跳到主要內容

臺灣博碩士論文加值系統

(44.222.64.76) 您好!臺灣時間:2024/06/16 03:30
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:周威圻
研究生(外文):CHOU, WEI-CHI
論文名稱:基於深度卷積神經網路於紡織布料之瑕疵檢測
論文名稱(外文):Deep Convolutional Neural Network based Defect Detection for Fabric Inspection
指導教授:何昭慶何昭慶引用關係
指導教授(外文):HO, CHAO-CHING
口試委員:何昭慶范光照朱志良章明
口試委員(外文):HO, CHAO-CHINGFAN, KUANG-CHAOCHU, CHIH-LIANGCHANG, MING
口試日期:2021-07-29
學位類別:碩士
校院名稱:國立臺北科技大學
系所名稱:機械工程系機電整合碩士班
學門:工程學門
學類:機械工程學類
論文種類:學術論文
論文出版年:2021
畢業學年度:109
語文別:中文
論文頁數:88
中文關鍵詞:深層卷積神經網路瑕疵檢測自動光學檢測數位影像處理深度學習網路優化剪枝參數
外文關鍵詞:Deep convolutional neural networkDefect detectionAutomated optical inspectionDigital image processingDeep learning network optimizationPruning parameter
相關次數:
  • 被引用被引用:0
  • 點閱點閱:232
  • 評分評分:
  • 下載下載:3
  • 收藏至我的研究室書目清單書目收藏:0
本研究針對布料表面進行瑕疵檢測與深度學習網路優化,由於影像處理無法有效解決具有複雜背景之布料,本研究採用深度學習進行瑕疵檢測,然而現今網路架構主要著重於自然影像(Natural image)辨識如動物或人臉而非瑕疵,導致使用於瑕疵檢測之網路架構具有較多冗餘神經元,造成推理速度降低,因此本研究使用網路剪枝搭配貝氏優化演算法自動調整網路剪枝參數,再將剪枝後網路進行再訓練,檢測過程則利用上述剪枝後網路預測缺陷信心圖,再根據本研究提出之影像處理流程作為布料瑕疵最終判斷依據,並將提出之方法驗證於二筆自製資料集與二筆公開資料集;另於提出之網路優化結果部分,四種資料集交聯比(Intersection over Union, IoU)相對於原始模型下降1.26%、1.13%、1.21%與2.15%,但於Geforce 2080Ti計算下可縮短預測時間為原預測時間的20.84%、40.52%、23.02與23.33%,且於嵌入式系統AGX Xavier中同樣也可縮短預測時間為原始時間的17.56%、37.03%、19.67%與22.26%;於影像處理部分,四種資料集準確度則可分別達92.75%、94.87%、95.6%與81.82%,而本研究更使用布料瑕疵訓練Yolov4,結果可得該架構並不利於預測狹長型布料瑕疵。
This research is aimed to detect defects on the surface of the fabric and deep learning model optimization. Since defect detection cannot effectively solve the fabric with complex background by image processing, this research uses deep learning to identify defects. However, the current network architecture mainly focuses on natural images rather than the defect detection. As a result, the network architecture used for defect detection has more redundant neurons, which reduces the inference speed. In order to solve the above problems, we propose network pruning with Bayesian optimization algorithm to automatically tune the network pruning parameters, and then retrain the network after pruning. The training and detection process uses the above-mentioned pruning network to predict the defect feature map, and then uses the image processing flow proposed in this research for the final judgment during fabric defect detection. The proposed method is verified in the two self-made datasets and two public datasets. In the part of the proposed network optimization results, the Intersection over Union(IoU) of four datasets are dropped by 1.26%, 1.13%, 1.21%, and 2.15% compared to the original network model, but the inference time is reduced to 20.84%, 40.52%, 23.02 and 23.33% of the original network model using Geforce 2080 Ti. Furthermore, the inference time is also reduced to 17.56%, 37.03%, 19.67% and 22.26% using the embedded system AGX Xavier. After the image processing part, the accuracy of the four datasets can reach 92.75%, 94.87%, 95.6% and 81.82%, respectively. In this research, Yolov4 is also trained with fabric defects, and the results showed this model is not conducive to detecting long and narrow fabric defects.
