利用邊緣裝置結合物件偵測技術應用於農場偵測害蟲系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	黃裕誠 Yu-Cheng Huang
論文名稱：	利用邊緣裝置結合物件偵測技術應用於農場偵測害蟲系統 Utilizing edge devices combined with object detection technology for pest detection system in agricultural farms
指導教授：	吳中實 Jung-Shyr Wu
口試委員:
學位類別：	碩士 Master
系所名稱：	資訊電機學院 - 通訊工程學系 Department of Communication Engineering
論文出版年：	2024
畢業學年度：	112
語文別：	中文
論文頁數：	88
中文關鍵詞：	智慧農業、邊緣裝置、物件偵測
外文關鍵詞：	smart agriculture, edge devices, object detection
相關次數：	點閱：11 下載：0
分享至:	分享至facebook 分享至twitter

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

隨者務農人數意願降低而逐年減少，智慧農業為一種當前重要的發展，目前的務農人以中高年齡居多，而隨著年齡的增長，體能上不如過往，生產力可能也會降低，故在農業問題上面臨了人力短缺的問題。且還有近幾年的全球暖化問題讓外在環境逐漸顯惡，在長時間高溫曝曬的環境可能也會讓人體無法負荷。於是結合邊緣裝置結合物件偵測技術，利用監控的方式讓農民可以不用長時間的待在高溫曝曬的環境中。網路技術的興起，物聯網技術也會跟著網路技術的興起而有所進步。利用物聯網技術來幫助農業上的問題，可以降低相關人力與成本。本論文提出了一套利用邊緣裝置結合物件偵測技術來辨識害蟲的系統，使用樹莓派用來做邊緣裝置，結合YOLOv5物件偵測技術實現辨識害蟲的功能，並將偵測到的影像截圖利用Line notify通知使用者進行後續處理，同時將影像偵測結果上傳備份至AWS雲端儲存桶內。本地端進行邊緣端的偵測模型訓練，並將訓練好的模型上傳至AWS雲端，以AWS雲端技術做為本地端與邊緣裝置之間的連結。這種方式大幅降低的資料傳輸延遲、降低成本、降低中心壓力及增加隱私安全性。

As the number of farmers willing to engage in agriculture decreases annually, smart agriculture has become an important current development. Currently, most farmers are of middle to advanced age, and as they grow older, their physical abilities may decline, potentially reducing productivity and leading to a shortage of labor in agriculture. Furthermore, the escalating global warming issue in recent years has worsened external environmental conditions, making prolonged exposure to high temperatures unbearable for humans.To address these agricultural challenges, integrating edge devices with object detection technology allows farmers to monitor conditions without prolonged exposure to extreme heat. With the rise of internet technology, IoT (Internet of Things) capabilities have advanced accordingly, offering solutions to agricultural issues that can reduce labor costs.This paper proposes a system that utilizes edge devices combined with object detection technology to identify pests. Specifically, it employs a Raspberry Pi as an edge device and integrates YOLOv5 object detection to achieve pest identification. Detected images are captured and notified to users via Line notify for further action, while detection results are also backed up in an AWS cloud storage bucket. Training of the edge-side detection model is conducted locally, and the trained model is uploaded to AWS cloud to establish the connection between local and edge devices using AWS cloud technologies. This approach significantly reduces data transmission latency, lowers costs, alleviates central processing burdens, and enhances privacy and security.

