基於背景顏色可變之收集樣本影像與建立物件辨識資料集之方法

簡易檢索 / 詳目顯示

回結果列表

研究生：	張仁佑 Jen-You Zhang
論文名稱：	基於背景顏色可變之收集樣本影像與建立物件辨識資料集之方法 A Method of Collecting the Sample Images with the Color-Changeable Background and Establishing a Dataset for Object Recognition
指導教授：	李維楨 Wei-Chen Lee
口試委員:	孫沛立 Pei-li Sun 洪詩涵 Shih-Han Hung
學位類別：	碩士 Master
系所名稱：	工程學院 - 機械工程系 Department of Mechanical Engineering
論文出版年：	2019
畢業學年度：	107
語文別：	中文
論文頁數：	98
中文關鍵詞：	電腦視覺、影像處理、深度學習、物件辨識、Faster R-CNN 、影像合成、機械手臂
外文關鍵詞：	Computer Vision, Image Processing, Deep Learning, Object Recognition, Faster R-CNN, Image Synthesis, Robotic Arm
相關次數：	點閱：411 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

上一筆

摘要 I
Abstract II
誌謝 III
圖目錄 VII
表目錄 XI

第一章 緒論 1
1 研究動機 1
2 文獻回顧 2
3 研究目的 5

第二章 相關原理介紹 6
1 彩色影像之門檻值分割法 6
1.1 YCbCr色彩空間 6
2 影像去背與影像合成 7
2.1 影像去背(Image Matting) 7
2.2 Alpha合成(Alpha composition) 8
3 四種常見的影像辨識任務種類 10
4 卷積神經網路(Convolution neural network，CNN) 11
5 遷移式學習(Transfer learning) 12
6 Faster R-CNN物件偵測模型 13

第三章 實驗情境與實驗流程架構 14
1 實驗情境假設 14
2 實驗流程架構 15

第四章 基於可變背景收集商品影像 16
1 搭配機械手臂自動取得不同視野位置的物品影像 16
2 實驗硬體架構 19
3 收集影像流程 24
3.1 前置作業-取得彩色影像分割閾值 24
3.2 前置作業-取得實際距離與影像像素距離比值 27
3.3 尋找前景位置與計算機械手臂移動座標 28

第五章 提取商品前景與商品影像資料庫建立 37
1 提取商品前景影像處理流程 37
2 基於可變背景拍攝物品影像缺點 40
3 商品影像資料庫 41
4 收集真實影像集 43
4.1 人工影像資料標註 45

第六章 以影像合成生成訓練影像與自動標註 46
1 影像合成之商品擺放組合定義 46
1.1 物件影像插入背景影像可移動範圍定義 47
1.2 情境一、兩物體接觸且長軸相平行 48
1.3 情境二、兩物體接觸且長軸相垂直 49
1.4 情境三、兩物體分離 50
2 商品組合影像合成於背景與自動標註 51
2.1 影像合成流程 53
2.2 影像自動標註(Object detection標註類型) 56
2.3 影像自動標註(Instance segmentation標註類型) 57
3 以影像合成建立訓練資料集 58

第七章 實驗結果與討論 59
1 Faster R-CNN模型訓練參數設置 59
2 目標檢測模型評估指標 60
3 實驗一: 使用合成與真實訓練Faster R-CNN模型 62
3.1 實驗一: 訓練資料與測試資料分配 62
3.2 實驗一: Faster R-CNN訓練結果評估 63
4 實驗二: 單個及多個視野位置樣本影像訓練集模型評估 68
4.1 實驗二: 訓練資料與測試資料分配 68
4.2 實驗二: Faster R-CNN訓練結果評估 69

第八章 結論與未來展望 71
1 結論 71
2 未來展望 72

參考文獻 73
附錄 i


                                