摘要 i
ABSTRACT ii
誌謝 iv
目錄 v
表目錄 vii
圖目錄 viii
第一章 緒論 1
1.1 前言 1
1.2 研究動機 2
1.3 文獻回顧 2
1.3.1 紡織品缺陷檢測 2
1.3.1.1 統計方法 3
1.3.1.2 光譜方法 4
1.3.1.3 深度學習方法 5
1.3.2 模型壓縮 9
1.3.2.1 網路剪枝 9
1.3.2.2 知識蒸餾 10
1.3.2.3 參數量化 11
1.3.2.4 架構設計 12
1.4 論文架構 13
第二章 檢測資料集製作 14
2.1 自製資料集 14
2.1.1 取像機構 14
2.1.1.1 布料A 16
2.1.1.2 布料B 17
2.2 公開資料集 18
2.2.1 針織資料集 18
2.2.2 奈米纖維 18
第三章 深度學習與瑕疵檢測 20
3.1 神經網路介紹 20
3.1.1 卷積層 22
3.1.2 池化層 22
3.1.3 激勵函數 24
3.1.4 損失函數 25
3.1.5 反向傳遞 26
3.1.6 梯度下降法 27
3.2 瑕疵分類與瑕疵分割 28
3.2.1 瑕疵分類 29
3.2.1.1 布料A預測結果 31
3.2.1.2 布料B預測結果 33
3.2.1.3 針織資料集預測結果 34
3.2.1.4 奈米纖維預測結果 35
3.2.2 瑕疵分割 36
3.2.2.1 布料A預測結果 38
3.2.2.2 布料B預測結果 39
3.2.2.3 針織資料集預測結果 39
3.2.2.4 奈米纖維預測結果 41
3.2.3 實驗結果分析 41
第四章 網路優化演算方法 43
4.1 網路優化研究方法 43
4.1.1 網路優化方法 43
4.1.1.1 UNet++ 43
4.1.1.2 網路剪枝與再訓練 44
4.1.1.3 貝氏超參數優化 46
4.1.2 網路優化流程 46
4.2 影像處理研究方法 48
4.2.1 影像處理流程 48
第五章 實驗結果與討論 50
5.1 網路優化結果與討論 50
5.1.1 布料A 50
5.1.2 布料B 53
5.1.3 針織資料集 56
5.1.4 奈米纖維資料集 59
5.1.5 網路優化於嵌入式系統 62
5.1.6 網路優化結果分析 66
5.2 影像處理結果 68
5.2.1 布料A 68
5.2.2 布料B 71
5.2.3 針織資料集 74
5.2.4 奈米纖維資料集 77
5.2.5 影像處理結果分析 81
第六章 結論與未來展望 82
6.1 結論 82
6.2 未來展望 82
文獻參考 84

[1] S. S. T. Selvi and G. Nasira, "An effective automatic fabric defect detection system using digital image processing," J. Environ. Nanotechnol, vol. 6, no. 1, pp. 79-85, 2017.
[2] O. Ronneberger, P. Fischer, and T. Brox, "U-net: Convolutional networks for biomedical image segmentation," in International Conference on Medical image computing and computer-assisted intervention, 2015: Springer, pp. 234-241.
[3] A. Bochkovskiy, C.-Y. Wang, and H. Liao, "YOLOv4: Optimal Speed and Accuracy of Object Detection," ArXiv, vol. abs/2004.10934, 2020.
[4] Y.-F. Chen, F.-S. Yang, E. Su, and C.-C. Ho, "Automatic defect detection system based on deep convolutional neural networks," in 2019 International Conference on Engineering, Science, and Industrial Applications (ICESI), 2019: IEEE, pp. 1-4.
[5] D. Zhu, R. Pan, W. Gao, and J. Zhang, "Yarn-dyed fabric defect detection based on autocorrelation function and GLCM," Autex research journal, vol. 15, no. 3, pp. 226-232, 2015.
[6] C.-h. Chan and G. K. Pang, "Fabric defect detection by Fourier analysis," IEEE transactions on Industry Applications, vol. 36, no. 5, pp. 1267-1276, 2000.
[7] T. O. Ayodele, "Types of machine learning algorithms," New advances in machine learning, vol. 3, pp. 19-48, 2010.
[8] P. Bergmann, S. Löwe, M. Fauser, D. Sattlegger, and C. Steger, "Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoders," in VISIGRAPP, 2019.
[9] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, "Image quality assessment: from error visibility to structural similarity," IEEE transactions on image processing, vol. 13, no. 4, pp. 600-612, 2004.
[10] J. Liu, C. Wang, H. Su, B. Du, and D. Tao, "Multistage GAN for fabric defect detection," IEEE Transactions on Image Processing, vol. 29, pp. 3388-3400, 2019.
[11] K. Simonyan and A. Zisserman, "Very Deep Convolutional Networks for Large-Scale Image Recognition," CoRR, vol. abs/1409.1556, 2015.
