請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/101789完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 莊永裕 | zh_TW |
| dc.contributor.advisor | Yung-Yu Chuang | en |
| dc.contributor.author | 邱彥慈 | zh_TW |
| dc.contributor.author | Yen-Tzu Chiu | en |
| dc.date.accessioned | 2026-03-04T16:34:50Z | - |
| dc.date.available | 2026-03-05 | - |
| dc.date.copyright | 2026-03-04 | - |
| dc.date.issued | 2026 | - |
| dc.date.submitted | 2026-02-10 | - |
| dc.identifier.citation | Daochen Zha, Zaid Pervaiz Bhat, Kwei-Herng Lai, Fan Yang, Zhimeng Jiang, Shaochen Zhong, and Xia Hu. Data-centric artificial intelligence: A survey. ACM Comput. Surv., 57(5):1-42, May 2025.
Luis Augusto Libório Oliveira Fonseca, Yuzo Iano, Gabriel Gomes de Oliveira, Gabriel Caumo Vaz, Giulliano Paes Carnielli, Júlio César Pereira, and Rangel Arthur. Automatic printed circuit board inspection: a comprehensible survey. Discover Artificial Intelligence, 4(1):10, 2024. Nikhila Ravi, Valentin Gabeur, Yuan-Ting Hu, Ronghang Hu, Chaitanya Ryali, Tengyu Ma, Haitham Khedr, Roman Rädle, Chloe Rolland, Laura Gustafson, et al. Sam 2: Segment anything in images and videos. arXiv preprint arXiv:2408.00714, 2024. Sanli Tang, Fan He, Xiaolin Huang, and Jie Yang. Online pcb defect detector on a new pcb defect dataset, 2019. Weibo Huang, Peng Wei, Manhua Zhang, and Hong Liu. Hripcb: a challenging dataset for pcb defects detection and classification. The Journal of Engineering, 2020, 05 2020. Shengping Lv, Bin Ouyang, Zhihua Deng, Tairan Liang, Shixin Jiang, Kaibin Zhang, Jianyu Chen, and Zhuohui Li. A dataset for deep learning based detection of printed circuit board surface defect. Scientific Data, 11(1):811, 2024. Diulhio Candido de Oliveira, Bogdan Tomoyuki Nassu, and Marco Aurelio Wehrmeister. Image-based detection of modifications in assembled pcbs with deep convolutional autoencoders. Sensors, 23(3), 2023. Minghui Shen, Yujie Liu, Jing Chen, Kangqi Ye, Heyuan Gao, Jie Che, Qingyang Wang, Hao He, Jian Liu, Yan Wang, and Ye Jiang. Defect detection of printed circuit board assembly based on yolov5. Scientific Reports, 14, 08 2024. Gianmauro Fontana, Maurizio Calabrese, Leonardo Agnusdei, Gabriele Papadia, and Antonio Del Prete. Soldef_ai: An open source pcb dataset for mask r-cnn defect detection in soldering processes of electronic components. Journal of Manufacturing and Materials Processing, 8(3), 2024. Raffaele Mineo, Amelia Sorrenti, Robin Faro, Gabriele Mineo, Francesco Cancelliere, and Alberto Faro. Pcb-said: A low-cost camera-based dataset for few-shot smd assembly inspection. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, pages 1351-1357, October 2025. Christopher Pramerdorfer and Martin Kampel. A dataset for computer-vision-based pcb analysis. In 2015 14th IAPR International Conference on Machine Vision Applications (MVA), pages 378-381, 2015. Gayathri Mahalingam, Kevin Marshall Gay, and Karl Ricanek. Pcb-metal: A pcb image dataset for advanced computer vision machine learning component analysis. In 2019 16th International Conference on Machine Vision Applications (MVA), pages 1-5, 2019. Hangwei