簡易檢索 / 詳目顯示

回結果列表
研究生： 鄭瓊娥
Trinh, Quynh Nga
論文名稱： 人工智慧追蹤懸浮藻毯技術在自來水水質管理的應用研究
AI-Based Tracking of Floating Algal Mats for Drinking Water Quality Management
指導教授： 林財富
Lin, Tsair-Fuh
共同指導教授: 薛欣達
Hsueh, Hsin-Ta
學位類別： 碩士
Master
系所名稱： 工學院 - 環境工程學系
Department of Environmental Engineering
論文出版年： 2025
畢業學年度： 113
語文別： 英文
論文頁數： 112
中文關鍵詞： 藻毯 、物件追蹤 、水質監測 、軌跡估算 、YOLOv11
外文關鍵詞： YOLOv11, Algal Mats, Object Tracking, Water Quality Monitoring, Trajectory Estimation
相關次數： 點閱：18  下載：3 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 藍綠浮動藍綠菌藻毯在天然水體與水處理設施中，對操作與水質管理皆構成重大挑戰。本研究提出一套以人工智慧（AI）為基礎之藻毯偵測與追蹤框架，結合 YOLOv11（You Only Look Once 第十一版）物件偵測模型與 SORT 追蹤演算法，以即時監測浮動藻毯之大小、移動軌跡與空間分布。

    本研究探討兩種實際應用場景：(1) 流動水道（如石門渠道與鯤鯓湖公園），此類場域中藻毯隨水流方向漂移，可能導致臭味化合物（如 2-MIB）入侵；(2) 自來水廠之沉澱池（如路竹自來水處理廠），藻毯易於累積，需即時人工清除。

    研究採集真實場域之影像進行分析，並透過相機校正，將邊界框尺寸轉換為實際物理量測。系統能準確估算藻毯尺寸與移動軌跡，並透過持續追蹤物件識別編號（ID），實現速度與覆蓋率之量化分析。最終成果整合至 Dash 平台介面，供現場人員與操作人員即時決策使用。

    模型於交集覆蓋率（IoU）為 0.5 條件下之平均精準率（mAP@0.5）達 0.68，展現良好之偵測效能與穩定的追蹤編號一致性。系統能提供藻毯移速（cm/s）、軌跡視覺化與百分比覆蓋率等指標，協助水廠即時監控與早期預警。

    研究結果顯示，本研究可有效應用於動態（渠道）與靜態（沉澱池）環境。在流動水道中，可用以預警藻毯入侵並量化覆蓋率；於水處理廠內，則有助於視覺化偵測與即時干預。整體而言，本研究提出一套簡易、可擴展之水質監測工具，對於藍綠菌暴發管理具有實際應用價值。

    Floating cyanobacterial mats present significant operational and water quality challenges in both natural water bodies and treatment facilities. This study proposes an Artificial Intelligence Based (AI-based) detection and tracking framework using the You Only Look Once version 11(YOLOv11) object detection model combined with the SORT tracking algorithm to monitor the size, movement, and spatial distribution of floating algal mats in real-time. Two practical scenarios were explored: (1) flowing water channels (e.g., Shihmen and Hulupi Park), where mats drift directionally and may contribute to odor compound (2-MIB) intrusion; and (2) sedimentation tanks at a water treatment plant (e.g., Luzhu Water Treatment Plant), where mats accumulate and require timely manual removal.
    Field videos were collected under real-world conditions, and camera calibration was performed to convert bounding box dimensions into physical measurements. The system accurately estimated mat dimensions and movement trajectories, with identification (ID) continuity maintained across frames for velocity and coverage estimation. The model was visualized using a Dash-based interface to support decision-making.
    The system achieved good detection accuracy (mean Average Precision at Intersection over Union - mAP@ IoU 0.5: 0.68) and stable ID tracking, allowing for velocity estimation (in cm/s), trajectory visualization, and percent coverage analysis. These indicators were integrated into a Dash-based user interface to support real-time decision-making by plant operators and field personnel.
