簡易檢索 / 詳目顯示

回結果列表
研究生： 陳忠順
Toni Setiawan Jaya
論文名稱： 利用物聯網圖像處理於森林火災偵測研究
Forest Fires Detection Using IoT Image Processing
指導教授： 賴維祥
Lai, Wei-Hsiang
學位類別： 碩士
Master
系所名稱： 工學院 - 能源工程國際碩博士學位學程
International Master/Doctoral Degree Program on Energy Engineering
論文出版年： 2022
畢業學年度： 110
語文別： 英文
論文頁數： 64
中文關鍵詞： 森林火災 、火災偵測 、影像處理 、網絡服務器 、無人飛機
外文關鍵詞： Forest fires, Fire detection, Image processing, IoT, Unmanned aerial vehicle
相關次數： 點閱：106  下載：0 
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

    • 森林火災是造成森林損失的形式。如眾所周知，森林可以吸收二氧化碳，產生氧氣，成為瀕危動植物群的棲息地，且森林可以減少全球變暖。因此森林火災是全球關注的問題。 20世紀初，發生了無數的野火，尤其是在印度尼西亞。本研究提出一種向消防隊發送檢測信息以即早發現森林火災並儘快做出反應的方法，這樣工作人員就不必經常檢查監視情況。其中它具有內置的人工智能（Yolov5）。 Raspberry Pi 4 或Jetson Nano 等平台與 GoPro / Camera Surveillance 等相機操作整合，允許添加多個獨立的 Raspberry Pi 4 或 Jetson Nanos 以進行無人機監控。在這個項目中使用網路(Web)伺服器相當可靠，使本方法可以使用無人機監控森林，並使用 640x360 的圖像進行接近即時檢測，其中Jetson Nano 可以識別高達 10fps，而 Raspberry Pi 4 可以識別高達 0.4fps。

    Forest fires are a form of loss arising from deforestation. As we know that forest can absorb carbon dioxide, produce oxygen, become habitats for endangered flora and fauna, and forests can reduce global warming. Forest fires are a global concern. At the beginning of the 20th century, there were numerous wildfires, especially in Indonesia. A method of sending detections to the fire brigade to recognize forest fires early and respond as quickly as possible, so that staff do not have to constantly check for surveillance. It has built-in artificial intelligence (Yolov5). Platforms such as the Raspberry Pi 4 and Jetson Nano are integrated with camera actions such as GoPro / Camera Surveillance, allowing you to add multiple standalone Raspberry Pi 4 or Jetson Nanos for drone monitoring. Using a web server for this project is more reliable because you can monitor the forest with a drone. As the result with the image 640x360 for detecting, the Jetson Nano can recognize up to 10fps and the Raspberry Pi 4 can recognize up to 0.4fps close to real time information.

    摘要 i Abstract ii Acknowledgement iii Table of Contents iv List of Tables vii List of Figures viii Chapter 1 Introduction 1 1.1 Background 1 1.2 A Schematic of algorithm 2 1.3 Motivation and objectives 3 1.4 Thesis Architecture 3 1.5 Limitation 4 Chapter 2 Literature Review 5 2.1 Type of forest fire detection 5 2.2 Forest Guardian 5 2.3 IoT System for Forest Monitoring 6 2.4 Low Cost LoRa based Network for Forest Fire Detection 6 2.5 Development of a Surveillance System for Forest Fire Detection and Monitoring Using Drones 7 2.6 Forestry Monitoring System using LoRa and Drone 7 2.8 Aerial Forest Fire Surveillance – Evaluation of Forest Fire Detection Model Using Aerial Videos 8 2.9 Fire Detection System using Raspberry Pi 8 2.10 Raspberry Pi 9 2.11 Open-CV 9 2.12 Machine Learning (Tensor Flow) 9 2.13 Matplotlib 10 2.14 Seaborn 10 2.15 Cython 10 2.16 Libatlas-base-dev 10 2.17 Yolov5 11 2.18 Pyyaml 11 Chapter 3 Research Methods and Experimental Equipment 12 3.1 Proposed System 12 3.2 Experimental Equipment 12 3.3 Build the Component 13 3.4 Build the AI on Raspberry Pi 4 14 3.5 Build the AI on Jetson Nano 14 3.6 Training AI for Detect Fire Using Google Colab 15 3.7 Training AI for Detect Fire and Person Using Google Colab 15 Chapter 4 Testing and Observation Results 17 4.1 Testing the Model from Yolov5 17 4.2 Training AI Fire Model 17 4.2 Detect AI Fire Model 19 4.3 Testing AI Fire Model on Raspberry Pi 4 21 4.4 Testing AI Fire Model on Jetson Nano 22 4.5 Training AI Person and Fire Model 23 4.6 Detect AI Person and Fire Model 27 4.7 Testing AI person and Fire Model on Raspberry Pi 4 27 4.8 Testing AI Person and Fire Model on Jetson Nano 29 4.9 Detecting Video from UAV Capture 31 4.10 Build Web Server 33 4.11 Testing Web Server 37 4.12 GPS and 4G module 39 4.13 Test in High Building (6th Floor) 41 Chapter 5 Conclusions and Suggestions 43 References 44 Appendices 47 Program Python in Jeston Nano 47 Server Program 55