Lu, Dhwani Mehta, Olivia P. Paradis, Navid Asadizanjani, Mark Mohammad Tehranipoor, and D. Woodard. Fics-pcb: A multi-modal image dataset for automated printed circuit board visual inspection. IACR Cryptol. ePrint Arch., 2020:366, 2020. Nathan Jessurun, Olivia P. Dizon-Paradis, Jacob Harrison, Shajib Ghosh, Mark M. Tehranipoor, Damon L. Woodard, and Navid Asadizanjani. Fpic: A novel semantic dataset for optical pcb assurance. J. Emerg. Technol. Comput. Syst., 19(2), May 2023. Elias Arbash, Margret Fuchs, Behnood Rasti, Sandra Lorenz, Pedram Ghamisi, and Richard Gloaguen. Pcb-vision: A multiscene rgb-hyperspectral benchmark dataset of printed circuit boards. IEEE Sensors Journal, 24(10):17140-17158, 2024. Paul Bergmann, Kilian Batzner, Michael Fauser, David Sattlegger, and Carsten Steger. Beyond dents and scratches: Logical constraints in unsupervised anomaly detection and localization. International Journal of Computer Vision, 130:947-969, 2022. Paul Bergmann, Michael Fauser, David Sattlegger, and Carsten Steger. Mvtec ad -a comprehensive real-world dataset for unsupervised anomaly detection. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 9584-9592, 2019. Fupei Wu, Xianmin Zhang, Yongcong Kuan, and Zhenzhen He. An aoi algorithm for pcb based on feature extraction. In 2008 7th World Congress on Intelligent Control and Automation, pages 240-247, 2008. Labellerr. Automated labeling: Revolutionizing data annotation with ai, February 2025. Accessed: 2025-10-13. Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer Whitehead, Alexander C. Berg, Wan-Yen Lo, Piotr Dollár, and Ross Girshick. Segment anything. arXiv:2304.02643, 2023. Nicolas Carion, Laura Gustafson, Yuan-Ting Hu, Shoubhik Debnath, Ronghang Hu, Didac Suris, Chaitanya Ryali, Kalyan Vasudev Alwala, Haitham Khedr, Andrew Huang, Jie Lei, Tengyu Ma, Baishan Guo, Arpit Kalla, Markus Marks, Joseph Greer, Meng Wang, Peize Sun, Roman Rädle, Triantafyllos Afouras, Effrosyni Mavroudi, Katherine Xu, Tsung-Han Wu, Yu Zhou, Liliane Momeni, Rishi Hazra, Shuangrui Ding, Sagar Vaze, Francois Porcher, Feng Li, Siyuan Li, Aishwarya Kamath, Ho Kei Cheng, Piotr Dollár, Nikhila Ravi, Kate Saenko, Pengchuan Zhang, and Christoph Feichtenhofer. Sam 3: Segment anything with concepts, 2025. Rishi Bom Masani, Drew A. Hudson, Ehsan Adeli, Russ Altman, Simran Arora, Sydney von Arx, Michael S. Bernstein, Jeannette Bohg, Antoine Bosselut, Emma Brunskill, Erik Brynjolfsson, Shyamal Buch, Dallas Card, Rodrigo Castellon, Niladri Chatterji, Annie Chen, Kathleen Creel, Jared Quincy Davis, Dora Demszky, Chris Donahue, Moussa Doumbouya, Esin Durmus, Stefano Ermon, John Etchemendy, Kawin Ethayarajh, Li Fei-Fei, et al. On the opportunities and risks of foundation models, 2021. Zheda Mai, Ping Zhang, Cheng-Hao Tu, Hong-You Chen, Quang-Huy Nguyen, Li Zhang, and Wei-Lun Chao. Lessons and insights from a unifying study of parameter-efficient fine-tuning (peft) in visual recognition. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 14845-14857, 2025. Tianrun Chen, Lanyun Zhu, Chaotao Ding, Runlong Cao, Yan Wang, Shangzhan Zhang, Zejian Li, Lingyun Sun, Ying Zang, and Papa Mao. Sam-adapter: Adapting segment anything in underperformed scenes. In 2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), pages 3359-3367, 2023. Tianrun Chen, Ankang Lu, Lanyun Zhu, Chaotao Ding, Chunan Yu, Deyi Ji, Zejian Li, Lingyun Sun, Papa Mao, and Ying Zang. Sam2-adapter: Evaluating and adapting segment anything 2 in downstream tasks: Camouflage, shadow, medical image segmentation, and more, 2024. Tianrun Chen, Runlong Cao, Xinda Yu, Lanyun Zhu, Chaotao Ding, Deyi Ji, Cheng Chen, Qi Zhu, Chunyan Xu, Papa Mao, and Ying Zang. Sam3-adapter: Efficient adaptation of segment anything 3 for camouflage object segmentation, shadow detection, and medical image segmentation, 2025. Enkai Zhang, Jingjing Liu, Anda Cao, Zhen Sun, Haofei Zhang, Huiqiong Wang, Li Sun, and Mingli Song. Rs-sam: Integrating multi-scale information for enhanced remote sensing image segmentation. In Proceedings of the Asian Conference on Computer Vision (ACCV), pages 994-1010, December 2024. Guangyu Ren, Hengyan Liu, Michalis Lazarou, and Tania Stathaki. Multi-modal segment anything model for camouflaged scene segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 19882-19892, October 2025. Jan-Martin O. Steitz and Stefan Roth. Adapters strike back. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 23449-23459, June 2024. Edward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. Lora: Low-rank adaptation of large language models. In ICLR. OpenReview.net, 2022. Shih-Yang Liu, Chien-Yi Wang, Hongxu Yin, Pavlo Molchanov, Yu-Chiang Frank Wang, Kwang-Ting Cheng, and Min-Hung Chen. Dora: weight-decomposed low-rank adaptation. In Proceedings of the 41st International Conference on Machine Learning, ICML'24. JMLR.org, 2024. Wei Dong, Xing Zhang, Bihui Chen, Dawei Yan, Zhijun Lin, Qingsen Yan, Peng Wang, and Yang Yang. Low-Rank Rescaled Vision Transformer Fine-Tuning: A Residual Design Approach. In 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 16101-16110, Los Alamitos, CA, USA, June 2024. IEEE Computer Society. Mouin Ben Ammar, Arturo Mendoza, Nacim Belkhir, Antoine Manzanera, and Gianni Franchi. Foundation models and transformers for anomaly detection: A survey. Information Fusion, 126:103517, 2026. Xiaofan Li, Zhizhong Zhang, Xin Tan, Chengwei Chen, Yanyun Qu, Yuan Xie, and Lizhuang Ma. Promptad: Learning prompts with only normal samples for few-shot anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 16838-16848, June 2024. Bin-Bin Gao, Yue Zhou, Jiangtao Yan, Yuezhi Cai, Weixi Zhang, Meng Wang, Jun Liu, Yong Liu, Lei Wang, and Chengjie Wang. Adaptclip: Adapting clip for universal visual anomaly detection. In AAAI, 2026. Simon Damm, Mike Laszkiewicz, Johannes Lederer, and Asja Fischer. Anomaly-dino: Boosting patch-based few-shot anomaly detection with dinov2. In Proceedings of the Winter Conference on Applications of Computer Vision (WACV 2025), 2025. Fatih Cagatay Akyon, Sinan Onur Altinuc, and Alptekin Temizel. Slicing aided hyper inference and fine-tuning for small object detection. In 2022 IEEE International Conference on Image Processing (ICIP), 2022. Bowen