    Results demonstrate the framework's effectiveness in both dynamic (channel) and static (tank) environments. In flowing channels, it enabled early warning for algal intrusion and quantification of mat coverage. In treatment tanks, it supported visual detection and timely intervention. Overall, this research contributes a lightweight and adaptable tool for water quality monitoring, with practical implications for cyanobacterial bloom management in diverse aquatic settings.

    Abstract ii 摘要 iv Acknowledgement vi Table of Contents vii Table of Figure x List of Table xii Chapter 1 Introduction 1 1.1 Background and Importance of Algal Mats 1 1.2 Application Scenarios in Water Quality Monitoring 3 1.3 Challenges in Monitoring and Detection 6 Chapter 2 Literature review 8 2.1 Global Drivers for Algal-Bloom Monitoring 8 2.2 Impacts of Floating Algal Mats in Reservoirs and Water Treatment Systems 9 2.3 Conventional Monitoring Techniques 12 2.4 Optical and Imaging-Based Methods in Water-Treatment Settings 14 2.4.1 Fixed Cameras and Time-Lapse Systems 14 2.4.2 UAVs and Remote Sensing in Local Facilities 15 2.4.3 Conventional Image Processing Techniques 16 2.4.4 Limitations and Need for Deep Learning-Based Solutions 17 2.5 Deep Learning for Aquatic Algae Detection 17 2.6 Object Tracking Algorithms for Bloom Dynamics 20 2.6.1 Traditional Tracking Approaches: Kalman Filter and Optical Flow 20 2.6.2 DeepSORT and ByteTrack: Tracking with Appearance Embedding 22 2.6.3 SORT: A Lightweight, Effective Tracking Baseline 22 2.7 End-to-End AI Pipelines in Drinking Water Utilities 24 2.8 Identified Gaps and Rationale for this study 27 Chapter 3 Material and Methods 32 3.1 Overview 32 3.2 Study Area and Data Collection 34 3.2.1 Environmental Context and Site Characteristics 35 3.2.2 Data Acquisition Methods 39 3.3 Dataset Preparation and Annotation 40 3.3.1 Frame Extraction from Field Videos 41 3.3.2 Manual Annotation and Class Definition 41 3.3.3 Dataset Structure and Splitting 42 3.3.4 Dataset Summary 43 3.4 Detection Model Architecture (YOLOv11) 44 3.4.1 Rationale for Choosing YOLOv11 44 3.4.2 Input and Output Format 47 3.4.3 Training Configuration 47 3.4.4 Inference and Postprocessing 48 3.5 Tracking Algorithm: SORT 49 3.5.1 Tracking-by-Detection Framework 49 3.5.2 Algorithmic Components of SORT 49 3.5.3 Integration into the Monitoring Pipeline 50 3.6 Integrated Detection and Tracking Pipeline 50 3.7 Performance Evaluation Metrics 54 3.7.1 Detection Evaluation Metrics (YOLOv11) 55 3.8 Hardware Setup and Deployment Environment 56 3.8.1 Cloud-based Training Environment 56 3.8.2 Local CPU-Based Testing Environment 57 3.8.3 Video Input and Data Capture 58 3.8.4 Deployment Considerations and Scalability 59 3.8.5 Summary 59 Chapter 4 Results and Discussion 60 4.1 Overview the chapter 60 4.2 Model training results 61 4.2.1 Early Epoch Behavior 61 4.2.2 Training Stability and Loss Convergence 62 4.2.3 Justification for Epoch Limit 69 4.3 Real-World Application Results 69 4.4 Advantages of the developed method 77 4.5 Discussion of Limitations 80 4.6 Integration with Water Management – Potential for Development 83 4.7 Summary of Key Findings 87 Chapter 5 Conclusions and Suggestions 89 5.1 Conclusions 89 5.2 Suggestions 90 References 92