摘要    I
Abstract    II
誌謝    III
目錄    IV
圖目錄    VIII
表目錄    XI
第一章 緒論    1
1-1 前言    1
1-2 研究動機    2
1-3 論文架構    4
第二章 相關研究背景    5
2-1 物聯網(Internet of Things)    5
    2-1-1 邊緣運算（Edge Computing）    6
2-2 終端裝置與設備    8
    2-2-1 樹莓派(Raspberry Pi)    8
    2-2-2 Pi-camera    10
2-3 加密傳輸(Encrypted Transmission)    11
    2-3-1 安全外殼協定(Secure Shell Protocol)    11
    2-3-2 安全複製協定（Secure Copy Protocol）    12
2-4 神經網路介紹    13
    2-4-1 卷積神經網路(Convolutional Neural Network)    13
    2-4-2 基於區域的卷積神經網路(R-CNN)    15
2-5 物件偵測技術(Object Detection)    16
    2-5-1 YOLO (You Only Look Once)    18
2-6 機器學習(Machine Learning)    20
    2-6-1 Python簡介    21
    2-6-2 資料強化(Data Augmentation)    22
    2-6-3 增量學習(Incremental Learning)    22
2-7 Line簡介    23
    2-7-1 Line notify    23
2-8 亞馬遜雲端技術(Amazon Web Services)    24
    2-8-1 AWS IAM(Identity and Access Management)    25
    2-8-2 AWS S3 (Simple Storage Service)    26
第三章 系統情境與架構    27
3-1 伺服器端系統設計    28
    3-1-1 資料蒐集與標記    28
    3-1-2 模型訓練    30
    3-1-3 對原始資料進行資料強化    31
    3-1-4 終端裝置取得模型    32
3-2 邊緣端系統設計    35
    3-2-1 影像偵測    35
    3-2-2 發送通知    36
    3-2-3 雲端權限開通    37
3-3 模型評估指標    39
    3-3-1 混淆矩陣(Confusion Matrix)    39
    3-3-2 精確率(Precision)    40
    3-3-3 召回率(Recall)    40
    3-3-4 F1值(F1-Score)    41
    3-3-5 重疊度(IoU)    41
    3-3-6 信心分數(Confidence Score)    42
第四章 模擬與分析    43
4-1 模擬環境與硬體設備架構    44
4-2 模擬結果與分析    46
    4-2-1 實驗數據分析    47
    4-2-2 辨識實驗結果    59
第五章 結論與未來研究方向    66
參考文獻    67

                                