[1] A. Krizhevsky, I. Sutskever, and G. E. Hinton, "ImageNet classification with deep convolutional neural networks," presented at the Proceedings of the 25th International Conference on Neural Information Processing Systems - Volume 1, Lake Tahoe, Nevada, 2012.
[2] M. Everingham, L. Gool, C. K. Williams, J. Winn, and A. Zisserman, "The Pascal Visual Object Classes (VOC) Challenge," Int. J. Comput. Vision, vol. 88, no. 2, pp. 303-338, 2010.
[3] J. Deng, W. Dong, R. Socher, L. Li, L. Kai, and F.-F. Li, "ImageNet: A large-scale hierarchical image database," in 2009 IEEE Conference on Computer Vision and Pattern Recognition, 20-25 June 2009 2009, pp. 248-255.
[4] T.-Y. Lin et al., "Microsoft COCO: Common Objects in Context."
[5] S. Mahapatra. Why Deep Learning over Traditional Machine Learning? Retrieved from https://towardsdatascience.com/why-deep-learning-is-needed-over-traditional-machine-learning-1b6a99177063 (June 5,2019).
[6] B. Sapp, A. Saxena, and A. Y. Ng, A Fast Data Collection and Augmentation Procedure for Object Recognition. 2008, pp. 1402-1408.
[7] H. Jo, Y. Na, and J. Song, "Data augmentation using synthesized images for object detection," in 2017 17th International Conference on Control, Automation and Systems (ICCAS), 18-21 Oct. 2017 2017, pp. 1035-1038.
[8] S. Ren, K. He, R. Girshick, and J. Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137-1149, 2017.
[9] D. Dwibedi, I. Misra, and M. Hebert, "Cut, Paste and Learn: Surprisingly Easy Synthesis for Instance Detection," in 2017 IEEE International Conference on Computer Vision (ICCV), 22-29 Oct. 2017 2017, pp. 1310-1319.
[10] J. Long, E. Shelhamer, and T. Darrell, "Fully convolutional networks for semantic segmentation," in 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 7-12 June 2015 2015, pp. 3431-3440.
[11] P. P, M. Gangnet, and A. Blake, "Poisson image editing," ACM Trans. Graph., vol. 22, no. 3, pp. 313-318, 2003.
[12] T. Kiyokawa, K. Tomochika, J. Takamatsu, and T. Ogasawara, "Fully Automated Annotation With Noise-Masked Visual Markers for Deep-Learning-Based Object Detection," IEEE Robotics and Automation Letters, vol. 4, no. 2, pp. 1972-1977, 2019.
[13] R. Dalal and T. Moh, "Fine-Grained Object Detection Using Transfer Learning and Data Augmentation," in 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 28-31 Aug. 2018 2018, pp. 893-896.
[14] W. Liu et al., "SSD: Single Shot MultiBox Detector."
[15] C. Szegedy et al., "Going deeper with convolutions," in 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 7-12 June 2015 2015, pp. 1-9.
[16] C. A. Poynton, A Technical Introduction to Digital Video. Wiley, 1996.
[17] A. R. Smith and J. F. Blinn, "Blue screen matting," presented at the Proceedings of the 23rd annual conference on Computer graphics and interactive techniques, 1996.
[18] J. Sun, Y. Li, S. B. Kang, and H.-Y. Shum, "Flash matting," ACM Trans. Graph., vol. 25, no. 3, pp. 772-778, 2006.
[19] MathWorks. Convolutional Neural Network 3 things you need to know. Retrieved from https://www.mathworks.com/solutions/deep-learning/convolutional-neural-network.html?s_tid=srchtitle (June 10,2019)
[20] 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, 2010.
[21] R. Girshick, "Fast R-CNN," in 2015 IEEE International Conference on Computer Vision (ICCV), 7-13 Dec. 2015 2015, pp. 1440-1448.
[22] A. a. Z. Dutta, Andrew, "The VIA Annotation Software for Images, Audio and Video," arXiv preprint arXiv:1904.10699, 2019, Art no. dutta2019vgg.

全文公開日期 2024/08/15 (校內網路)
全文公開日期本全文未授權公開 (校外網路)
全文公開日期本全文未授權公開 (國家圖書館：臺灣博碩士論文系統)

簡易檢索 / 詳目顯示

相關論文