    Abdul Gaffar Sheik, A. K., Reeza Patnaik, Sheena Kumari & Faizal, & Bux. (2023). Machine learning-based design and monitoring of algae blooms: Recent trends and future perspectives – A short review. https://doi.org/10.1080/10643389.2023.2252313
    Abdullah, Ali, S., Khan, Z., Hussain, A., Athar, A., & Kim, H.-C. (2022). Computer Vision Based Deep Learning Approach for the Detection and Classification of Algae Species Using Microscopic Images. 14(14). https://doi.org/10.3390/w14142219
    Affonso, C., Rossi, A. L. D., Vieira, F. H. A., & de Leon Ferreira, A. C. P. (2017). Deep learning for biological image classification. https://doi.org/10.1016/j.eswa.2017.05.039
    Anderson, D. (2014). HABs in a changing world: a perspective on harmful algal blooms, their impacts, and research and management in a dynamic era of climactic and environmental change. Harmful algae 2012: proceedings of the 15th International Conference on Harmful Algae: October 29-November 2, 2012, CECO, Changwon, Gyeongnam, Korea/editors, Hak Gyoon Kim, Beatriz Reguera, Gustaaf M. Hallegraeff, Chang Kyu Lee, M.,
    Avidan, S., Brostow, G., Cissé, M., Farinella, G. M., & Hassner, T. (2022). Computer Vision–ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XX (Vol. 13680). Springer Nature.
    Cao, X., Guo, S., Lin, J., Zhang, W., & Liao, M. (2020). Online tracking of ants based on deep association metrics: method, dataset and evaluation. Pattern Recognition, 103. https://doi.org/10.1016/j.patcog.2020.107233
    Chang, C.-C. (2025). Analysis of 2-MIB sources in water supply channels National Cheng Kung University].
    Chapra, S. C., Boehlert, B., Fant, C., Bierman, V. J., Jr., Henderson, J., Mills, D., Mas, D. M. L., Rennels, L., Jantarasami, L., Martinich, J., Strzepek, K. M., & Paerl, H. W. (2017). Climate Change Impacts on Harmful Algal Blooms in U.S. Freshwaters: A Screening-Level Assessment. Environ Sci Technol, 51(16), 8933-8943. https://doi.org/10.1021/acs.est.7b01498
    Chorus, I., and Martin Welker. (2021). Toxic cyanobacteria in water: a guide to their public health consequences, monitoring and management., 858. https://doi.org/10.1201/9781003081449
    Chorus, I., Falconer, I. R., Salas, H. J., & Bartram, J. (2000). Health risks caused by freshwater cyanobacteria in recreational waters. J Toxicol Environ Health B Crit Rev, 3(4), 323-347. https://doi.org/10.1080/109374000436364
    Clark, J. M., Schaeffer, B. A., Darling, J. A., Urquhart, E. A., Johnston, J. M., Ignatius, A., Myer, M. H., Loftin, K. A., Werdell, P. J., & Stumpf, R. P. (2017). Satellite monitoring of cyanobacterial harmful algal bloom frequency in recreational waters and drinking source waters. Ecol Indic, 80, 84-95. https://doi.org/10.1016/j.ecolind.2017.04.046
    Corporation, T. W. (2025). TAIWAN WATER CORPORATION-Water Quality Report.