[12] J. Johnson, A. Alahi, and L. Fei-Fei, "Perceptual losses for real-time style transfer and super-resolution," in European conference on computer vision, 2016: Springer, pp. 694-711.
[13] J. F. Jing, H. Ma, and H. H. Zhang, "Automatic fabric defect detection using a deep convolutional neural network," Coloration Technology, vol. 135, no. 3, pp. 213-223, 2019.
[14] J. Jing, Z. Wang, M. Rätsch, and H. Zhang, "Mobile-Unet: An efficient convolutional neural network for fabric defect detection," Textile Research Journal, p. 0040517520928604, 2020.
[15] D. Eigen and R. Fergus, "Predicting depth, surface normals and semantic labels with a common multi-scale convolutional architecture," in Proceedings of the IEEE international conference on computer vision, 2015, pp. 2650-2658.
[16] W. Wen, C. Wu, Y. Wang, Y. Chen, and H. H. Li, "Learning Structured Sparsity in Deep Neural Networks," in NIPS, 2016.
[17] G. Hinton, O. Vinyals, and J. Dean, "Distilling the knowledge in a neural network," arXiv preprint arXiv:1503.02531, 2015.
[18] S. Christodoulidis, M. Anthimopoulos, L. Ebner, A. Christe, and S. Mougiakakou, "Multisource transfer learning with convolutional neural networks for lung pattern analysis," IEEE journal of biomedical and health informatics, vol. 21, no. 1, pp. 76-84, 2016.
[19] S. Han, H. Mao, and W. Dally, "Deep Compression: Compressing Deep Neural Network with Pruning, Trained Quantization and Huffman Coding," arXiv: Computer Vision and Pattern Recognition, 2016.
[20] A A. G. Howard et al., "Mobilenets: Efficient convolutional neural networks for mobile vision applications," arXiv preprint arXiv:1704.04861, 2017.
[21] M. E. Stivanello, S. Vargas, M. L. Roloff, and M. R. Stemmer, "Automatic detection and classification of defects in knitted fabrics," IEEE Latin America Transactions, vol. 14, no. 7, pp. 3065-3073, 2016.
[22] D. Carrera, F. Manganini, G. Boracchi, and E. Lanzarone, "Defect detection in SEM images of nanofibrous materials," IEEE Transactions on Industrial Informatics, vol. 13, no. 2, pp. 551-561, 2016.
[23] R. Salakhutdinov, A. Mnih, and G. Hinton, "Restricted Boltzmann machines for collaborative filtering," presented at the Proceedings of the 24th international conference on Machine learning, Corvalis, Oregon, USA, 2007. [Online]. Available: https://doi.org/10.1145/1273496.1273596.
[24] G. E. Hinton and R. R. Salakhutdinov, "Reducing the dimensionality of data with neural networks," science, vol. 313, no. 5786, pp. 504-507, 2006.
[25] M. Lin, Q. Chen, and S. Yan, "Network In Network," CoRR, vol. abs/1312.4400, 2014.
[26] C.-Y. Lee, S. Xie, P. Gallagher, Z. Zhang, and Z. Tu, "Deeply-supervised nets," in Artificial intelligence and statistics, 2015: PMLR, pp. 562-570.
[27] S. Liu, L. Qi, H. Qin, J. Shi, and J. Jia, "Path Aggregation Network for Instance Segmentation," 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8759-8768, 2018.
[28] Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh, and J. Liang, "Unet++: A nested u-net architecture for medical image segmentation," in Deep learning in medical image analysis and multimodal learning for clinical decision support: Springer, 2018, pp. 3-11.
[29] H. Li, A. Kadav, I. Durdanovic, H. Samet, and H. P. Graf, "Pruning filters for efficient convnets," arXiv preprint arXiv:1608.08710, 2016.
[30] A. Renda, J. Frankle, and M. Carbin, "Comparing Rewinding and Fine-tuning in Neural Network Pruning," International Conference on Learning Representations, vol. abs/2003.02389, 2020.
[31] W. Chen, X. Zhong, and J. Zhang, "Optimization Research and Defect Object Detection of Aeroengine Blade Boss Based on YOLOv4," in Journal of Physics: Conference Series, 2021, vol. 1746, no. 1: IOP Publishing, p. 012076.
[32] H. Deng, J. Cheng, T. Liu, B. Cheng, and Z. Sun, "Research on iron surface crack detection algorithm based on improved YOLOv4 network," in Journal of physics: Conference series, 2020, vol. 1631, no. 1: IOP Publishing, p. 012081.

電子全文 電子全文(網際網路公開日期:20260908)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