[1] Song Chaohui; Ma Longtao; Zhu Jiayun; Jia Pengfei; Yang Ning; Li Lihua,”Research on State Sensing Technology for Substation Equipment Based on Internet of Things”. 2023 IEEE 7th Conference on Energy Internet and Energy System Integration (EI2),2023,pp3832-3837.
[2] Y. J. Qu; X.G. Ming; S. Q. Qiu; Z. W Liu; X.Y. Zhang; Z.T. Hou,” A Framework for Smart Manufacturing Systems Based on the Stakeholders' Value”. 2018 IEEE International Conference on Advanced Manufacturing (ICAM),2018,pp239-242.
[3] Byeongkwan Kang; Seunghwan Kim; Myeong-In Choi,” Analysis of Types and Importance of Sensors in Smart Home Services”, 2016 IEEE 18th International Conference on High Performance Computing and Communications; IEEE 14th International Conference on Smart City; IEEE 2nd International Conference on Data Science and Systems (HPCC/SmartCity/DSS),2016,pp1388-1389.
[4] Mahbubur Rahman; Md Saidur Rahman Kohinoor; Aftar Ahmad Sami,” Enhancing Poultry Farm Productivity Using IoT-Based Smart Farming Automation System”, 2023 26th International Conference on Computer and Information Technology (ICCIT),2023,pp.1-6.
[5] 網路資料 online resources: Banana Pi。取自https://www.banana-pi.org/
[6] 網路資料 online resources: Jetson Nano 將現代人工智慧的強大功能帶至邊緣裝置。取自 https://www.nvidia.com/zh-tw/autonomous-machines/embedded-systems/jetson-nano/product-development/
[7] Jiran Sithiyopasakul; Tuanjai Archevapanich; Saran Sithiyopasakul, “Implementation of Cloud Computing and Internet of Things (IoT) by Performance Evaluation”, 2024 12th International Electrical Engineering Congress (iEECON),2024,pp.1-6.
[8] 網路資料 online resources：物聯網技術大解析。取自：https://www.ithome.com.tw/news/90461
[9] Deok Hee Nam,” A Comparative Study of Mobile Cloud Computing, Mobile Edge Computing, and Mobile Edge Cloud Computing”, 2023 Congress in Computer Science, Computer Engineering, & Applied Computing (CSCE),2023,pp.1219-1224
[10] 網路資料 online resources: 邊緣運算。取自https://www.fsp-group.com/tw/knowledge-app-42.html
[11] 網路資料 online resources: 樹莓派。取自https://zh.wikipedia.org/zh-tw/%E6%A0%91%E8%8E%93%E6%B4%BE
[12] 網路資料 online resources: GPIO簡介-奇特威科技。取自https://www.chipwaygo.com/doc/gpio_pin.php
[13] 網路資料 online resources: Raspberry Pi Camera Module V2 8MP 樹莓派相機模組 8百萬像素 | 攝像頭 | Sony IMX219。取自 https://piepie.com.tw/product/camera-module-v2
[14] Computer Networking A Top-Down Approach 7th edtion國際版中文翻譯本補充資料,第八章網路安全,pp8.11-8.17
[15] Tatu Ylonen,” SSH Key Management Challenges and Requirements”, 2019 10th IFIP International Conference on New Technologies, Mobility and Security (NTMS),2019,pp1-5
[16] Mufid Ridlo Effendi; Ridwan Sidik Al-Falah; Sarbini; Nanang Ismail,” IoT-Based Battery Monitoring System in Solar Power Plants with Secure Copy Protocol (SCP)”, 2021 7th International Conference on Wireless and Telematics (ICWT),2021,pp 1-4
[17] Junxun Huang, Xuemei Ma, Yulong Wang, Xiao Li, “Real-time Video Fire Detection via Convolutional Neural Networks”, 2023 IEEE 2nd International Conference on Electrical Engineering, Big Data and Algorithms(EEBDA), 2023, pp.475-481.
[18] 網路資料 online resources：資料分析&機器學習。取自：https://medium.com/jameslearningnote/%E8%B3%87%E6%96%99%E5%88%86%E6%9E%90-%E6%A9%9F%E5%99%A8%E5%AD%B8%E7%BF%92-%E7%AC%AC5-1%E8%AC%9B-%E5%8D%B7%E7%A9%8D%E7%A5%9E%E7%B6%93%E7%B6%B2%E7%B5%A1%E4%BB%8B%E7%B4%B9-convolutional-neural-network-4f8249d65d4f
[19] 網路資料 online resources： A Comprehensive Guide to Convolutional Neural Networks。取自：https://towardsdatascience.com/a-comprehensive-guide-to-convolutional-neural-networks-the-eli5-way-3bd2b1164a53
[20] Ross Girshick; Jeff Donahue; Trevor Darrell; Jitendra Malik, “Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation”, 2014 IEEE Conference on Computer Vision and Pattern Recognition,2014,pp580-587.
[21] 網路資料 online resources：深度學習-什麼是one stage，什麼是two stage 物件偵測。取自：
https://chih-sheng-huang821.medium.com/%E6%B7%B1%E5%BA%A6%E5%AD%B8%E7%BF%92-%E4%BB%80%E9%BA%BC%E6%98%AFone-stage-%E4%BB%80%E9%BA%BC%E6%98%AFtwo-stage-%E7%89%A9%E4%BB%B6%E5%81%B5%E6%B8%AC-fc3ce505390f
