臺灣博碩士論文加值系統

摘要 I
ABSTRACT II
致謝 IV
目錄 V
圖目錄 IX
表目錄 XII
第一章、 緒論 1
1.1 研究動機 1
1.2 系統架構與流程 2
1.3 論文大綱 3
第二章、 相關習知技術與知識 4
2.1 縮減取樣 4
2.1.1 最鄰近插值法(Nearest Neighbor Interpolation) 4
2.1.2 雙線性插值法(Bilinear Interpolation) 5
2.1.3 雙三次插值法(Bicubic Interpolation) 6
2.2 卷積神經網路 6
2.2.1 卷積層(Convolutional Layer) 7
2.2.2 反卷積層(Deconvolutional Layer) 8
2.2.3 池化層(Pooling Layer) 9
2.2.4 全連接層(Fully Connected Layer) 9
2.2.5 激勵函數(Activation Function) 10
2.3 神經網路架構 11
2.3.1 2D U-Net 11
2.4 神經網路特徵傳遞方法 12
2.4.1 殘差網路(Residual Network) 13
2.4.2 密集連接網路(Densely Connected Networks) 14
2.5 通道注意力機制 16
2.5.1 SENet(Squeeze-and-Excitation Networks) 16
第三章、 相關方法探討 19
3.1 Abnormal Event Detection in Videos using Spatiotemporal Autoencoder 19
3.2 Future Frame Prediction for Anomaly Detection - A New Baseline 22
3.3 Video Anomaly Detection with Sparse Coding Inspired Deep Neural Networks 24
3.4 公開數據集 26
3.4.1 CUHK Avenue Dataset 27
3.4.2 ShanghaiTech Campus Dataset 28
3.4.3 UCSD Dataset 28
3.4.4 UCF Crime Dataset 29
第四章、 本系統與方法 31
4.1 影像前處理模組 33
4.1.1 輸入影像前處理 34
4.1.2 連續幀通道串接 35
4.2 異常偵測神經網路模組 36
4.2.1 未來幀預測網路 37
4.3.1.1 通道注意力層(Channel Attention) 40
4.3.1.2 密集塊(Dense Block) 40
4.2.2 計算損失函數 42
4.3 計分門檻系統模組 43
4.3.1 計分門檻 44
4.3.2 使用ROC曲線分析門檻值 48
4.4 網路訓練 51
4.5 數據集拍攝 52
4.5.1 公園 54
4.5.2 校園 55
4.5.3 住家門口 56
4.5.4 電梯門口 57
第五章、 實驗結果 58
5.1 實驗設備與環境 58
5.2 實驗結果與分析 59
5.2.1 主觀結果評估與比較 59
5.2.1.1 CUHK Avenue 實際測試畫面 60
5.2.1.2 ShanghaiTech 實際測試畫面 65
5.2.1.3 Our Dataset 實際測試畫面 72
5.2.2 客觀結果評估與比較 77
5.2.3 應用方面比較 79
5.2.4 整體性比較 81
5.3 錯誤討論與解決途徑 81
5.3.1 自評 81
5.3.2 遭遇之困難與解決途徑 82
5.3.2.1 檢測異常事件 82
5.3.2.2 神經網路模型的選擇與設計 83
5.3.2.3 串流影像中的畫面傳輸問題 83
5.3.2.4 驗證模型的可行性 84
第六章、 結論與未來方向 85
6.1 結論 85
6.2 未來方向 86
參考文獻 87

[1] Guansong Pang; Chunhua Shen; Longbing Cao; Anton Van Den Hengel;, "Deep Learning for Anomaly Detection: A Review," ACM Computing Surveys, vol. 54, no. 38, pp. 1-38, 5 12 2020.
[2] Olaf Ronneberger; Philipp Fischer; Thomas Brox;, "U-Net: Convolutional Networks for Biomedical Image Segmentation," in International Conference on Medical Image Computing and Computer-Assisted Intervention, 2015.
[3] R. Keys, "Cubic convolution interpolation for digital image processing," IEEE Transactions on Acoustics, Speech, and Signal Processing, vol. 29, no. 6, pp. 1153-1160, 12 1981.
[4] K. Fukushima, "Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position," Biological Cybernetics, vol. 36, p. pages193–202, 4 1980.
[5] Lecun, Y. and Bottou, L. and Bengio, Y. and Haffner, P., "Gradient-based learning applied to document recognition," Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, 1998.
[6] Zeiler, Matthew D. and Krishnan, Dilip and Taylor, Graham W. and Fergus, Rob, "Deconvolutional networks," in 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010.
[7] Zeiler, Matthew D. and Taylor, Graham W. and Fergus, Rob, "Adaptive deconvolutional networks for mid and high level feature learning," in 2011 International Conference on Computer Vision, 2011.
