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：32

、訪客IP：3.145.82.50

姓名	吳明憲(Ming-Sian Wu)  查詢紙本館藏	畢業系所	資訊工程學系
論文名稱	使用無紋理之3D CAD工業零件模型結合長度檢測實現細粒度真實工業零件影像分類 (Fine-grained Real-world Industrial Component Image Classification with Untextured 3D CAD Industrial Component Model and Length Detection)
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 ( 永不開放)
摘要(中)	近年來，深度學習技術被大量應用在多個不同的領域，其中虛擬實境頭盔結合深度學習技術常被使用來提升生產效率。但是因為在訓練深度學習模型時會需要大量的訓練資料，而工廠的作業流程通常都是在接到訂單之後才會開始生產，因此很難預先取得大量的訓練資料供深度學習模型來訓練，故使用 3D 零件模型影像來訓練模型及辨識真實零件影像便是一個較為可行的解決方式。雖然使用 3D 零件模型影像來訓練模型可以解決缺乏訓練資料的問題，但是工廠提供的 3D 零件模型中並不一定會包含紋理資訊，導致其和真實零件還是有很大的差距，因此若要使用這些 3D 零件模型影像來訓練模型辨識真實零件會是一項困難的挑戰。而在生成的 3D 零件模型影像中，部分角度之零件影像會容易使模型混淆，且因為零件之間具有很高的相似性，所以導致模型會有預測準確度不佳的問題，因此本研究提出了依據零件面積比例來過濾資料集，去除掉較容易使模型混淆之影像，此外，本研究也提出了長度過濾模組來輔助模型推論，通過長度資訊篩選掉較不符合之類別，實驗結果顯示，我們提出的方法可以顯著提升模型在細粒度真實工業零件分類問題的表現。
摘要(英)	In recent years, deep learning technology has been widely used in various fields. One common application is combining virtual reality helmets with deep learning to improve production efficiency. However, training deep learning models requires a lot of data, which is difficult to obtain in factories where production starts only after orders are received. Using 3D models of components to train the model and recognize real component images is a feasible solution, but the lack of texture information in 3D models provided by the factory poses a significant challenge in accurately identifying real components. To address this, our study proposes a filtering method based on component area ratios to eliminate confusing images, and introduces a length filter module to assist model inference by filtering out mismatching size categories. Experimental results show that our methods significantly improve model performance in fine-grained real world industrial component classification tasks.
關鍵字(中)	★ 細粒度工業零件分類 ★ 3D CAD 模型影像合成 ★ 過濾資料集 ★ 長度過濾	關鍵字(英)	★ Fine-grained industrial component classification ★ 3D CAD based image rendering ★ Filter dataset ★ Length filter
論文目次	摘要 vi Abstract vii 目錄 viii 圖目錄 x 表目錄 xii 一、 緒論 1 1.1 研究背景 1 1.2 研究動機與目的 2 1.3 問題定義 4 1.4 研究貢獻 4 1.5 論文架構 5 二、 相關研究 6 2.1 基於 3D CAD 的資料增量 6 2.2 細粒度分類模型 7 三、 解決方案 9 3.1 3D 物件影像擷取及過濾模組 11 3.1.1 3D 物件模型影像擷取 11 3.1.2 資料集過濾 15 3.2 HERBS 分類模型模組 19 3.2.1 資料前處理 19 3.2.2 模型訓練及推論 22 3.3 長度過濾模組 23 viii 目錄 四、 實驗與結果討論 26 4.1 實驗資料集介紹 26 4.2 評估方法 33 4.3 交叉驗證 33 4.4 實驗一：準確度性能實驗 34 4.4.1 實驗動機與目的 34 4.4.2 實驗模型選擇及參數 34 4.4.3 實驗方法 35 4.4.4 實驗結果 37 4.5 實驗二：相似零件性能實驗 40 4.5.1 實驗動機與目的 40 4.5.2 實驗模型選擇及參數 40 4.5.3 實驗方法 40 4.5.4 實驗結果 40 4.6 Ablation study 44 4.6.1 實驗三：最佳擷取影像數量實驗 44 4.6.2 實驗四：最佳資料區間性能實驗 47 4.6.3 實驗五：長度過濾模組性能實驗 50 4.6.4 實驗六：相似零件之消融研究 (Ablation study) 實驗 54 五、 結論與未來展望 60 5.1 結論 60 5.2 未來展望 60 六、 參考文獻 62
參考文獻	[1] “Microsoft hololens: Mixed reality technology for business.” https://www.microsoft.com/en-us/hololens. Accessed: 2024-03-21. [2] S. J. Pan and Q. Yang, “A survey on transfer learning,” IEEE Transactions on knowledge and data engineering, vol. 22, no. 10, pp. 1345–1359, 2009. [3] K. Židek, P. Lazorík, J. Pitel’, and A. Hošovskỳ, “An automated training of deep learning networks by 3d virtual models for object recognition,” Symmetry, vol. 11, no. 4, p. 496, 2019. [4] J. Cohen, C. F. Crispim-Junior, C. Grange-Faivre, and L. Tougne, “Cad-based learning for egocentric object detection in industrial context,” in 15th International Conference on Computer Vision Theory and Applications, vol. 5, pp. 644–651, SCITEPRESS-Science and Technology Publications; SCITEPRESS-Science and …, 2020. [5] J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, “Imagenet: A large-scale hierarchicalimage database, ” in 2009 IEEE conference on computer vision and pattern recognition, pp. 248–255, Ieee, 2009. [6] J. Redmon