[22] Thammi Reddy Konala; Anusha Nammi; Divya Sree Tella,” Analysis of Live Video Object Detection using YOLOv5 and YOLOv7”, 2023 4th International Conference for Emerging Technology (INCET),2023,pp1-6
[23] Thomas Rincy N; Roopam Gupta,” A Survey on Machine Learning Approaches and Its Techniques”, 2020 IEEE International Students' Conference on Electrical,Electronics and Computer Science (SCEECS),2020,pp1-6
[24] Danushka Senarathna; Spyros Tragoudas; Kiriti Nagesh Gowda,” Detection and Quantization of Data Drift in Image Classification Neural Networks”, 2023 IEEE 24th International Conference on High Performance Switching and Routing (HPSR),2023,pp38-42
[25] Hailong Chen; Chunli Liu,”Research and application of cluster analysis algorithm”, Proceedings of 2013 2nd International Conference on Measurement, Information and Control,2013,pp575-579
[26] K Manjunath; K Suresh,” Adaptive Task Scheduling Procedure By Using Improvised Q-Learning Technique In Internet Of Things”, 2022 International Conference on Computing, Communication, Security and Intelligent Systems (IC3SIS),2022,pp1-6
[27] 網路資料 online resources: 什麼是python。取自:
https://aws.amazon.com/tw/what-is/python/
[28] 網路資料 online resources: Pytorch 介紹。取自:
https://ithelp.ithome.com.tw/articles/10273262?sc=rss.iron
[29] Ariel Yonatan Alin; Kusrini; Kusrini; Kumara Ari Yuana,” Data Augmentation Method on Drone Object Detection with YOLOv5 Algorithm”, 2023 Eighth International Conference on Informatics and Computing (ICIC),2023,pp1-6
[30] Muhammad Abdullah Hafeez; Adnan Ul-Hasan; Faisal Shafait,” Incremental Learning of Object Detector with Limited Training Data”, 2021 Digital Image Computing: Techniques and Applications (DICTA),2021,pp1-8
[31] 網路資料 online resources: LINE。取自:
https://zh.wikipedia.org/zh-tw/LINE
[32] 網路資料 online resources: 什麼是 LINE Notify？。取自:
https://ithelp.ithome.com.tw/articles/10241078
[33] Jiran Sithiyopasakul; Tuanjai Archevapanich; Saran Sithiyopasakul, “Implementation of Cloud Computing and Internet of Things (IoT) by Performance Evaluation”, 2024 12th International Electrical Engineering Congress (iEECON),2024,pp.1-6
[34] 網路資料 online resources: Amazon Web Services。取自: https://aws.amazon.com/tw/free/?gclid=EAIaIQobChMIjY3Qj7O0hgMVoG0PAh1iThLuEAAYASAAEgLnt_D_BwE&trk=cf0164b0-a143-4575-9482-010451b8f1d5&sc_channel=ps&ef_id=EAIaIQobChMIjY3Qj7O0hgMVoG0PAh1iThLuEAAYASAAEgLnt_D_BwE:G:s&s_kwcid=AL!4422!3!595905315029!e!!g!!aws!17115100998!136234409996&all-free-tier.sort-by=item.additionalFields.SortRank&all-free-tier.sort-order=asc&awsf.Free%20Tier%20Types=*all&awsf.Free%20Tier%20Categories=*all
[35] 網路資料 online resources: 什麼是 IAM？。取自: https://docs.aws.amazon.com/zh_tw/IAM/latest/UserGuide/introduction.html
[36] 網路資料 online resources: 探索AWS：從菜鳥到熟練的完全指南。取自: https://hackmd.io/@gdw7l5sPTOyNv76kZ_twjA/SJdVjf1D2
[37] 網路資料 online resources: 雲端物件儲存–Amazon S3。取自: https://aws.amazon.com/tw/s3/
[38] 網路資料 online resources: YOLOv5 in PyTorch。取自: https://github.com/ultralytics/yolov5
[39] Xinfei Jin; Fulin Su,“Aircraft Recognition Using ISAR Image Based on Quadrangle-points Affine Transform”,2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI),2022,pp1-6
[40] 網路資料 online resources: 機器學習\統計方法。取自:https://dysonma.github.io/2020/12/05/%E6%A9%9F%E5%99%A8%E5%AD%B8%E7%BF%92%E5%B8%B8%E8%A6%8B%E7%9A%84%E8%A9%95%E4%BC%B0%E6%8C%87%E6%A8%99/
[41] Luc-Etienne Pommé; Romain Bourqui; Romain Giot,“Relative Confusion Matrix: Efficient Comparison of Decision Models”,2022 26th International Conference Information Visualisation (IV) ,2022,pp98-103.
[42] 網路資料 online resources:深度學習中IoU(Intersection over Union)的概念理解。取自: https://blog.csdn.net/IAMoldpan/article/details/78799857
[43] Joseph Redmon; Santosh Divvala; Ross Girshick; Ali Farhadi,” You Only Look Once: Unified, Real-Time Object Detection”, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2016,pp779-788

簡易檢索 / 詳目顯示

相關論文