Cheng, Ross Girshick, Piotr Dollar, Alexander C. Berg, and Alexander Kirillov. Boundary iou: Improving object-centric image segmentation evaluation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 15334-15342, June 2021. Ranjan Sapkota, Konstantinos I. Roumeliotis, and Manoj Karkee. The sam2-to-sam3 gap in the segment anything model family: Why prompt-based expertise fails in concept-driven image segmentation, 2025. Yunjie Tian, Qixiang Ye, and David Doermann. YOLOv12: Attention-centric real-time object detectors. In The Thirty-ninth Annual Conference on Neural Information Processing Systems, 2025. Chunyu Xie, Bin Wang, Fanjing Kong, Jincheng Li, Dawei Liang, Ji Ao, Dawei Leng, and Yuhui Yin. Fg-clip 2: A bilingual fine-grained vision-language alignment model, 2025. | - |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/101789 | - |
| dc.description.abstract | 工業異常定位（Industrial Anomaly Localization）由於像素級標註稀缺，在高解析度且元件密集分佈的印刷電路板組裝（PCBA）圖像中仍面臨挑戰。本文提出一個源自真實 SMT 生產線的大規模數據集 AnomalyPCB，其特點在於包含空間對齊的元件圖像，涵蓋邏輯、結構和元件級異常的多樣化缺陷。為確保質量同時擴展標註規模，我們提出了一個自動標註工作流程，包含偽標籤生成（Pseudo-label Generation）、人工質量檢驗與迭代重訓練機制。此外，我們提出了一個用於自動掩碼生成的基礎架構, 將參數高效微調（Parameter-Efficient Fine-Tuning）後的 SAM2 與於空間對齊的異常定位模塊相結合。實驗顯示，現有的少樣本方法（Few-shot Methods）難以應對 PCBA 圖像中特定的領域複雜性。因此 AnomalyPCB 可作為工業檢測研究中其一具挑戰性的基準。 | zh_TW |
| dc.description.abstract | Industrial anomaly localization remains a significant challenge due to the scarcity of pixel-level annotations, particularly in high-resolution, densely populated printed circuit board assembly (PCBA) imagery. In this work, we introduce AnomalyPCB, a large-scale dataset curated from online SMT production lines. It is characterized by spatially aligned components and heterogeneous defect patterns, including logical, structural, and component-level anomalies. To scale annotation while maintaining quality, we propose an iterative labeling pipeline that integrates pseudo-label generation, quality refinement, and iterative retraining.Furthermore, we establish a baseline framework for automatic mask generation by leveraging a parameter-efficiently adapted SAM2 coupled with a spatially aligned anomaly localization module. Extensive experiments reveal that existing few-shot methods struggle with the domain-specific complexities of PCBA images. Consequently, AnomalyPCB establishes a novel, challenging benchmark for advancing future research in industrial inspection. | en |
| dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2026-03-04T16:34:50Z No. of bitstreams: 0 | en |
| dc.description.provenance | Made available in DSpace on 2026-03-04T16:34:50Z (GMT). No. of bitstreams: 0 | en |
| dc.description.tableofcontents | Verification Letter from the Oral Examination Committee i
