臺灣四面環海，沿岸常受颱風外圍環流與東北季風影響，異常漫堤海浪或瘋狗浪事件因突發性高且不易即時辨識，對海岸活動安全與防災管理構成威脅。然而，傳統浮標、潮位站與數值模式多偏向整體海況描述，較難直接反映近岸越波漫堤衝擊，現有研究亦較少針對海岸影像建立可應用於短時間預警之事件級辨識流程。因此，本研究以海岸 CCTV 影像為基礎，建立一套海浪漫堤事件之短時間預警方法，作為未來低成本海岸智慧監測與防災系統之參考。
本研究以安平港堤岸 CCTV 影像為主要資料來源，先進行灰階化、事件判定與降採樣等前處理，再於畫面兩側分別設定目標區域與特徵提取帶，建構影像時間序列特徵，並結合波高、週期、水位與流況等靜態海象參數，建立長短期記憶網路（LSTM）預測模型；評估方式除傳統幀級指標外，另採事件級指標評估，以更貼近實務警報對漏報與誤報的需求。
結果顯示，本研究所建立之影像式流程具有預測越波事件之潛力；在固定門檻值下，基本模型於 A 區之事件級召回率達 0.76，B 區達 0.90。比較結果指出，靜態海象參數可提升事件段辨識能力，但其效果並非隨參數數量增加而單調改善；當目標區域與特徵提取帶距離增加時，整體辨識能力明顯衰減；提前預警約 3 秒仍具可行性，若延長至 4 秒以上則誤報增加且穩定性下降；提高 LSTM 層數與單元數對預測表現之改善有限。整體而言，本研究驗證了結合海岸影像、時序特徵與 LSTM 模型進行漫堤短時間預警之可行性，在特定條件下事件級召回率可達 0.95 以上，並提出分區配置、精簡模型與分級警報之部署建議。

Taiwan seashore is frequently affected and suffered from typhoons and northeastern monsson. Under these conditions, anomalous waves or rogue-wave-related events may cause sudden coastal overtopping and threaten public safety. This study developed a short-term warning approach for overtopping events using images captured from coastal CCTV and machine learning. CCTV videos collected at Anping Harbor were used as the main data source. Temporal image features were extracted from target regions and feature extraction strips, and static oceanographic parameters were added to selected model settings. The LSTM network was used as the main prediction model. Event-level evaluation was adopted to better reflect practical warning needs. The results show that the proposed framework can identify overtopping-related signals from coastal CCTV image and has potential for short-term warning. Model performance was affected by static oceanographic parameters, the spatial relationship between the target region and the feature extraction strip, and the selected lead time. This study shows that the proposed method can be used as a low-cost auxiliary warning tool for coastal overtopping events and can support future coastal monitoring and hazard mitigation systems.

1. Akbarifard, S., & Radmanesh, F. (2018). Predicting sea wave height using Symbiotic Organisms Search (SOS) algorithm. Ocean Engineering, 167, 348-356.
2. Akhmediev, N., Kibler, B., Baronio, F., Belić, M., Zhong, W.-P., Zhang, Y., Chang, W., Soto-Crespo, J. M., Vouzas, P., Grelu, P., Lecaplain, C., Hammani, K., Rica, S., Picozzi, A., Tlidi, M., Panajotov, K., Mussot, A., Bendahmane, A., Szriftgiser, P., Genty, G., Dudley, J., Kudlinski, A., Demircan, A., Morgner, U., Amiraranashvili, S., Bree, C., Steinmeyer, G., Masoller, C., Broderick, N. G. R., Runge, A. F. J., Erkintalo, M., Residori, S., Bortolozzo, U., Arecchi, F. T., Wabnitz, S., Tiofack, C. G., Coulibaly, S., & Taki, M. (2016). Roadmap on optical rogue waves and extreme events. Journal of Optics, 18(6), 063001.
3. Azevedo, L., Meyers, S., Pleskachevsky, A., Pereira, H. P., & Luther, M. (2022). Characterizing rogue waves at the entrance of Tampa Bay (Florida, USA). Journal of Marine Science and Engineering, 10(4), 507.
4. Bernier, N. B., & Thompson, K. R. (2015). Deterministic and ensemble storm surge prediction for Atlantic Canada with lead times of hours to ten days. Ocean Modelling, 86, 114-127.
5. Chang, H. K., Liou, J. C., Liu, S. J., & Liaw, S. R. (2011). Simulated wave-driven ANN model for typhoon waves. Advances in Engineering Software, 42(1-2), 25-34.
