隨著極端天氣和氣候變化，淹水和高強度降雨開始頻繁發生。在下雨或淹水之後，裝有積水的容器就可能成為蚊子的孳生地。蚊子會傳播一種名為登革熱的嚴重傳染病。台灣政府希望能有一種更有效的方法來找出蚊子的孳生地。因此台灣政府提供了照片資料集並舉行比賽來徵求最佳的圖片檢測模型。在這篇研究中，我們建構一個圖像物件檢測模型來檢測可能會產生積水的容器。首先我們會檢查並過濾不適合的訓練數據。然後通過實驗來找到一個好的超參數來訓練我們的檢測模型。最後我們的檢測模型獲得了很高的準確性，並在該比賽中贏得了第一名。

With the extreme weather and climate change, floods and high-intensity rainfall will increase frequently. After raining or flooding, the containers that hold stagnant water may become mosquito breeding sites. Mosquitoes spread a serious epidemic disease, Dengue Fever. Taiwan government hopes to have a more efficient approach to find out the mosquito breeding sites. Therefore, the Taiwan government provides the image dataset and hold a competition to ask for the best detection model. In this work, we build an image object detection model to detect the containers that may hold stagnant water. First, we check and filter the inappropriate training data. Afterward, we find good hyperparameters to train our detection model with several experiments. Finally, our detection model achieved great accuracy and won first place in the competition.

I. Introduction 1
II. Related Work 4
III. Object Detection Model 8
A. Training Dataset 8
1) Data Annotations 8
2) Data Augmentation 9
B. Faster R-CNN Model 10
1) Faster R-CNN Architecture and Hyperparameters 10
2) Gradient Descent Optimizer 16
3) Transfer Learning 17
IV. Experiment Results 18
A. Evaluation metric 18
B. Experiment Result 19
V. Conclusion 25
References 27

[1] World Health Organization, "Dengue and severe dengue," from https://www.who.int/news-room/fact-sheets/detail/dengue-and-severe-dengue
[2] Wikipedia contributors. (2019, September 14). Dengue fever. In Wikipedia, The Free Encyclopedia. From https://en.wikipedia.org/w/index.php?title=Dengue_fever&oldid=915595019
[3] Taiwan Center for Disease Control, from https://www.cdc.gov.tw/Bulletin/Detail/IJYsuoOa_YZ7xOfBW7KddQ?typeid=9
[4] T. K. Bee, K. H. Lye and T. S. Yean, "Modeling Dengue Fever Subject to Temperature Change," Proceedings of 2009 Sixth International Conference on Fuzzy Systems and Knowledge Discovery, Tianjin, 2009, pp. 61-65.
[5] https://aidea-web.tw
[6] Z. Zheng, Y. Yang, X. Niu, H. Dai, and Y. Zhou, "Wide and Deep Convolutional Neural Networks for Electricity-Theft Detection to Secure Smart Grids," IEEE Transactions on Industrial Informatics, vol. 14, no. 4, pp. 1606-1615, April 2018.
[7] V. T. Bickel, C. Lanaras, A. Manconi, S. Loew, and U. Mall, "Automated Detection of Lunar Rockfalls Using a Convolutional Neural Network," in IEEE Transactions on Geoscience and Remote Sensing, vol. 57, no. 6, pp. 3501-3511, June 2019. doi: 10.1109/TGRS.2018.2885280
[8] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You Only Look Once: Unified, Real-Time Object Detection," 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, 2016, pp. 779-788.
[9] S. Ren, K. He, R. Girshick and J. Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137-1149, 1 June 2017.
[10] Huang J, Rathod V, Sun C, Zhu M, Korattikara A, Fathi A, Fischer I, Wojna Z, Song Y, Guadarrama S, Murphy K, "Speed/accuracy trade-offs for modern convolutional object detectors," CVPR 2017
[11] K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning for Image Recognition," 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, 2016, pp. 770-778.
[12] C. Szegedy et al., "Going deeper with convolutions," 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, 2015, pp. 1-9.
[13] Sergey Ioffe, Christian Szegedy, "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift," ICML'15 Proceedings of the 32nd International Conference on International Conference on Machine Learning - Volume 37, 2015, pp. 448-456.
[14] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens and Z. Wojna, "Rethinking the Inception Architecture for Computer Vision," 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, 2016, pp. 2818-2826.
[15] Christian Szegedy, Sergey Ioffe, Vincent Vanhoucke, Alex Alemi, "Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning," arXiv 1602.07261 (2016)
[16] Krizhevsky, Alex & Sutskever, Ilya & Hinton, Geoffrey, "ImageNet Classification with Deep Convolutional Neural Networks," Neural Information Processing Systems (2012). 25. 10.1145/3065386.
[17] R. Girshick, J. Donahue, T. Darrell and J. Malik, "Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation," 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, 2014, pp. 580-587.
[18] R. Girshick, "Fast R-CNN," 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, 2015, pp. 1440-1448.
[19] Liu, Wei et al. "SSD: Single Shot MultiBox Detector, " Lecture Notes in Computer Science (2016): 21–37. Crossref. Web.
[20] Joseph Redmon, Ali Farhadi, "YOLOv3: An Incremental Improvement", arXiv:1804.02767 (2018).
[21] S. J. Pan and Q. Yang, "A Survey on Transfer Learning," in IEEE Transactions on Knowledge and Data Engineering, vol. 22, no. 10, pp. 1345-1359, Oct. 2010.
[22] R. A. Aral, Ş. R. Keskin, M. Kaya and M. Hacıömeroğlu, "Classification of TrashNet Dataset Based on Deep Learning Models," 2018 IEEE International Conference on Big Data (Big Data), Seattle, WA, USA, 2018, pp. 2058-2062.
[23] C. Bircanoğlu, M. Atay, F. Beşer, Ö. Genç and M. A. Kızrak, "RecycleNet: Intelligent Waste Sorting Using Deep Neural Networks," 2018 Innovations in Intelligent Systems and Applications (INISTA), Thessaloniki, 2018, pp. 1-7.
[24] Vinod Nair, Geoffrey E. Hinton, "Rectified linear units improve restricted boltzmann machines," in ICML'10 Proceedings of the 27th International Conference on International Conference on Machine Learning, 2010, pp. 807-814
[25] Wikipedia contributors. (2019, July 9). Jaccard index. In Wikipedia, The Free Encyclopedia. From https://en.wikipedia.org/w/index.php?title=Jaccard_index&oldid=905462536
[26] Ilya Sutskever, James Martens, George Dahl, and Geoffrey Hinton, "On the importance of initialization and momentum in deep learning, " 2013 Proceedings of the 30th International Conference on International Conference on Machine Learning - Volume 28 (ICML'13), Sanjoy Dasgupta and David McAllester (Eds.), Vol. 28. JMLR.org III-1139-III-1147.
[27] https://storage.googleapis.com/openimages/web/download_v4.html
[28] Lin TY. et al. (2014), "Microsoft COCO: Common Objects in Context, " In Computer Vision – ECCV 2014. ECCV 2014. Lecture Notes in Computer Science, vol 8693. Springer, Cham