臺灣博碩士論文加值系統

車牌偵測是車牌辨識中的關鍵步驟，本論文探討以最新深度學習方法解決開放 環境下的車牌偵測問題。本研究為極少數運用最新的深度學習方法應用於車牌偵測， 包含在各種挑戰的情境下，如惡劣天氣環境、照明與交通等。本研究探討目前最新 的即時物件檢測架構，包括 SSD (Single Shot MultiBox Detector) [1]和 YOLO (You- Only-Look-Once) [2]，並針對在開放環境下之車牌偵測做了一些改善。主要修改包括 以下內容:1)在 YOLO 上調整全連接層;2)在 SSD 上調整多尺度特徵圖，錨點與 寬高比;3)建置更完整和嚴格的 AOLPE (Application-Oriented License Plate Extended) 影像資料。AOLPE 影像資料庫是 AOLP (Application-Oriented License Plate) [3]資料庫 的延伸版，加入了許多在挑戰性場景下拍攝的影像樣本。原始的 YOLO 和 SSD 不是 針對車牌進行偵測，所以本論文提出 YOLO 和 SSD 的車牌偵測版本，在 AOLPE 影 像資料庫上進行評估。本論文的主要貢獻是將車牌偵測從實驗室向應用推進了一大 步，不僅實現了實時的車牌偵測方法，而且還釋放了更接近真實場景的 AOLPE 影像資料庫，並取得了良好的測試效果。

License Plate Detection (LPD) is the pivotal step for License Plate Recognition. In this work, we explore and customize state-of-the-art detection approaches for exclusively handling the LPD in the wild. In-the-wild LPD considers license plates captured in challenging conditions caused by bad weathers, lighting, traffics, and other factors. As conventional methods failed to handle these inevitable conditions, we explore the latest real- time deep learning based detectors, namely YOLO (You- Only-Look-Once) and SSD (Single Shot Multi-Box Detector), and customize them for effectively handling the LPD. The prime modifications include the following: 1) Modifying of fully connected layer on YOLO, 2) Tuning of multi-scale feature maps, anchor boxes, and aspect ratios on SSD, and 3) Creation of a more complete and rigorous AOLPE (Application-Oriented License Plate Extended) database for robust performance evaluation. The AOLPE database is an extended version of the AOLP (Application-Oriented License Plate) database with additional images taken under extreme but realistic conditions. As the original YOLO and SSD are not intended for LPD and they failed miserably as LPDs, the performances of the proposed customized versions of both YOLO and SSD are directly evaluated on the AOLPE database. The contributions made in this study is not only a pioneering customized exploration of state-of- the-art real- time deep learning approaches for handling in-the-wild LPD, but also involves the release of the AOLPE database and evaluation protocol to define a novel and practical benchmark for LPD.

中文摘要
Abstract
致謝
目錄
圖目錄
表目錄
第一章 簡介
1.1 研究背景和動機
1.2 方法概述
1.3 論文貢獻
1.4 論文架構
第二章 資料庫介紹與相關文獻探討
2.1 車牌影像資料庫
2.2 車牌偵測相關文獻
2.3 物件辨識相關文獻
第三章 主要方法與流程
3.1 開放環境下之車牌偵測挑戰影像 AOLPE
3.2 以 YOLO 為基礎之即時車牌偵測系統設計與分析
3.3 以 SSD 為基礎之即時車牌偵測系統設計與分析
3.4 系統比較
第四章 實驗設置與結果
4.1 車牌訓練影像
4.2 參數設置
4.3 車牌偵測實驗
4.4 車牌偵測實驗結果分析
第五章 結論與未來發展方向
參考文獻
附錄 A-字彙表
附錄 B-YOLOv2

