在視頻監控中船隻偵測是一個重大議題，透過電腦視覺分析船隻的位置。傳統的偵測方法是透過監督式學習來判斷，但真實世界的情況較為複雜，無法每一種角度或大小的船隻都可以偵測出來。有鑑於以上方法的缺點，我們設計一套船隻偵測的方法，可以利用空拍機影像，在不同的角度（含仰角、俯角、方位角）、不同顏色、不同形狀下都可以順利被偵測出來。
本論文提出一個可應用於在各種海域中,即時自動偵測出船隻的方法。首先我們會訓練一個可以偵測所有角度，所有顏色的船隻模型，透過卷積式類神經網路的機器學習方法，得到一個可以偵測多角度的船隻模型。在船隻偵測的程序中，透過灰階化，高斯矩陣的應用，我們做到二值化的效果；並將前景切成一個個物件，以框架的方式將我們感興趣的區域輸入至已分類完成的模型裡，最後輸出我們的偵測影片。
實驗的部份我們先針對模型的準確度進行分析，得到最終的準確度為98%左右；此外針對不同環境進行分析，如有大量白煙的海域、光照很強的海域、船隻會旋轉的海域、很多顏色的船的場域、還有陸地的場域。我們提出的方法可以正確偵測到船隻，且可以區分陸地與船隻，在白煙海域的準確性是96.6％，在光照強的海域是96％，在船隻旋轉的場域是89.8％，在很多顏色的船的場域是96.1％，在陸地的場域為91.1%，且整體執行時間很短，一張影像的處理時間約0.05到0.06秒，達到即時偵測。

Ship detection is a major issue in video surveillance, analyzing the location of the ship through computer vision. The traditional detection method is judged by supervised learning, but the real world situation is more complicated and cannot be detected by ships of every angle or size. In view of the shortcomings of the above methods, we have designed a method of ship detection that can be used to detect images at different angles (including elevation, depression, and azimuth), different colors, and different shapes.
This paper proposes a method that can be applied to automatically detect ships in various sea areas. First we will train a ship model that can detect all angles and all colors. Through a machine learning method, convolutional neural network, a ship model capable of detecting multiple angles is obtained. In the ship detection process, we use the gray-scale, Gaussian matrix application, we achieve the effect of binarization; cut the foreground into individual objects, and input the regions of interest to classify the ships by the pre-trained model. Then we output our detection video at last.
In the experimental part, we first analyze the accuracy of the model, and the final accuracy is about 98%. In addition, we analyze the different environments, such as the sea with a lot of white smoke, the sea with strong illumination, and the sea where the ship will rotate, the field of many colors of the ship, and the field of the land. The method proposed in this thesis can correctly detect the ship and distinguish between land and ships. The accuracy in the white smoke area is 96.6%, 96% in the sea with strong illumination, 89.8% in the field where the ship rotates, the color of the ship's field is 96.1%, and the land field is 91.1%. The overall execution time is very fast, and the processing time of an aerial image is about 0.05 to 0.06 seconds to achieve real-time multi-angle ship detection.