Acknowledgements ii 摘要 iii Abstract iv Contents vi List of Figures x List of Tables xii Denotation xiii Chapter 1 Introduction 1 1.1 PCBA Anomaly Dataset 2 1.1.1 Gaps in General-Purpose Industrial Dataset and PCBA Dataset 3 1.2 Scope and Contributions 3 1.3 Thesis Organization 4 Chapter 2 AnomalyPCB Dataset 7 2.1 Overview of Dataset 7 2.2 Definition 8 2.2.1 Component ID (CID) 8 2.2.2 Defect Types 8 2.2.3 Format 9 2.2.4 Labeling Scheme 9 2.3 Labeling Pipeline 10 2.3.1 Data Collection 10 2.3.2 Data Preprocessing 11 2.3.3 Pseudo Label Generation 11 2.3.4 Quality Assertion 12 Chapter 3 Related Work 13 3.1 Segment Anything 13 3.2 Domain-Specific Adaptation of Segment Anything 13 3.3 PEFT 14 3.3.1 Adapter-Based Methods 14 3.3.2 Low-Rank Adaptation 15 3.4 Visual Anomaly Detection and Semantic Alignment 15 3.4.1 Unsupervised Visual Anomaly Detection 15 3.4.2 Vision-Language Models and Semantic Alignment 16 Chapter 4 Methodology 17 4.1 Pseudo-Label Generation 17 4.1.1 Problem Formulation 17 4.1.2 Overview of Labeling Framework 17 4.2 Segmentation Module 19 4.2.1 SAHI 19 4.2.2 ViT Adaptation 19 4.3 Anomaly Detection Module 19 4.3.1 Multi-Scale Feature Aggregation 20 4.3.2 Spatially-Aligned Anomaly Calculation 20 4.4 Postprocessing Module 21 4.4.1 Instance-Aware Gating and Thresholding 21 4.4.2 Defect-Specific Anomaly Assignment 22 4.4.2.1 Instance-Anomaly Matching 23 4.4.2.2 Defect-Specific Assignment Logic 24 Chapter 5 Experiments 27 5.1 Experimental Settings 27 5.1.1 Dataset 27 5.1.2 Implementation Details 27 5.1.3 Evaluation Metrics 28 5.2 Evaluation of Parameter-Efficient Fine-Tuning 29 5.2.1 Upper Bound Capacity Analysis 29 5.2.2 Generalization and Robustness Analysis 30 5.3 Comparison of Integration Strategy 31 5.4 Anomaly Detection 32 5.4.1 Defect Scheme and Preliminary 32 5.4.2 Analysis of CLIP-Based Approaches 34 5.4.3 Analysis of DINO-Based Approaches 35 5.4.4 Analysis of SAM2 Adaptation 35 5.5 Iterative Training and Analysis 36 5.6 Summary of Experimental Results 38 Chapter 6 Conclusion 43 6.1 Conclusion 43 6.2 Future Work 45 References 47 Appendix A — XML 55 A.1 CID 55 | - |
| dc.language.iso | zh_TW | - |
| dc.subject | 印刷電路板 | - |
| dc.subject | 異常檢測資料集 | - |
| dc.subject | 異常檢測 | - |
| dc.subject | 影像分割 | - |
| dc.subject | 視覺基礎模型 | - |
| dc.subject | Printed Circuit Board (PCB) | - |
| dc.subject | Industrial Anomaly Detection Dataset | - |
| dc.subject | Anomaly Detection | - |
| dc.subject | Image Segmentation | - |
| dc.subject | Vision Foundation Model | - |
| dc.title | AnomalyPCB：面向 PCB 邏輯與結構異常檢測的資料集建立 | zh_TW |
| dc.title | AnomalyPCB: A Comprehensive Dataset for Logical and Structural PCB Anomaly Detection | en |
| dc.type | Thesis | - |
| dc.date.schoolyear | 114-1 | - |
| dc.description.degree | 碩士 | - |
| dc.contributor.oralexamcommittee | 吳賦哲;葉正聖 | zh_TW |
| dc.contributor.oralexamcommittee | Fu-Che Wu;Jeng-Sheng Yeh | en |
| dc.subject.keyword | 印刷電路板,異常檢測資料集異常檢測影像分割視覺基礎模型 | zh_TW |
| dc.subject.keyword | Printed Circuit Board (PCB),Industrial Anomaly Detection DatasetAnomaly DetectionImage SegmentationVision Foundation Model | en |
| dc.relation.page | 55 | - |
| dc.identifier.doi | 10.6342/NTU202600394 | - |
| dc.rights.note | 未授權 | - |
| dc.date.accepted | 2026-02-10 | - |
| dc.contributor.author-college | 電機資訊學院 | - |
| dc.contributor.author-dept | 資訊網路與多媒體研究所 | - |
| dc.date.embargo-lift | N/A | - |
| 顯示於系所單位： | 資訊網路與多媒體研究所 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-114-1.pdf 未授權公開取用 | 5.54 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。