[8] Vinod Nair; Geoffrey E. Hinton;, "Rectified Linear Units Improve Restricted Boltzmann Machines," in International Conference on Machine Learning, 2010.
[9] Kaiming He; Xiangyu Zhang; Shaoqing Ren; Jian Sun;, "Deep Residual Learning for Image Recognition," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016.
[10] Gao Huang; Zhuang Liu; Laurens van der Maaten; Kilian Q. Weinberger;, "Densely Connected Convolutional Networks," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017.
[11] Jie Hu; Li Shen; Samuel Albanie; Gang Sun; Enhua Wu;, "Squeeze-and-Excitation Networks," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018.
[12] Yong Shean Chong; Yong Haur Tay;, "Abnormal Event Detection in Videos using Spatiotemporal Autoencoder," in International symposium on neural networks, 2017.
[13] Cewu Lu; Jianping Shi; Jiaya Jia;, "Abnormal Event Detection at 150 FPS in MATLAB," in Proceedings of the IEEE international conference on computer vision, 2013.
[14] Amit Adam; Ehud Rivlin; Ilan Shimshoni; Daviv Reinitz;, "Robust Real-Time Unusual Event Detection using Multiple Fixed-Location Monitors," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 30, no. 3, pp. 555-560, 21 1 2008.
[15] Vijay Mahadevan; Weixin Li; Viral Bhalodia; Nuno Vasconcelos;, "Anomaly detection in crowded scenes," in IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Francisco, CA, USA, 2010.
[16] Cong, Yang and Yuan, Junsong and Liu, Ji, "Sparse reconstruction cost for abnormal event detection," in CVPR 2011, 2011.
[17] Zhao, Bin and Fei-Fei, Li and Xing, Eric P., "Online detection of unusual events in videos via dynamic sparse coding," in CVPR 2011, 2011.
[18] Li, Ce and Han, Zhenjun and Ye, Qixiang and Jiao, Jianbin, "Abnormal Behavior Detection via Sparse Reconstruction Analysis of Trajectory," in 2011 Sixth International Conference on Image and Graphics, 2011.
[19] Mahmudul Hasan; Jonghyun Choi; Jan Neumann; Amit K. Roy-Chowdhury; Larry S. Davis;, "Learning Temporal Regularity in Video Sequences," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016.
[20] Chandola, Varun and Banerjee, Arindam and Kumar, Vipin, "Anomaly Detection A Survey," Association for Computing Machinery, vol. 41, no. 3, 7 2009.
[21] Medel, J. R. and Savakis, A., "Anomaly Detection in Video Using Predictive Convolutional Long Short-Term Memory Networks," arXiv e-prints, 12 2016.
[22] Wen Liu; Weixin Luo; Dongze Lian; Shenghua Gao;, "Future Frame Prediction for Anomaly Detection -- A New Baseline," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018.
[23] Philipp Fischer; Alexey Dosovitskiy; Eddy Ilg; Philip Häusser; Caner Hazırbaş; Vladimir Golkov; Patrick van der Smagt; Daniel Cremers; Thomas Brox;, "FlowNet: Learning Optical Flow with Convolutional Networks," in Proceedings of the IEEE international conference on computer vision, 2015.
[24] Luo, Weixin and Liu, Wen and Gao, Shenghua, "Remembering history with convolutional LSTM for anomaly detection," in 2017 IEEE International Conference on Multimedia and Expo (ICME), 2017.
[25] Weixin Luo; Wen Liu; Dongze Lian; Jinhui Tang; Lixin Duan; Xi Peng; Shenghua Gao;, "Video Anomaly Detection with Sparse Coding Inspired Deep Neural Networks," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 3, pp. 1070-1084, 2021.
[26] Wisdom, Scott and Powers, Thomas and Pitton, James and Atlas, Les, "Building recurrent networks by unfolding iterative thresholding for sequential sparse recovery," in 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2017.
[27] Bin Zhao; Li Fei-Fei; Eric P. Xing;, "Online detection of unusual events in videos via dynamic sparse coding," in CVPR, 2011.
[28] Waqas Sultani; Chen Chen; Mubarak Shah;, "Real-world Anomaly Detection in Surveillance Videos," in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018.
[29] Andrew P. Bradley, "The use of the area under the ROC curve in the evaluation of machine learning algorithms," Pattern Recognition, vol. 30, no. 7, pp. 1145-1159, 7 1997.
[30] J.A. Hanley, B.J. McNeil, "The meaning and use of the area under a receiver operating characteristic (ROC) curve," Radiology, vol. 143, no. 1, pp. 29-36, 1982.