    Cullen, J. J., Ciotti, Á. M., Davis, R. F., & Lewis, M. R. (2003). Optical detection and assessment of algal blooms. Limnology and Oceanography, 42(5part2), 1223-1239. https://doi.org/10.4319/lo.1997.42.5_part_2.1223
    Du, Y., Zhao, Z., Song, Y., Zhao, Y., Su, F., Gong, T., & Meng, H. . (2023). StrongSORT: Make DeepSORT Great Again. 25. https://doi.org/10.1109/TMM.2023.3240881
    Fu, C., Liu, R., Fan, X., Chen, P., Fu, H., Yuan, W., ... & Luo, Z. . (2023). Rethinking general underwater object detection: Datasets, challenges, and solutions. https://doi.org/10.1016/j.neucom.2022.10.039
    Ghojogh, B., & Crowley, M. (2019). The Theory Behind Overfitting, Cross Validation, Regularization, Bagging, and Boosting: Tutorial. https://doi.org/arXiv:1905.12787
    Glorot, X., & Bengio, Y. (2010). Understanding the difficulty of training deep feedforward neural networks https://proceedings.mlr.press/v9/glorot10a.html
    Henderson, R., Parsons, S. A., & Jefferson, B. (2008). The impact of algal properties and pre-oxidation on solid-liquid separation of algae. Water Research, 42(8-9), 1827-1845. https://doi.org/10.1016/j.watres.2007.11.039
    Hidayatullah, P., Syakrani, N., Sholahuddin, M. R., Gelar, T., & Tubagus, R. (2025). YOLOv8 to YOLO11: A Comprehensive Architecture In-depth Comparative Review. arXiv preprint arXiv:2501.13400. https://doi.org/10.48550/arXiv.2501.13400
    Huisman, J., Codd, G. A., Paerl, H. W., Ibelings, B. W., Verspagen, J. M. H., & Visser, P. M. (2018). Cyanobacterial blooms. Nat Rev Microbiol, 16(8), 471-483. https://doi.org/10.1038/s41579-018-0040-1
    Jegham, N., Koh, C. Y., Abdelatti, M., & Hendawi, A. (2024). Yolo evolution: A comprehensive benchmark and architectural review of yolov12, yolo11, and their previous versions. Yolo11, and Their Previous Versions. https://doi.org/10.13140/RG.2.2.15952.83201
    Jenkins, W. M., & Masterton, R. B. (1982). Sound localization: effects of unilateral lesions in central auditory system. Journal of Neurophysiology, 47(6), 987-1016.
    Jocher, G., & Munawar, M. R. (2024). Ultralytics YOLO11. https://docs.ultralytics.com/models/yolo11/
    Jung, Y. C. a. C. (2019). Single Image Reflection Removal Using Convolutional Neural Networks. IEEE Transactions on Image Processing, 28, 1954-1966. https://doi.org/10.1109/TIP.2018.2880088
    Khanam, R., & Hussain, M. (2024). Yolov11: An overview of the key architectural enhancements. arXiv preprint arXiv:2410.17725. https://doi.org/10.48550/arXiv.2410.17725
    Knappe, D. R. (2004). Algae detection and removal strategies for drinking water treatment plants. American Water Works Association. https://books.google.com.tw/books?hl=en&lr=&id=UtuTFyk6pQ4C&oi=fnd&pg=PR13&dq=algae+in+water+treatment+plant&ots=F9VlfXPkii&sig=aPquBVLiYfY7zzZt-SFGf11KVyo&redir_esc=y#v=onepage&q=algae%20in%20water%20treatment%20plant&f=false
    Kudela, R., Berdalet, E., & Urban, E. (2015). Harmful algal blooms: a scientific summary for policy makers.
