7 deep learning model, were used to accurately identify and quantify defects, particularly those with irregular shapes. By implementing a refined dataset-splitting strategy and optimizing the model′s loss function, a 10% increase in detection accuracy was achieved. Additionally, the automated segmentation reduced manual annotation time by 75%, improving data consistency and increasing the prediction rate by 18%. These findings demonstrate that the automated system significantly improves defect detection in graphene, streamlines quality control processes, and boosts production efficiency. The advancements are expected to support the development of high-performance graphene-based technologies, providing a reliable solution to the limitations of traditional inspection methods.

關鍵字(中)

★ One keyword per line

關鍵字(英)

★ One keyword per line

論文目次

CONTENT
Library Authorization for Thesis/Dissertation i
Advisor’s Recommendation ii
Verification Letter from the Oral Examination Committee iii
摘要 iv
ABSTRACT v
Acknowledgement vi
CONTENT vii
List of Figures ix
List of Tables xi
1. INTRODUCTION 1
2. METHODOLOGY 5
2.1 Experimental Flowchart 6
2.2 Data Collection 7
2.3 Defect Type Classification 7
2.4 Automated Image Segmentation and JSON Conversion 9
2.5 JSON to YOLO Conversion 11
2.6 Dataset Splitting 11
2.7 Hyperparameter Tuning 13
2.7.1 Tuning Loss Function Components 13
2.7.2 Key Adjustments and Impact 16
2.7.3 HSV Parameter Tuning 17
2.8 Configuration of Training Model and Implementation 19
2.9 Local Environment Setup and Execution 20
2.10 Output Metrics and Automated Reporting 21
2.11 Accuracy, Recall and Precision 21
2.12 Workflow for Detecting Defect Analysis 22
3. RESULTS AND DISCUSSION 25
3.1 Automated Segmentation and JSON Format for Defect Detection in Graphene 25
3.2 Comparison of Defect Detection Methods 27
3.3 Comparison of Manual and Automated Labeling Methods for Defect Detection 29
3.4 Hyperparameter Tuning 31
3.4.1 Tuning Loss Function to Enhance Defect Detection at 100μm Scale (X1000) 31
3.4.2 Tuning Loss Function to Enhance Defect Detection at 20μm Scale (X200) 35
3.4.3 Impact of HSV Parameter Tuning on Optimized Model Performance 39
3.5 Graphene Defect Detection: From High Defect to Defect-Free 44
3.6 Shape and Density Variations in Complex Nucleation Domains 49
3.7 Defects Analysis in CVD Graphene Transfer Methods 51
4. CONCLUSION 54
REFERENCES xii
Supplementary Section: xiv
SS-01: A Python Code for Automated Image Segmentation and JSON Conversion xiv
SS-02: Dataset Splitting xvii
SS-03: Output Metric and Automated Reporting xx
SS-04: Analysis of Defect Nucleation in Graphene Growth xxii
SS-05: Analysis of Defects in CVD Transfer Methods xxiv

