臺灣博碩士論文加值系統

目錄
摘要 I
目錄 III
表目錄 IV
圖目錄 V
第一章 簡介 1
第二章 文獻探討 4
第三章 皮革檢測的可行性分析 7
3.1 皮革瑕疵樣品與取像環境 8
3.2 YOLOV3模型訓練 16
第四章 雛形系統展示 32
4.1 深度學習影像辨識雛型系統 32
第五章 結論 40
參考文獻 42
附錄一 色差檢測方法的分析評估 46
附錄二 混淆矩陣 52
 
表目錄
表1：第一批皮革瑕疵取像的瑕疵種類 2
表2：第一階段的分析結果 2
表3：第二階段的分析結果 2
表4：第三階段的分析結果 2
表5：第二批皮革瑕疵取像的瑕疵種類 2
表6：第四階段的分析結果 2
表7：第四階段模型對每一種類瑕疵的召回率 2
表8：第五階段的分析結果 2
表9：第五階段模組對每一種類瑕疵的召回率 2
表10：圖像檢測時間分析 2
表11：三張圖像的RGB數據分析 2
表12：在不同環境配置下的取像對比結果 2
 
圖目錄
圖1：人造皮革 1
圖2：公司所提供的人造皮革 2
圖3：不同顏色與花紋的皮革 3
圖4：YOLOv3與其他物件偵測演算法的性能比較[20] 7
圖5：NG的樣品 8
圖6：不同光源模組下的取像情況 10
圖7：實際硬體的架設環境 11
圖8：皮革瑕疵影像 13
圖9：皮革瑕疵影像 14
圖10：皮革瑕疵影像 15
圖11：瑕疵影像的標記示意圖 15
圖12：第一階段檢測結果 18
圖13：“點”與“線”的標記示意圖 19
圖14：第二階段檢測結果 20
圖15：第三階段檢測結果 22
圖16：一、二批取像的差異 23
圖17：第四階段測試結果 27
圖18：第五階段測試結果 31
圖19：系統主頁面 32
圖20：系統運行流程圖 34
圖21：連接相機 36
圖22：輸出影像串流 37
圖23：檢測與標記 38
圖24：讀取影像 39
圖25：檢測與標記 39
圖26：不同光源的取像結果 47
圖27：不同解析度的取像結果 48
圖28：不同解析度下，三種皮革影像的差異 49

參考文獻
[1]M. Dill, “Leather,” Wikipedia[Online]. Available: https://en.wikipedia.org/wiki/Leather, 14 July 2020.
[2]M. K. Kasi, J B. Rao and V. K. Sahu, “Identification of Leather Defects using An Autoadaptive Edge Detection Image Processing Algorithm,” 2014 International Conference on High Performance Computing and Applications (ICHPCA), 19 February 2015.
[3]T.-H. H. Lee, “Edge Detection Analysis,” [Online]. Available: http://disp.ee.ntu.edu.tw/henry/edge_detection.pdf, 27 May 2020.
[4]T. Qiu, Y. Yan, G. Lu, “An Autoadaptive Edge-Detection Algorithm for Flame and Fire Image Processing,” IEEE Transactions on Instrumentation and Measurement, VOL. 61, NO. 5, MAY 2012.
[5]T. Qiu, Y. Yan and G. Lu “A New Edge Detection Algorithm for Flame Image Processing,” 2011 IEEE International Instrumentation and Measurement Technology Conference, 10-12 May 2011.
[6]Ap, “Laplace transform,” Wikipedia[Online]. Available: https://en.wikipedia.org/wiki/Laplace_transform, 18 July 2020.
[7]F. Dahuang, D. Lei and D. Jiehang, “Automatic Detection and Localization of Surface Defects for Whole Piece of Ultrahigh-definition Leather Images,” 2019 IEEE 4th International Conference on Computer and Communication Systems (ICCCS), 02 September 2019.
[8]J. Niesen, “Generalized minimal residual method,” Wikipedia[Online]. Available: https://en.wikipedia.org/wiki/Generalized_minimal_residual_method, 1 July 2020.
[9]M. Mathematicae, “Otsu's method,” Wikipedia[Online]. Available: https://en.wikipedia.org/wiki/Otsu%27s_method, 11 May 2020.
[10]C. Matthews, “Discrete wavelet transform,” Wikipedia[Online]. Available: https://en.wikipedia.org/wiki/Discrete_wavelet_transform, 9 July 2020.
[11]D.-M. Tsai and B. Hsiao, “Automatic surface inspection using wavelet reconstruction,” Pattern Recognition, Volume 34, Issue 6, Pages 1285-1305, June 2001.
[12]P. J. Navarro, C. Fernández-Isla, P. M. Alcover and J. Suardíaz, “Defect Detection in Textures through the Use of Entropy as a Means for Automatically Selecting the Wavelet Decomposition Level,” Sensors, 17 July 2016.
[13]L. Jian, H. Wei and H. Bin, “Research on Inspection and Classification of Leather Durface Defects Based on Neural Network and Decision Tree,” 2010 International Conference On Computer Design and Applications, 09 August 2010.
[14]Klbrain, “Feedforward neural network,” Wikipedia[Online]. Available: https://en.wikipedia.org/wiki/Feedforward_neural_network, 5 June 2020.
[15]P. Villar, M. Mora and P. Gonzalez, “A New Approach for Wet Blue Leather Defect Segmentation,” Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications, pp 591-598, 2011.
[16]J. M. R.S. Tavares, J. R.M. Natal, “Defect Detection in Raw Hide and Wet Blue Leather,” Proceedings of the International Symposium CompIMAGE 2006, 20-21 October 2006.
 
[17]C. Yeha and D.-B. Perng, “Establishing a Demerit Count Reference Standard for the Classification and Grading of Leather Hides,” The International Journal of Advanced Manufacturing Technology, volume 18, pages731–738, 2001.
[18]W. P. Amorim and R, Viana, “Desenvolvimento de um Software de Processamento e Gerac¸ao de Imagens para ˜ Classificac¸ao de Couro Bovino,” [Online]. http://www.gpec.ucdb.br/pistori/publicacoes/amorim_sibgrapi2006.pdf, 2006.
[19]Arcenciel, “Confusion matrix,” Wikipedia[Online]. Available: https://en.wikipedia.org/wiki/Feature_selection, 2 July 2020.
[20]H.-Q. Bong, Q.-B. Truong, H.-C. Nguyen and M.-T. Nguyen, “Vision-based Inspection System for Leather Surface Defect Detection and Classification,” 2018 5th NAFOSTED Conference on Information and Computer Science (NICS), 10 January 2019.
[21]Sepervinco, “Support vector machine,” Wikipedia[Online]. Available: https://en.wikipedia.org/wiki/Support_vector_machine, 17 July 2020.
[22]J. Redmon and A. Farhadi, “YOLOv3: An Incremental Improvement,” arXiv[Online]. Available: https://arxiv.org/pdf/1804.02767.pdf, 8 April 2018.
[23]T.-Y. Lin, P. Goyal, R. Girshick, K. He and P. Dollar, “Focal Loss for Dense Object Detection,” The IEEE International Conference on Computer Vision (ICCV), 2017.
[24]A. Bochkovskiy, C.-Y. Wang, H.-Y. M. Liao, “YOLOv4: Optimal Speed and Accuracy of Object Detection,” arXiv[Online]. Available: https://arxiv.org/pdf/2004.10934.pdf, 23 April 2020.
[25]C. Larrea, “Confusion matrix,” Wikipedia[Online]. Available: https://en.wikipedia.org/w/index.php?title=Confusion_matrix&action=history, 13 July 2020.