    Kuo, L.-C., & Tai, C.-C. (2022). Robust Image-Based Water-Level Estimation Using Single-Camera Monitoring. IEEE, 71. https://doi.org/10.1109/TIM.2022.3161691
    LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436-444. https://doi.org/10.1038/nature14539
    Li, J., Yang, X., & Sitzenfrei, R. (2020). Rethinking the Framework of Smart Water System: A Review. Water, 12(2). https://doi.org/10.3390/w12020412
    Liciotti, D., Paolanti, M., Frontoni, E., Zingaretti, P. . (2017). New Trends in Image Analysis and Processing – ICIAP 2017. https://doi.org/10.1007/978-3-319-70742-6_20
    Matthews, M. W. (2014). Eutrophication and cyanobacterial blooms in South African inland waters: 10years of MERIS observations. Remote Sensing of Environment, 155, 161-177. https://doi.org/10.1016/j.rse.2014.08.010
    Mikołajczyk, A., & Grochowski, M. (2018). Data augmentation for improving deep learning in image classification problem. https://doi.org/10.1109/IIPHDW.2018.8388338
    Miller, M. A., Kudela, R. M., Mekebri, A., Crane, D., Oates, S. C., Tinker, M. T., Staedler, M., Miller, W. A., Toy-Choutka, S., Dominik, C., Hardin, D., Langlois, G., Murray, M., Ward, K., & Jessup, D. A. (2010). Evidence for a novel marine harmful algal bloom: cyanotoxin (microcystin) transfer from land to sea otters. PLoS One, 5(9). https://doi.org/10.1371/journal.pone.0012576
    Mishra, S., Kumari, N., Häder, D.-P., & Sinha, R. P. (2022). Cyanobacterial Blooms and Their Implications in the Changing Environment. Advances in Environmental and Engineering Research, 3(1), 1-1. https://doi.org/10.21926/aeer.2201011
    Moreira, C., Vasconcelos, V., & Antunes, A. (2022). Cyanobacterial Blooms: Current Knowledge and New Perspectives. Earth, 3(1), 127-135. https://doi.org/10.3390/earth3010010
    O’Neil, J. M., Davis, T. W., Burford, M. A., & Gobler, C. J. (2012). The rise of harmful cyanobacteria blooms: the potential roles of eutrophication and climate change. Harmful algae, 14, 313-334. https://doi.org/10.1016/j.hal.2011.10.027
    O'Reilly, C. M., Sharma, S., Gray, D. K., Hampton, S. E., Read, J. S., Rowley, R. J., Schneider, P., Lenters, J. D., McIntyre, P. B., Kraemer, B. M., Weyhenmeyer, G. A., Straile, D., Dong, B., Adrian, R., Allan, M. G., Anneville, O., Arvola, L., Austin, J., Bailey, J. L.,…Zhang, G. (2015). Rapid and highly variable warming of lake surface waters around the globe. Geophysical Research Letters, 42(24). https://doi.org/10.1002/2015gl066235
    Paerl, H. W., Hall, N. S., & Calandrino, E. S. (2011). Controlling harmful cyanobacterial blooms in a world experiencing anthropogenic and climatic-induced change. Sci Total Environ, 409(10), 1739-1745. https://doi.org/10.1016/j.scitotenv.2011.02.001
    Paerl, H. W., & Paul, V. J. (2012). Climate change: links to global expansion of harmful cyanobacteria. Water Res, 46(5), 1349-1363. https://doi.org/10.1016/j.watres.2011.08.002
    Pereira, R., Carvalho, G., Garrote, L., & Nunes, U. J. (2022). Sort and Deep-SORT Based Multi-Object Tracking for Mobile Robotics: Evaluation with New Data Association Metrics. https://doi.org/10.3390/app12031319
    Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. Proceedings of the IEEE conference on computer vision and pattern recognition,
    Richardson, L. L. (1996). Remote Sensing of Algal Bloom Dynamics: New research fuses remote sensing of aquatic ecosystems with algal accessory pigment analysis. BioScience, 46(7), 492-501. https://doi.org/10.2307/1312927
    Rodriguez, E., Onstad, G. D., Kull, T. P., Metcalf, J. S., Acero, J. L., & von Gunten, U. (2007). Oxidative elimination of cyanotoxins: comparison of ozone, chlorine, chlorine dioxide and permanganate. Water Res, 41(15), 3381-3393. https://doi.org/10.1016/j.watres.2007.03.033
    Scholin, C., Doucette, G., Jensen, S., Roman, B., Pargett, D., Marin Iii, R., Preston, C., Jones, W., Feldman, J., & Everlove, C. (2009). Remote detection of marine microbes, small invertebrates, harmful algae, and biotoxins using the Environmental Sample Processor (ESP). Oceanography, 22(2), 158-167.