6. Chang, M., XING, Y. Y., ZHANG, Q. Y., HAN, S. J., & Kim, M. (2020). A CNN Image Classification Analysis for ‘Clean-Coast Detector’as Tourism Service Distribution. Journal of Distribution Science, 18(1), 15-26.
7. Chen, Y. W., Yu, T. T., & Peng, W. F. (2025). Early Surge Warning Using a Machine Learning System with Real-Time Surveillance Camera Images. Journal of Marine Science and Engineering, 13(2), 193.
8. Chien, H. W. A., Kao, C. C., & Chuang, L. Z. (2002). On the characteristics of observed coastal freak waves. Coastal engineering journal, 44(04), 301-319.
9. Dai, H. F., Bian, H. W., Wang, R. Y., & Ma, H. (2020). An INS/GNSS integrated navigation in GNSS denied environment using recurrent neural network. Defence technology, 16(2), 334-340.
10. De Laurentiis, L., Jones, C. E., Holt, B., Schiavon, G., & Del Frate, F. (2020). Deep learning for mineral and biogenic oil slick classification with airborne synthetic aperture radar data. IEEE Transactions on Geoscience and Remote Sensing, 59(10), 8455-8469.
11. de Silva, A., Mori, I., Dusek, G., Davis, J., & Pang, A. (2021). Automated rip current detection with region based convolutional neural networks. Coastal Engineering, 166, 103859.
12. Doong, D. J., Chen, S. T., Chen, Y. C., & Tsai, C. H. (2020). Operational probabilistic forecasting of coastal freak waves by using an artificial neural network. Journal of Marine Science and Engineering, 8(3), 165.
13. Doong, D. J., Peng, J. P., & Chen, Y. C. (2018). Development of a warning model for coastal freak wave occurrences using an artificial neural network. Ocean Engineering, 169, 270-280.
14. Ducournau, A., & Fablet, R. (2016, December). Deep learning for ocean remote sensing: An application of convolutional neural networks for super-resolution on satellite-derived SST data. In 2016 9th IAPR workshop on pattern recogniton in remote sensing (PRRS) (pp. 1-6). IEEE.
15. Durap, A. (2024). Data-driven models for significant wave height forecasting: Comparative analysis of machine learning techniques. Results in Engineering, 24, 103573.
16. Dusek, G., Hernandez, D., Willis, M., Brown, J. A., Long, J. W., Porter, D. E., & Vance, T. C. (2019). WebCAT: Piloting the development of a web camera coastal observing network for diverse applications. Frontiers in Marine Science, 6, 353.
17. Dysthe, K., Krogstad, H. E., & Müller, P. (2008). Oceanic rogue waves. Annu. Rev. Fluid Mech., 40(1), 287-310.
18. Elman, J. L. (1990). Finding structure in time. Cognitive science, 14(2), 179-211.
19. Fallati, L., Polidori, A., Salvatore, C., Saponari, L., Savini, A., & Galli, P. (2019). Anthropogenic Marine Debris assessment with Unmanned Aerial Vehicle imagery and deep learning: A case study along the beaches of the Republic of Maldives. Science of The Total Environment, 693, 133581.
20. Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780.
21. Hsu, W. Y., Huang, Z. C., Na, B., Chang, K. A., Chuang, W. L., & Yang, R. Y. (2019). Laboratory observation of turbulence and wave shear stresses under large scale breaking waves over a mild slope. Journal of Geophysical Research: Oceans, 124(11), 7486-7512.
22. Hu, H., van der Westhuysen, A. J., Chu, P., & Fujisaki-Manome, A. (2021). Predicting Lake Erie wave heights and periods using XGBoost and LSTM. Ocean Modelling, 164, 101832.
23. Hu, Z., Zhang, X., Li, Y., Li, X., & Qin, H. (2021). Numerical simulations of super rogue waves in a numerical wave tank. Ocean Engineering, 229, 108929.
24. Jaseena, K. U., & Kovoor, B. C. (2021). Decomposition-based hybrid wind speed forecasting model using deep bidirectional LSTM networks. Energy Conversion and Management, 234, 113944.
25. Kharif, C., & Pelinovsky, E. (2003). Physical mechanisms of the rogue wave phenomenon. European Journal of Mechanics-B/Fluids, 22(6), 603-634.
26. Lawyer, C., An, L., & Goharian, E. (2023). A review of climate adaptation impacts and strategies in coastal communities: From agent-based modeling towards a system of systems approach. Water, 15(14), 2635.