[1]　Gaoya Wang, Huiyuan Fu, and Yingxin Liu, “Real Time Abnormal Crowd　Behavior Detection Based On Adjacent Flow Location Estimation,” in Proceedings of the Conference of Computational Interdisciplinary Science, Brazil , pp. 476-479, 2016.
[2] C. Corbane, E. Pecoul, L. Demagistri, and M. Petit, “Fully Automatic Procedure for Ship Detection Using Optical Satellite Imagery,” in Proceeding of SPIE 7150, Remote Sensing of Inland, Coastal, and Oceanic Waters, 71500R, Poitiers, France, Dec. 2008.
[3]　C. Zhu, H. Zhou, R. Wang, and J. Guo, “A Novel Hierarchical Method of Ship Detection from Spaceborne Optical Images Based on Shape and Texture Features,” IEEE Transactions on Geosciences and Remote Sensing, Pasadena, vol. 48, no. 9, pp. 3446-3456, 2010.
[4]　K. Eldhuset, “An Automatic Ship and Ship Wake Detection System for Spaceborne SAR images in Coastal Regions,” IEEE Transactions on Geosciences and Remote Sensing, Pasadena, vol. 34 no. 4, pp. 1010-1019, Jul. 1996.
[5]　M. Tello, C. Lopez-Martinez, and J. J. Mallorqui, “A Novel Algorithm for Ship Detection in SAR Imagery Based on the Wavelet Transform,” IEEE Transaction Geosciences and Remote Sensing, Pasadena, vol. 2, no. 2, pp. 201-205, Apr. 2005.
[6]　Juan Ram´on Ant´on Morillas, Irene Camino Garc´ıa and Udo Z¨olzer, “Ship Detection Based on SVM Using Color and Texture Features,” in Proceedings of the Intelligent Computer Communication and Processing, Cluj-Napoca, Romania, pp. 343-350, 2015.
[7]　O. Rusch, C. Ruwwe, and U. Z¨olzer, “Image Segmentation in Naval Ship Images,” in Proceedings of the 11th Workshop Farbbildverarbeitung der GFaI, Berlin, Germany, pp. 63-70, 2005.
[8]　 X. Bao, S. Zinger, R. Wijnhoven, and P. de With, “Water Region Detection Supporting Ship Identification in Port Surveillance,” Advanced Concepts for Intelligent Vision Systems, Sydney, Australia, vol. 7517, pp. 444-454, 2012.
[9]　 I. Camino Garc´ıa and U. Z¨olzer, “A Fully Automatic Ship Localization Algorithm for Complex Backgrounds,” in Proceedings of the International Conference on Image Processing, Computer Vision and Pattern Recognition, Las Vegas, Jul. 2013.
[10]　A. Rankin and L. Matthies, “Daytime Water Detection Based on Color Variation,” in Proceedings of the IEEE International Conference on Intelligent Robots and Systems, Taipei, Taiwan, pp. 215-221, 2010.
[11] Yang Cong, Junsong Yuan, and Ji Liu, “Sparse Reconstruction Cost for Abnormal Event Detection,” in Proceedings of the 24th IEEE Conference on Computer Vision and Pattern Recognition, Colorado, pp. 3449-3456, 2011.
[12] Weixin Lin, Vijay Mahadevan, and Nuno Vasconcelos, “Anomaly Detection and Localization in Crowded Scenes,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 36, no. 1, pp. 18-32, 2014.
[13] Ship - Google Search. (n.d.). Retrieved April 5th, 2018, from Google: https://www.google.com.tw/search?q=ship&source=lnms&tbm=isch&sa=X&ved=0ahUKEwiCj4f87JPcAhUUZt4KHYEyAZgQ_AUICigB&biw=1422&bih=653

[14] Ship Index – ShipSpotting.com – Ship Photos and Ship Trackers (n.d.). Retrieved April 5th, 2018, from ShipSpotting: http://www.shipspotting.com/ships/index.php
[15] Dan Xu, Elisa Ricci, Yan Yan, Jingkuan Song, and Nicu Sebe, “Learning Deep Representations of Appreance and Motion for Anomalous Event Detection,” in Proceedings of the British Machine Vision Conference, Swansea, UK, pp. 8.1-8.12, 2015.
[16] Y. Feng, Y. Yuan, and X. Lu, “Deep Representation for Abnormal Event Detection in Crowded Scenes,” in Proceedings of the 2016 ACM Conference on Multimedia Conference, Amsterdam, Netherlands, pp. 591-595, 2016.
[17] L. Itti and P. Baldi, “A Principled Approach to Detecting Surprising Events in Video,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, San Diego, pp. 631-637, 2005.
[18] X. Cui, Q. Liu, and M. Gao, “Abnormal Detection Using Interaction Energy Potential,” in Proceedings of the 24th IEEE Conference on Computer Vision and Pattern Recognition, Colorado, pp. 3161-3167, 2011.
[19] Hawook Jeong, Hyung Jin Chang, and Jin Young Choi, “Modeling of Moving　Object Trajectory by Spatio-temporal Learning for Abnormal Behavior Detection,” in Proceedings of the IEEE International Conference on Advanced Video and Signal-Based Surveillance, Klagenfurt, Austria, pp. 119-123, 2011.
[20] Tian Wang, and Hichem Snoussi, “Detection of Abnormal Visual Event via Global Optical Flow Orientation Histogram,” IEEE Transactions on Information Forensics and Security, vol. 9, no. 6, pp. 988-998, 2014.
[21] X. Bao, S. Zinger, de With, P.H.N., Wijnhoven, R.G.J., and Han, J. “Water Region and Multiple Ship Detection for Port Surveillance,” in Proceedings of the 33rd WIC Symposium on Information Theory, Boekeloo, Netherlands, pp. 20-27, 2012.
[22] V.D. Aleksey, A. K. Pavel, and N. B. Nikolay, “Ship Detection in Images Obtained from the Unmanned Aerial Vehicle (UAV)”, Indian Journal of Science and Technology, vol. 9, no. 46, December, 2016.
[23] Zechuang Chen, Bin Li, Lian Fang Tian, “Automatic Detection and Tracking of Ship Based on Mean Shift in Corrected Video Sequences,” in Proceedings of 2017 2nd International Conference on Image, Vision and Computing, Guangzhou, China, pp. 449-453, 2017.