    Schunck, B. K. P. H. a. B. G. (1981). Determining Optical Flow. Artificial Intelligence 17(1-3), 185-203. https://doi.org/10.1016/0004-3702(81)90024-2
    Shi, K., Zhang, Y., Qin, B., & Zhou, B. (2019). Remote sensing of cyanobacterial blooms in inland waters: present knowledge and future challenges. Sci Bull (Beijing), 64(20), 1540-1556. https://doi.org/10.1016/j.scib.2019.07.002
    Vaiciute, D., Bucas, M., Bresciani, M., Dabuleviciene, T., Gintauskas, J., Mezine, J., Tiskus, E., Umgiesser, G., Morkunas, J., De Santi, F., & Bartoli, M. (2021). Hot moments and hotspots of cyanobacteria hyperblooms in the Curonian Lagoon (SE Baltic Sea) revealed via remote sensing-based retrospective analysis. Sci Total Environ, 769, 145053. https://doi.org/10.1016/j.scitotenv.2021.145053
    Vekaria, D., & Sinha, S. (2024). aiWATERS: an artificial intelligence framework for the water sector. AI in Civil Engineering, 3(1). https://doi.org/10.1007/s43503-024-00025-7
    Water, C. D. (2014). Guidelines for Canadian drinking water quality. Retrieved from https://www.canada.ca/content/dam/hc-sc/documents/programs/consultation-draft-guidelines-canadian-drinking-water-quality-trihalomethanes/consultation-draft-guidelines-canadian-drinking-water-quality-trihalomethanes.pdf
    Watson, S. B., Ridal, J., & Boyer, G. L. (2008). Taste and odour and cyanobacterial toxins: impairment, prediction, and management in the Great Lakes. Canadian Journal of Fisheries and Aquatic Sciences, 65(8), 1779-1796. https://doi.org/10.1139/f08-084
    Welch, G., & Bishop, G. (1995). An introduction to the Kalman filter.
    Wojke, N., Bewley, A., & Paulus, D. (2017). Simple online and realtime tracking with a deep association metric. 2017 IEEE international conference on image processing (ICIP),
    Wu, D., Li, R., Zhang, F., & Liu, J. (2019). A review on drone-based harmful algae blooms monitoring. Environ Monit Assess, 191(4), 211. https://doi.org/10.1007/s10661-019-7365-8
    Wu, J., Cao, Y., Wu, S., Parajuli, S., Zhao, K., & Lee, J. (2025). Current Capabilities and Challenges of Remote Sensing in Monitoring Freshwater Cyanobacterial Blooms: A Scoping Review. Remote Sensing, 17(5). https://doi.org/10.3390/rs17050918
    X. Li, C. L., W. Wang, G. Li, L. Yang and J. Yang. (2023). Generalized Focal Loss: Towards Efficient Representation Learning for Dense Object Detection. https://doi.org/10.1109/TPAMI.2022.3180392
    Zahir, M., Su, Y., Shahzad, M. I., Ayub, G., Rahman, S. U., & Ijaz, J. (2024). A review on monitoring, forecasting, and early warning of harmful algal bloom. Aquaculture, 593. https://doi.org/10.1016/j.aquaculture.2024.741351
    Zamyadi, A., Choo, F., Newcombe, G., Stuetz, R., & Henderson, R. K. (2016). A review of monitoring technologies for real-time management of cyanobacteria: Recent advances and future direction. TrAC Trends in Analytical Chemistry, 85, 83-96. https://doi.org/10.1016/j.trac.2016.06.023
    Zamyadi, A., Ho, L., Newcombe, G., Bustamante, H., & Prevost, M. (2012). Fate of toxic cyanobacterial cells and disinfection by-products formation after chlorination. Water Res, 46(5), 1524-1535. https://doi.org/10.1016/j.watres.2011.06.029

    下載圖示 校內：立即公開
    校外：立即公開
    QR CODE