[1] W. Liu, D. Anguelov, D. Erhan, C. Szegedy, and S. Reed, "SSD: Single Shot MultiBox Detector," arXiv preprint arXiv:1512.02325, 2015.
[2] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You only look once: Unified, real- time object detection," arXiv preprint arXiv:1506.02640, 2015.
[3] G.-S. Hsu, J.-C. Chen, and Y.-Z. Chung, "Application-oriented license plate recognition," IEEE transactions on vehicular technology, vol. 62, pp. 552-561, 2013.
[4] C. P. Tips and C. Spotlights, "Deep Learning for Computer Vision with Caffe and cuDNN."
[5] M. Everingham, L. Van Gool, C. Williams, J. Winn, and A. Zisserman, "The pascal visual object classes challenge 2007 (voc 2007) results (2007)," ed, 2008.
[6] A. Krizhevsky, I. Sutskever, and G. E. Hinton, "Imagenet classification with deep convolutional neural networks," in Advances in neural information processing systems, 2012, pp. 1097-1105.
[7] K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," arXiv preprint arXiv:1409.1556, 2014.
[8] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, et al., "Going deeper with convolutions," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 1-9.
[9] R. Girshick, J. Donahue, T. Darrell, and J. Malik, "Rich feature hierarchies for accurate object detection and semantic segmentation," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2014, pp. 580-587.
[10] R. Girshick, "Fast r-cnn," in Proceedings of the IEEE International Conference on Computer Vision, 2015, pp. 1440-1448.
[11] 曾喜得, "動態與靜態影像之車牌辨識." 台灣科技大學碩士學位論文, 2015.
[12] 邱俊瑋, "機車車牌即時追蹤演算法之效能探討." 台灣科技大學碩士學位論文, 2016.
[13] H. Nam and B. Han, "Learning multi-domain convolutional neural networks for visual tracking," arXiv preprint arXiv:1510.07945, 2015.
[14] M. Danelljan, G. Häger, F. Khan, and M. Felsberg, "Accurate scale estimation for robust visual tracking," in British Machine Vision Conference, Nottingham, September 1-5, 2014, 2014.
[15] J. F. Henriques, R. Caseiro, P. Martins, and J. Batista, "High-speed tracking with kernelized correlation filters," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 37, pp. 583-596, 2015.
[16] K. Zhang, L. Zhang, Q. Liu, D. Zhang, and M.-H. Yang, "Fast visual tracking via dense spatio-temporal context learning," in European Conference on Computer Vision, 2014, pp. 127-141.
[17] H. Li and C. Shen, "Reading Car License Plates Using Deep Convolutional Neural Networks and LSTMs," arXiv preprint arXiv:1601.05610, 2016.
[18] L. Zheng, X. He, B. Samali, and L. T. Yang, "An algorithm for accuracy enhancement of license plate recognition," Journal of computer and system sciences, vol. 79, pp. 245-255, 2013.
[19] S. Ren, K. He, R. Girshick, and J. Sun, "Faster R-CNN: Towards real-time object detection with region proposal networks," in Advances in neural information processing systems, 2015, pp. 91-99.
[20] A. Geiger, P. Lenz, C. Stiller, and R. Urtasun, "Vision meets robotics: The KITTI dataset," The International Journal of Robotics Research, p. 0278364913491297, 2013.
[21] O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, et al., "Imagenet large scale visual recognition challenge," International Journal of Computer Vision, vol. 115, pp. 211-252, 2015.
[22] B. Tian, Y. Li, B. Li, and D. Wen, "Rear-view vehicle detection and tracking by combining multiple parts for complex urban surveillance," IEEE Transactions on Intelligent Transportation Systems, vol. 15, pp. 597-606, 2014.
[23] M. Asif, C. Qi, and M. M. S. Fareed, "Multiple License Plate Detection for Chinese Vehicles in Dense Traffic Scenarios," IET Intelligent Transport Systems, 2016.
[24] A. Graves, S. Fernández, F. Gomez, and J. Schmidhuber, "Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks," in Proceedings of the 23rd international conference on Machine learning, 2006, pp. 369-376.
[25] M. Jaderberg, A. Vedaldi, and A. Zisserman, "Deep features for text spotting," in European conference on computer vision, 2014, pp. 512-528.
[26] K. Wang, B. Babenko, and S. Belongie, "End-to-end scene text recognition," in 2011 International Conference on Computer Vision, 2011, pp. 1457-1464.
[27] S. M. Lucas, A. Panaretos, L. Sosa, A. Tang, S. Wong, and R. Young, "ICDAR 2003 Robust Reading Competitions," in ICDAR, 2003, p. 682.
[28] S. M. Lucas, "ICDAR 2005 text locating competition results," in Eighth International Conference on Document Analysis and Recognition (ICDAR'05), 2005, pp. 80-84.
[29] A. Shahab, F. Shafait, and A. Dengel, "ICDAR 2011 robust reading competition challenge 2: Reading text in scene images," in 2011 international conference on document analysis and recognition, 2011, pp. 1491-1496.
[30] A. Criminisi, "Microsoft research cambridge object recognition image database," ed, 2004.
[31] A. Vedaldi and K. Lenc, "Matconvnet: Convolutional neural networks for matlab," in Proceedings of the 23rd ACM international conference on Multimedia, 2015, pp. 689-692.
[32] J. R. Uijlings, K. E. Van De Sande, T. Gevers, and A. W. Smeulders, "Selective search for object recognition," International journal of computer vision, vol. 104, pp. 154-171, 2013.
[33] M. D. Zeiler and R. Fergus, "Visualizing and understanding convolutional networks," in European Conference on Computer Vision, 2014, pp. 818-833.
[34] M. A. Sadeghi and D. Forsyth, "30hz object detection with dpm v5," in European Conference on Computer Vision, 2014, pp. 65-79.
[35] M. Lin, Q. Chen, and S. Yan, "Network in network," arXiv preprint arXiv:1312.4400, 2013.
[36] K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," arXiv preprint arXiv:1512.03385, 2015.
[37] B. Hariharan, P. Arbeláez, R. Girshick, and J. Malik, "Hypercolumns for object segmentation and fine-grained localization," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 447-456.
[38] L.-C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, and A. L. Yuille, "Semantic image segmentation with deep convolutional nets and fully connected crfs," arXiv preprint arXiv:1412.7062, 2014.
[39] J. Redmon and A. Farhadi, "YOLO9000: Better, Faster, Stronger," arXiv preprint arXiv:1612.08242, 2016.