and A. Farhadi, “Yolov3: An incremental improvement,” arXiv preprint arXiv:1804.02767, 2018. [7] H. Tavakoli, S. Walunj, P. Pahlevannejad, C. Plociennik, and M. Ruskowski, “Small object detection for near real-time egocentric perception in a manual assembly scenario,” arXiv preprint arXiv:2106.06403, 2021. [8] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “Yolov4: Optimal speed and accuracy of object detection,” arXiv preprint arXiv:2004.10934, 2020. [9] P. Ruediger-Flore, M. Glatt, M. Hussong, and J. C. Aurich, “Cad-based data augmentation and transfer learning empowers part classification in manufacturing,” The International Journal of Advanced Manufacturing Technology, vol. 125, no. 11, pp. 5605–5618, 2023. [10] K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556, 2014. [11] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770–778, 2016. [12] M. Tan and Q. Le, “Efficientnet: Rethinking model scaling for convolutional neural networks,” in International conference on machine learning, pp. 6105–6114, PMLR, 2019. [13] F. Chollet, “Xception: Deep learning with depthwise separable convolutions,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 1251–1258, 2017. [14] G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely connected convolutionalnetworks,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 4700–4708, 2017. [15] X. Zhu, T. Bilal, P. Mårtensson, L. Hanson, M. Björkman, and A. Maki, “Towards sim-to-real industrial parts classification with synthetic dataset,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 4453–4462, 2023. [16] Z. Liu, H. Mao, C.-Y. Wu, C. Feichtenhofer, T. Darrell, and S. Xie, “A convnet for the 2020s,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 11976–11986, 2022. [17] Q. Diao, Y. Jiang, B. Wen, J. Sun, and Z. Yuan, “Metaformer: A unified meta framework for finegrained recognition,” arXiv preprint arXiv:2203.02751, 2022. [18] P.-Y. Chou, C.-H. Lin, and W.-C. Kao, “A novel plug-in module for fine-grained visual classification,” arXiv preprint arXiv:2202.03822, 202 ㄉ. [19] P.-Y. Chou, Y.-Y. Kao, and C.-H. Lin, “Fine-grained visual classification with high-temperature refinement and background suppression,” arXiv preprint arXiv:2303.06442, 2023. [20] T.-Y. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan, and S. Belongie, “Feature pyramid networks for object detection,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 2117–2125, 2017. [21] G. Hinton, O. Vinyals, and J. Dean, “Distilling the knowledge in a neural network,” arXiv preprint arXiv:1503.02531, 2015. [22] Z. Liu, Y. Lin, Y. Cao, H. Hu, Y. Wei, Z. Zhang, S. Lin, and B. Guo, “Swin transformer: Hierarchical vision transformer using shifted windows,” in Proceedings of the IEEE/CVF international conference on computer vision, pp. 10012–10022, 2021. [23] C. Wah, S. Branson, P. Welinder, P. Perona, and S. Belongie, “The caltech-ucsd birds-200-2011 dataset,” California Institute of Technology, 2011. [24] G. Van Horn, S. Branson, R. Farrell, S. Haber, J. Barry, P. Ipeirotis, P. Perona, and S. Belongie, “Building a bird recognition app and large scale dataset with citizen scientists: The fine print in fine-grained dataset collection,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 595–604, 2015. [25] A. Buslaev, V. I. Iglovikov, E. Khvedchenya, A. Parinov, M. Druzhinin, and A. A. Kalinin, “Albumentations: fast and flexible image augmentations,” Information, vol. 11, no. 2, p. 125, 2020.
指導教授	梁德容 王尉任 林家瑜(Deron Liang Wei-Jen Wang Chia-Yu Lin)	審核日期	2024-7-31
推文	facebook   plurk   twitter   funp   google   live   udn   HD   myshare   reddit   netvibes   friend   youpush   delicious   baidu
網路書籤	Google bookmarks   del.icio.us   hemidemi   myshare