27. Lees, T., Reece, S., Kratzert, F., Klotz, D., Gauch, M., De Bruijn, J., ... & Dadson, S. (2021). Hydrological concept formation inside long short-term memory (LSTM) networks. Hydrology and Earth System Sciences Discussions, 2021, 1-37.
28. Lin, J., Li, H., Liu, N., Gao, J., & Li, Z. (2020). Automatic lithology identification by applying LSTM to logging data: A case study in X tight rock reservoirs. IEEE Geoscience and Remote Sensing Letters, 18(8), 1361-1365.
29. Liu, P., Li, Y., Liu, B., Chen, P., & Xu, J. (2019). Semi-automatic oil spill detection on X-band marine radar images using texture analysis, machine learning, and adaptive thresholding. Remote Sensing, 11(7), 756.
30. Luo, J. Y., Irisson, J. O., Graham, B., Guigand, C., Sarafraz, A., Mader, C., & Cowen, R. K. (2018). Automated plankton image analysis using convolutional neural networks. Limnology and Oceanography: methods, 16(12), 814-827.
31. Luo, Q. R., Xu, H., & Bai, L. H. (2022). Prediction of significant wave height in hurricane area of the Atlantic Ocean using the Bi-LSTM with attention model. Ocean Engineering, 266, 112747.
32. Mahjoobi, J., & Mosabbeb, E. A. (2009). Prediction of significant wave height using regressive support vector machines. Ocean Engineering, 36(5), 339-347.
33. Noble, W. S. (2006). What is a support vector machine?. Nature biotechnology, 24(12), 1565-1567.
34. Onorato, M., Residori, S., Bortolozzo, U., Montina, A., & Arecchi, F. (2013). Rogue waves and their generating mechanisms in different physical contexts. Physics Reports, 528(2), 47-89.
35. Panakkat, A., & Adeli, H. (2009). Recurrent neural network for approximate earthquake time and location prediction using multiple seismicity indicators. Computer‐Aided Civil and Infrastructure Engineering, 24(4), 280-292.
36. Panchang, V. G., Cushman-Roisin, B., & Pearce, B. R. (1988). Combined refraction-diffraction of short-waves in large coastal regions. Coastal Engineering, 12(2), 133-156.
37. Pelinovsky, E., Talipova, T., & Kharif, C. (2000). Nonlinear-dispersive mechanism of the freak wave formation in shallow water. Physica D: Nonlinear Phenomena, 147(1-2), 83-94.
38. Sareen, K., Panigrahi, B. K., Shikhola, T., & Sharma, R. (2023). An imputation and decomposition algorithms based integrated approach with bidirectional LSTM neural network for wind speed prediction. Energy, 278, 127799.
39. Shrivastava, A., & Srivastava, D. K. (2016, June). A review on pixel-based binarization of gray images. In Proceedings of the International Congress on Information and Communication Technology: ICICT 2015, Volume 2 (pp. 357-364). Singapore: Springer Singapore.
40. Slunyaev, A., Sergeeva, A., & Didenkulova, I. (2016). Rogue events in spatiotemporal numerical simulations of unidirectional waves in basins of different depth. Natural Hazards, 84(Suppl 2), 549-565.
41. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: a simple way to prevent neural networks from overfitting. The journal of machine learning research, 15(1), 1929-1958.
42. Tsai, C. H., Su, M. Y., & Huang, S. J. (2004). Observations and conditions for occurrence of dangerous coastal waves. Ocean engineering, 31(5-6), 745-760.
43. Wang, C. Y., Huang, T. C., & Wu, Y. M. (2022). Using LSTM neural networks for onsite earthquake early warning. Seismological Society of America, 93(2A), 814-826.
44. Wang, J. Q., Du, Y., & Wang, J. (2020). LSTM based long-term energy consumption prediction with periodicity. energy, 197, 117197.
45. Wei, L., Guan, L., Qu, L., & Guo, D. (2020). Prediction of sea surface temperature in the China seas based on long short-term memory neural networks. Remote Sensing, 12(17), 2697.
46. Yang, C. C., & Hwang, C. H. (2010, August). Studies on typhoon-wave prediction by Artificial Neural Networks at Longdong, Taiwan. In 2010 Sixth International Conference on Natural Computation (Vol. 2, pp. 626-630). IEEE.
47. Zakharov, V. E., Dyachenko, A. I., & Prokofiev, A. O. (2006). Freak waves as nonlinear stage of Stokes wave modulation instability. European Journal of Mechanics-B/Fluids, 25(5), 677-692.
