請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/84107
完整後設資料紀錄
DC 欄位 | 值 | 語言 |
---|---|---|
dc.contributor.advisor | 傅楸善(Chiou-Shann Fuh) | |
dc.contributor.advisor | 傅楸善(Chiou-Shann Fuh | fuh@csie.ntu.edu.tw | ), | |
dc.contributor.author | Kai-Ching Yen | en |
dc.contributor.author | 閻楷青 | zh_TW |
dc.date.accessioned | 2023-03-19T22:04:51Z | - |
dc.date.copyright | 2022-07-29 | |
dc.date.issued | 2022 | |
dc.date.submitted | 2022-07-18 | |
dc.identifier.citation | [Abrams 2007] Deborah R. Abrams, “Introduction to Regression,” https://dss.princeton.edu/online_help/analysis/regression_intro.htm, 2007. [Anbarjafari 2014] G. Anbarjafari, “Digital Image Processing,” https://sisu.ut.ee/imageprocessing/book/1, 2014. [Bakshi 2020] C. Bakshi, “Random Forest Regression,” https://levelup.gitconnected.com/random-forest-regression-209c0f354c84, 2020. [Bevans 2020] R. Bevans, “Multiple Linear Regression | A Quick and Simple Guide,” https://www.scribbr.com/statistics/multiple-linear-regression/, 2020. [DeepCast 2022] DeepCast, “Going Beyond K-means for Effective Clustering of Production Curves,” https://www.deepcast.ai/blog/going-beyond-k-means-for-effective-clustering-of-production-curves, 2017. [Doğrul 2021] B. Doğrul, “Overview of Recommender Systems and Implementations,” https://medium.com/geekculture/overview-of-recommender-systems-and-implementations-cae13088369, 2021. [Educative 2022] Educative, Inc. Edpresso Team, “What Is Feature Extraction?,” https://www.educative.io/edpresso/what-is-feature-extraction, 2022. [Faysal 2021] Faysal Mh, “Polynomial Linear Regression,” https://medium.com/@faysal.miah01/polynomial-linear-regression-ee4cb9db34a5, 2021. [Fernando 2021] J. Fernando, “R-Squared,” https://www.investopedia.com/terms/r/r-squared.asp, 2021. [Firebolt 2022] Firebolt Analytics, “Data Flattening and Data Unflattening,” https://www.firebolt.io/glossary-items/data-flattening-and-data-unflattening, 2022. [Lee 2019] D. Lee, “Reinforcement Learning, Part 1: A Brief Introduction,” https://medium.com/ai%C2%B3-theory-practice-business/reinforcement-learning-part-1-a-brief-introduction-a53a849771cf, 2019. [MathWorks 2022] The MathWorks, Inc., “Feature Extraction For Machine Learning And Deep Learning,” https://www.mathworks.com/discovery/feature-extraction.html, 2022. [Mortensen 2020] D. Mortensen, “Converting nested JSON structures to Pandas DataFrames,” https://medium.com/swlh/converting-nested-json-structures-to-pandas-dataframes-e8106c59976e, 2020. [Netflix 2022] Netflix, “Netflix,” https://www.netflix.com/tw/, 2022. [Opengate 2015] Opengate, “Clustering: K-means Algorithm,” https://mropengate.blogspot.com/2015/06/ai-ch16-5-k-introduction-to-clustering.html, 2015. [Pandas 2022] Pandas, “pandas.DataFrame,” https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html, 2022. [Sayad 2022] S. Sayad, “Decision Tree - Regression,” https://www.saedsayad.com/decision_tree_reg.htm, 2022. [scikit-learn 2022] scikit-learn, “sklearn.cluster.KMeans,” https://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html, 2022. [Tech Wrench 2021] Tech Wrench, “System Design Interview: Recommendation System Design (As Used by YouTube, Netflix etc.),” https://medium.com/double-pointer/system-design-interview-recommendation-system-design-as-used-by-youtube-netflix-etc-c457aaec3ab, 2021. [TRI 2022] Test Research Inc., “Test Research Inc.,” https://www.tri.com.tw/en/index.html, 2022. [Thomas 2019] C. Thomas, “An Introduction to Convolutional Neural Networks,” https://towardsdatascience.com/an-introduction-to-convolutional-neural-networks-eb0b60b58fd7, 2019. [Tyagi 2019 BRIEF] D. Tyagi, “Introduction to BRIEF (Binary Robust Independent Elementary Features),” https://medium.com/data-breach/introduction-to-brief-binary-robust-independent-elementary-features-436f4a31a0e6 , 2019. [Tyagi 2019 FAST] D. Tyagi, “Introduction to FAST (Features from Accelerated Segment Test),” https://medium.com/data-breach/introduction-to-fast-features-from-accelerated-segment-test-4ed33dde6d65, 2019. [Tyagi 2019 ORB] D. Tyagi, “Introduction to ORB (Oriented FAST and Rotated BRIEF),” https://medium.com/data-breach/introduction-to-orb-oriented-fast-and-rotated-brief-4220e8ec40cf, 2019. [Verma 2019] S. Verma, “Train Your Lunar-Lander | Reinforcement Learning | OpenAIGYM,” https://shiva-verma.medium.com/solving-lunar-lander-openaigym-reinforcement-learning-785675066197, 2019. [Wendy 2020] Wendy, “Data Cleaning Techniques and Tools,” https://whatagraph.com/blog/articles/data-cleaning, 2020. | |
dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/84107 | - |
dc.description.abstract | 本篇論文提出了對於瑕疵檢測機台的檢測過程中,不同特性組成的印刷電路板影像,所對應的最適當機台檢測參數組,藉此達到良好的檢測品質,並且同時降低檢測時間成本。因為在機台做瑕疵檢測的過程當中,現場應用工程師 (FAE: Field Application Engineer),也就是客戶服務工程師會對客戶送進來的一批新的印刷電路板,根據自身經驗作分析並給出一個優良的參數組合讓以供機台檢測;然而每一輪的機台檢測結果都需要進一步調整,一次次的達到最合適的參數組,這個過程不僅費時費財,也消耗現場應用工程師的人力成本,因此我們提出一套系統做參數推薦,藉以替代現場應用工程師在前半段一步步調整參數組的動作,以期達到參數推薦自動化的效果,希望在新機台或新印刷電路板初次檢測時能快速縮短起始時間,找出優良檢測參數,達成優良檢測成果,也就是低漏檢率與低假警報。近年來,強化學習 (Reinforcement Learning)在機器學習與深度學習上都有蓬勃的發展,我們從影像抽取方向性與旋轉特徵 (ORB: Oriented FAST (Features from Accelerated Segment Test) and Rotated BRIEF (Binary Robust Independent Elementary Features))並分群 (Clustering),並根據每一群去做參數組初始值的設定;接著加上機器學習去讓我們的系統了解到每一群在每一個情境下最合適的下一步(如何調整各個參數)。在一組新的影像輸入後,我們的系統會找尋最符合這組影像的群,接著給予最佳的初始參數組以及後續的調整參數步驟,達到參數推薦自動化的效果。 | zh_TW |
dc.description.abstract | In this thesis, we propose a parameter recommendation system for defect inspection machine due to different characteristics of printed circuit board images, to reach excellent inspection quality and also reduce initial time cost. During defect inspection process, FAE (Field Application Engineer) must search and configure excellent parameter set for a batch of new printed circuit boards according to their own experiences. However, each result requires further adjustment to reach the most appropriate parameter set step by step. This procedure is not only time-consuming but also wasting labor costs. Therefore, we propose parameter recommendation system to accelerate those adjustments by FAE initially, so we can recommend parameter set automatically. Recently, machine learning flourishes. We extract image features to be inspected by ORB, cluster on those features, and set up the initial values for the parameter set. Next, we apply machine learning methods to let our system understand which is the best choice for next step on each cluster in every situation. When a batch of new printed circuit board images arrive, our system will find the most suitable cluster for them, give them good initial values as well as the following adjusting steps, to achieve automatically recommending parameter set. | en |
dc.description.provenance | Made available in DSpace on 2023-03-19T22:04:51Z (GMT). No. of bitstreams: 1 U0001-1507202220435300.pdf: 3582865 bytes, checksum: 5a65689c2c5e60c081e5480ad983da2c (MD5) Previous issue date: 2022 | en |
dc.description.tableofcontents | 誌謝 i 中文摘要 ii ABSTRACT iv CONTENTS vi LIST OF FIGURES ix LIST OF TABLES xii Chapter 1 Introduction 1 1.1 Recommendation System 1 1.1.1 Common Recommendation System 1 1.1.2 Our Parameter Recommendation System 2 1.2 Data Analysis 4 1.2.1 Data Cleaning 4 1.2.2 Data Parsing and Flattening 5 1.3 Image Processing 6 1.3.1 Feature Extraction 6 1.3.2 Clustering 8 1.4 Regression Modeling 9 1.5 Thesis Organization 11 Chapter 2 Related Works 12 2.1 Convolutional Neural Network 12 2.2 Reinforcement Learning 13 2.2.1 Lunar Lander 14 Chapter 3 Methodology 16 3.1 Overview 16 3.2 Data Augmentation 17 3.3 Steps of Parameter Set Recommendation 18 3.3.1 Recommendation System 18 3.3.2 Cold Start Problem for Initial Values 19 3.3.3 ORB Feature Extractions 20 3.3.4 Clustering 23 3.3.5 Data Cleaning, Parsing, and Analysis 24 3.3.6 Regression Modeling 28 3.3.7 Evaluation 31 Chapter 4 Experimental Results 33 4.1 Overview 33 4.2 Datasets 33 4.3 Results 38 4.4 Evaluation 44 4.5 Performance 46 Chapter 5 Conclusion and Future Works 48 References 50 | |
dc.language.iso | en | |
dc.title | 閻推薦: X光影像印刷電路板瑕疵檢測參數推薦系統 | zh_TW |
dc.title | YenRecommend: Parameter Recommendation for Printed Circuit Board Defect Inspection System for X-ray Images | en |
dc.type | Thesis | |
dc.date.schoolyear | 110-2 | |
dc.description.degree | 碩士 | |
dc.contributor.oralexamcommittee | 賴志宏(Jr-Hung Lai),沈立健(Li-Jian Shen), 巫宗昇(Tzung-Sheng Wu) | |
dc.subject.keyword | 瑕疵檢測,機器學習,迴歸分析,閻推薦,推薦系統,參數推薦,資料分析,資料清理,影像處理,強化學習, | zh_TW |
dc.subject.keyword | Defect Inspection,Machine Learning,Regression,YenRecommend,Recommendation System,Parameter Recommendation,Data Analysis,Data Cleaning,Image Processing,Reinforcement Learning, | en |
dc.relation.page | 53 | |
dc.identifier.doi | 10.6342/NTU202201486 | |
dc.rights.note | 同意授權(限校園內公開) | |
dc.date.accepted | 2022-07-18 | |
dc.contributor.author-college | 電機資訊學院 | zh_TW |
dc.contributor.author-dept | 資訊工程學研究所 | zh_TW |
dc.date.embargo-lift | 2027-07-15 | - |
顯示於系所單位: | 資訊工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
U0001-1507202220435300.pdf 目前未授權公開取用 | 3.5 MB | Adobe PDF | 檢視/開啟 |
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。