    [1]. Channel News Asia, (2021), Indonesia president warns of forest fires as hot spots detected. Channel News Asia.com. https://www.channelnewsasia.com/asia/indonesia-forest-fires-hot-spots-haze-joko-widodo-366856
    [2]. Kontan, (2020), KLHK catat ada 8.254 ha luas kebakaran hutan dan lahan periode Januari-Maret 2020. Kontan. https://nasional.kontan.co.id/news/klhk-catat-ada-8254-ha-luas-kebakaran-hutan-dan-lahan-periode-januari-maret-2020
    [3]. Sahin, Y. G., & Ince, T. , (2009), Early forest fire detection using radio-acoustic sounding system. Sensors (Basel, Switzerland), 9(3), 1485–1498. https://doi.org/10.3390/s90301485
    [4]. Papan, J., Jurecka, M., & Puchyova, J., (2012), WSN for forest monitoring to prevent illegal logging.
    [5]. BBC News, (2019), Indonesia haze: Why do forests keep burning? BBC News. https://www.bbc.com/news/world-asia-34265922
    [6]. Marcu, A., Suciu, G., Olteanu, E.L., Miu, D., Drosu, A., & Marcu, I.M., (2019), IoT System for Forest Monitoring. 2019 42nd International Conference on Telecommunications and Signal Processing (TSP), 629-632.
    [7]. Vega-Rodríguez, R., Sendra, S., Lloret, J., Romero-Díaz, P., & García-Navas, J.L., (2019), Low Cost LoRa based Network for Forest Fire Detection. 2019 Sixth International Conference on Internet of Things: Systems, Management and Security (IOTSMS), 177-184.
    [8]. Allauddin, M. S. Kiran, G. S. Kiran, G. R. Srinivas, G. Mouli, G. U. R. and Prasad, P. V., (2019), Development of a Surveillance System for Forest Fire Detection and Monitoring using Drones. doi:10.1109/IGARSS.2019.8900436
    [9]. COCO Consortium, (2015), COCO Common Object in Context. https://cocodataset.org/#download
    [10]. Q-engineering, (2021), Install PyTorch on Jetson Nano. https://qengineering.eu/install-pytorch-on-jetson-nano.html
    [11]. Ai-coodinator, (2021), xml_to_textYolo. Retrieved from https://github.com/ai-coodinator/xml_to_textYolo
    [12]. Kareemov1000, Wifi Drone, Retrieved from https://thenounproject.com/term/wifi-drone/1054424/
    [13]. Park, S., Yun, S., Kim, H., Kwon, R., Ganser, J., and Anthony, S., (2018), Forestry Monitoring System using LoRa and Drone, WIMS '18: Proceedings of the 8th International Conference on Web Intelligence, Mining and Semantics doi:10.1145/3227609.3227677
    [14]. Gaitan, N.-C., and Hojbota, P., (2020), Forest Fire Detection System using LoRa Technology, International Journal of Advanced Computer Science and Applications, 11(5). doi:10.14569/IJACSA.2020.0110503
    [15]. Hanh, Dang-Ngoc and Hiew, Nguyen-Trung, (2019), Aerial Forest Fire Surveillance - Evaluation of Forest Fire Detection Model using Aerial Videos, 2019 International Conference on Advanced Technologies for Communications (ATC) doi:10.1109/ATC.2019.8924547
    [16]. Khan, M. N. A. Tanveer, T. Khurshid, K. Zaki, H. and Zaidi S. S. I., (2019), Fire Detection System using Raspberry Pi. doi:10.1109/CISCT.2019.8777414
    [17]. PyYAML, (2021), PyPI. https://pypi.org/project/PyYAML/
    [18]. Raspberry Foundation, (2021), About. https://www.raspberrypi.org/about/
    [19]. Open CV, (2021), About. https://opencv.org/about/
    [20]. Google Developer, (2021), Introduction to TensorFlow. https://developers.google.com/machine-learning/crash-course/first-steps-with-tensorflow/toolkit?hl=id
    [21]. Matplotlib, (2021), Matplotlib: Visualization with Python. https://matplotlib.org/stable/index.html
    [22]. Seaborn, (2021), seaborn: statistical data visualization. https://seaborn.pydata.org/
    [23]. Cyhton, (2021), C-Extensions for python. https://cython.org/
    [24]. Debian, (2021), Libatlas-base-dev. https://packages.debian.org/sid/libatlas-base-dev
    [25]. Ultralytics, (2021), Pytorch. https://pytorch.org/hub/ultralytics_yolov5/
    [26] Shenzhen Micro Snow Electronics Co., Ltd., (2022), https://www.waveshare.net/wiki/SIM7600G-H_4G_HAT
    [27] Nvidia, (2021), Deep Learning Framework Documentation. https://docs.nvidia.com/deeplearning/frameworks/install-tf-jetson-platform/index.html

    下載圖示
    2026-01-14公開
    QR CODE