48. Zhang, S., Wang, Y., Liu, M., & Bao, Z. (2017). Data-based line trip fault prediction in power systems using LSTM networks and SVM. Ieee Access, 6, 7675-7686.
49. Zhang, W., Guo, W., Liu, X., Liu, Y., Zhou, J., Li, B., Lu, Q. & Yang, S. (2018). LSTM-based analysis of industrial IoT equipment. IEEE Access, 6, 23551-23560.
50. 何良勝、林受勳、江玟德、張富東、張惠華（2007），臺灣地區港灣波浪推算與構造物安全監測研究(2/4)，交通部運輸研究所
51. 吳基、林柏青、林受勳、何良勝、徐如娟、江金德、廖慶堂、張富東、蔡金吉（2004），安平港海氣象觀測、防波堤水工模型試驗以及數值模擬研究(6)--海氣象觀測，交通部運輸研究所
52. 吳基、徐如娟、林受勳、何良勝（2008），臺灣地區國內商港附近海域海氣象觀測分析研究(3/4)，交通部運輸研究所
53. 吳基、徐如娟、林受勳、蘇青和（2007），臺灣地區國內商港附近海域海氣象 觀測分析研究(2/4)，交通部運輸研究所
54. 張芹瑜（2024），基於長短期記憶網路的海岸瘋狗浪預測模型研究，國立成功大學水利及海洋工程研究所碩士論文
55. 張紘聞（2022），應用WWIII波浪模式於極端波高模擬之研究，國立成功大學水利及海洋工程研究所碩士論文
56. 郭晉甫（2025），生成對抗式卷積遞迴神經網路預測近岸颱風風浪之研究，國立臺灣海洋大學海洋環境資訊系碩士學位論文
57. 陳冠宇（2002），瘋狗浪的可能機制. 海洋工程學刊, 2(1), 93-106.
58. 陳威成（2021），人工智慧（AI）演算法在瘋狗浪機率預警系統建置之研究，國立成功大學水利及海洋工程研究所碩士論文
59. 陳威成、陳盈智、林芳如、董東璟、蔡政翰（2022），隨機森林機器學習法應用於瘋狗浪機率預警系統之研究，第44屆海洋工程研討會論文集，P505-510
60. 陳威成、楊文榮、傅仲偉、董東璟、廖建明（2022），人工智慧技術在海洋觀測之應用。土木水利，49(6)，31-39。
61. 陳盈智、王敘民、董東璟、蔡政翰、滕春慈（2022），利用海岸攝影機影像遙測瘋狗浪之研究. 航測及遙測學刊, 27(2), 121-129.
62. 陳盈智、董東璟、林芳如、滕春慈（2022），海岸異常波浪觀測之研究. 土木水利, 49(6), 40-45.
63. 楊大成、范揚洺、李卞軍、高家俊、王仲豪（2011），應用可適隨機搜尋法率定SWAN波浪模式，第33屆海洋工程研討會論文集
64. 董東璟、蔡政翰、蕭士俊（2020），異常海象指標建立之研究，交通部中央氣象局委託研究計畫(期末)成果報告。
65. 廖御廷（2023），基於影像的網格分析應用於 CCTV 觀測河川水位. 淡江大學資訊工程學系碩士班學位論文.
66. 廖慶堂、林達遠、羅冠顯、劉明鑫、曹勝傑、柯拓宇、陳子健、陳孟宏、李俊穎（2024），2022 年港灣海氣象觀測資料統計年報 (8 港域觀測潮汐資料)，交通部運輸研究所
67. 廖慶堂、徐如娟、林受勳、何良勝（2012），臺灣國內商港海域海氣象觀測分析研究(3/4)，交通部運輸研究所
68. 劉俊志（2006），近岸突浪的分析探討，國立中山大學海洋物理研究所碩士論文
69. 戴世杰、黃清哲、范揚洺（2017），WAVEWATCH III 多重網格模式應用於颱風波浪推算之研究，第39屆海洋工程研討會論文集
70. 羅冠顯、林達遠、廖慶堂、劉明鑫、曹勝傑、柯拓宇、陳子健、陳孟宏、李俊穎（2024）， 2022 年港灣海氣象觀測資料統計年報 (8港域觀測波浪資料)，交通部運輸研究所
71. 國立成功大學近海水文中心 (2026.2.25)， https://www.comc.ncku.edu.tw/
72. 台灣大學海洋研究所:海氣象即時觀測浮標 (2026.3.1)， https://po.oc.ntu.edu.tw/buoy/
73. 聯合知識庫 (2020.9.7) ，https://udndata.com/ndapp/udntag/Index