參考文獻

1. Qing, F., et al., Towards large-scale graphene transfer. Nanoscale, 2020. 12(20): p. 10890-10911.
2. Watson, A.J., et al., Transfer of large-scale two-dimensional semiconductors: challenges and developments. 2D Materials, 2021. 8(3): p. 032001.
3. Bhowmik, S. and A.G. Rajan, Chemical vapor deposition of 2D materials: A review of modeling, simulation, and machine learning studies. Iscience, 2022. 25(3).
4. He, S.-M., et al., Toward large-scale CVD graphene growth by enhancing reaction kinetics via an efficient interdiffusion mediator and mechanism study utilizing CFD simulations. Journal of the Taiwan Institute of Chemical Engineers, 2021. 128: p. 400-408.
5. He, S.M., et al., Plasma?Driven Selenization for Electrical Property Enhancement in Janus 2D Materials. Small Methods, 2024: p. 2400150.
6. Juang, Z.-Y., et al., Graphene synthesis by chemical vapor deposition and transfer by a roll-to-roll process. carbon, 2010. 48(11): p. 3169-3174.
7. Bhatt, M.D., H. Kim, and G. Kim, Various defects in graphene: a review. RSC advances, 2022. 12(33): p. 21520-21547.
8. Ullah, S., et al., Graphene transfer methods: A review. Nano Research, 2021: p. 1-17.
9. He, S.-M., et al., Wrinkle-free graphene films on fluorinated self-assembled monolayer-modified substrates for enhancing the electrical performance of transistors. ACS Applied Nano Materials, 2022. 5(4): p. 5793-5802.
10. He, S.-M., et al., Spectroscopic and electrical characterizations of low-damage phosphorous-doped graphene via ion implantation. ACS applied materials & interfaces, 2019. 11(50): p. 47289-47298.
11. Wu, J.-B., et al., Raman spectroscopy of graphene-based materials and its applications in related devices. Chemical Society Reviews, 2018. 47(5): p. 1822-1873.
12. De Silva, K.K.H., et al., Nanoscale electrical characterization of graphene-based materials by atomic force microscopy. Journal of Materials Research, 2022. 37(20): p. 3319-3339.
13. Hung, Y.-H., et al., Ultraclean and facile patterning of CVD graphene by a UV-light-assisted dry transfer method. ACS Applied Materials & Interfaces, 2023. 15(3): p. 4826-4834.
14. Dai, G.-P., et al., Square-shaped, single-crystal, monolayer graphene domains by low-pressure chemical vapor deposition. Materials Research Letters, 2013. 1(2): p. 67-76.
15. Wang, C.-Y., A. Bochkovskiy, and H.-Y.M. Liao. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2023.
16. Dehaerne, E., et al. Optimizing YOLOv7 for semiconductor defect detection. in Metrology, Inspection, and Process Control XXXVII. 2023. SPIE.
17. Ramezani, F., et al., Automatic detection of multilayer hexagonal boron nitride in optical images using deep learning-based computer vision. Scientific Reports, 2023. 13(1): p. 1595.
18. Yang, E., et al., Machine Learning-Assisted Identification of Single-Layer Graphene via Color Variation Analysis. Nanomaterials, 2024. 14(2): p. 183.
19. Yu, Q., et al., Control and characterization of individual grains and grain boundaries in graphene grown by chemical vapour deposition. Nature materials, 2011. 10(6): p. 443-449.
20. Wang, C., et al., Graphene wrinkling: formation, evolution and collapse. Nanoscale, 2013. 5(10): p. 4454-4461.
21. Baheti, P., Train test validation split: How to & best practices [2023]. V7.(),[Online]. Available: https://www. v7labs. com/blog/train-validationtest-set (visited on 05/20/2023), 2021.
22. Xu, Y. and R. Goodacre, On splitting training and validation set: a comparative study of cross-validation, bootstrap and systematic sampling for estimating the generalization performance of supervised learning. Journal of analysis and testing, 2018. 2(3): p. 249-262.
23. Isa, I.S., et al., Optimizing the hyperparameter tuning of YOLOv5 for underwater detection. IEEE Access, 2022. 10: p. 52818-52831.
24. Buhl, N., Training, Validation, Test split for machine learning Datasets, in Encord Blog. 2023.
25. Raheja, S., A Comprehensive Guide to Train-Test-Validation Split in 2024, in Analytics Vidhya. 2024: Analytics Vidhya.
26. Ruchix18, Stratified Sampling in Pandas, in Stratified Sampling in Pandas. 2021, GeeksforGeeks: GeeksforGeeks.
27. Zhang, L., et al., A lightweight detection algorithm for unmanned surface vehicles based on multi-scale feature fusion. Journal of Marine Science and Engineering, 2023. 11(7): p. 1392.
28. Kusharki, M.B., et al., Automatic classification of equivalent mutants in mutation testing of android applications. Symmetry, 2022. 14(4): p. 820.

指導教授

蘇清源(Su, Ching Yuan)

審核日期

2025-3-27

推文