簡易檢索 / 詳目顯示

回結果列表
研究生： Natrada Kanapornchai
Natrada Kanapornchai
論文名稱： 應用YOLO模型於IOS行動裝置辨識榴槤葉病之研究
Detecting Durian Leaf Disease using YOLO on iOS Mobile Devices
指導教授： 呂永和
Yung-Ho Leu
口試委員: 陳雲岫
Yun-Shiow Chen
楊維寧
Wei-Ning Yang
學位類別： 碩士
Master
系所名稱： 管理學院 - 資訊管理系
Department of Information Management
論文出版年： 2023
畢業學年度： 112
語文別： 英文
論文頁數： 55
外文關鍵詞： YOLO, Durian Leaf Diseases, Mobile Application
相關次數： 點閱：29  下載：5 
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

    • This research aims to address the challenge of accurately diagnosing durian leaf diseases, which can be difficult for non-experts and can lead to ineffective treatment or crop loss. The research involves utilizing a pre-trained machine learning model to classify durian leaf images into different disease categories and then integrating this model into an iOS mobile application for convenient and accessible disease diagnosis.
    The model was trained on a dataset of durian leaf images labeled with various disease classes, and its performance was evaluated using standard metrics such as accuracy and precision. The mobile application was developed using Xcode and Swift, and it includes a user-friendly interface for capturing or uploading images and viewing disease classification results.
    The resulting mobile application demonstrates the potential for machine learning to support accurate and convenient durian leaf disease diagnosis, with potential implications for precision agriculture.

    ACKNOWLEDGEMENT i ABSTRACT ii LIST OF FIGURES iii LIST OF TABLES iv Chapter 1 Introduction 1 1.2 Research Objectives 2 Chapter 2 Literature Review 3 2.1 Disease & Pest Detection 3 Chapter 3 Methodology 7 3.1 Data Collection Methodology 7 3.1.1 Primary Data Collection 7 3.1.2 Secondary Data Collection 7 3.1.3 Data Screening and Validation 8 3.1.4 Rationale for Data Source Selection 8 3.2 Research Method and Design 9 3.2.1 YOLOv8 model training process 9 3.2.2 PyTorch to CoreML Framework Conversion 11 3.2.3 ‘D-Leaf Diagnostics’ mobile application’s GUI 11 3.2.4 Recommendation Documents 13 3.4 Data Preprocessing 15 3.4.1 Image Annotation 15 3.4.2 Image Augmentation 16 3.5 YOLOv8 model 19 3.5.1 Detail of YOLOv8 blocks 19 3.5.2 YOLOv8 Architecture 24 Chapter 4 Results and Discussion 33 4.1.1 Data gathering and collection 33 4.1.2 Experimental environment and parameter setting 35 4.1.3 Performance comparison of YOLOv7-tiny, YOLOv7 and YOLOv8 36 4.1.4 YOLOv8n with 150 epochs results and discussion 38 4.1.5 Performance comparison of YOLOv8n model with parameter adjustment 40 4.1.6 Visualization comparison and result 42 4.1.7 Performance comparison of YOLOv8n with other detectors 45 4.2 ‘D-Leaf Diagnostic’ mobile application 45 4.2.1 Development environment setting 46 4.2.2 Model Integration 46 4.2.3 GUI of mobile application 47 Chapter 5 Conclusion 51 References 53

    [1] Skyline, A. (2023, May 9). Retrieved from https://www.facebook.com/photo.php?fbid=639863404850977&id=100064819685437&set=a.568390251998293
    [2] Nations, F. a. (2023). https://agfstorage.blob.core.windows.net/misc/FP_com/2023/11/30/Aduc.pdf.
    [3] Xijia, Q. (2023, May 29). Retrieved from https://www.globaltimes.cn/page/202305/1291568.shtml
    [4] China, G. a. (2019, 6 28). Retrieved from http://english.customs.gov.cn/Statics/b768df9a-48db-43d3-a2e1-5c7b58ce0947.html
    [5] Office of Agriculture Economics (Thailand), O. o. (2023, 07 21). Retrieved from https://www.oae.go.th/view/1/รายละเอียดข่าว/ข่าวทั้งหมด
    [6] Department, S. R. (2023, June 8). Retrieved from https://www.statista.com/statistics/1319645/thailand-monthly-export-value-of-durians/
    [7] English, K. (2023, August 23). Retrieved from https://www.khaosodenglish.com/featured/2023/08/23/thailand-acts-rapidly-after-china-rejected-worm-infested-durians/
    [8] Thailand, N. (2023, August 22). Worms put Chumphon durians at risk of export ban . Retrieved from https://www.nationthailand.com/thailand/economy/40030412
    [9] Intan Nur Ainni Mohamed Azni, S. S. (2019). Pathogenicity of Malaysian Phytophthora palmivora on cocoa, durian, rubber and oil palm determines the threat of bud rot disease. https://doi.org/10.1111/efp.12557.
    [10] Veeranee Tongsri, P. S. (2016). Leaf Spot Characteristics of Phomopsis Durionis on Durian (Durio Zibethinus Murray) and Latent Infection of the Pathogen. https://www.researchgate.net/publication/296474332_Leaf_Spot_Characteristics_of_Phomopsis_Durionis_on_Durian_Durio_Zibethinus_Murray_and_Latent_Infection_of_the_Pathogen.
    [11] Nishant Shelar, S. S. (2022). Plant Disease Detection Using Cnn. ITM Web of Conferences. 40. India: EDP Sciences. https://doi.org/10.1051/itmconf/20224403049
    [12] Vijai Singh, A. M. (2016). Detection of plant leaf diseases using image segmentation and soft computing techniques. China Agricultural University. http://dx.doi.org/10.1016/j.inpa.2016.10.005.
    [13] V.Asha. (2023). An Enhanced Deep Learning Algorithms for Image Recognition and Plant Leaf Disease Detection. Second International Conference on Augmented Intelligence and Sustainable Systems . India: DOI: https://doi.org/10.1109/ICAISS58487.2023.10250727
    [14] Muhammad Juman Jhatial, R. A. (2022). Deep Learning-Based Rice Leaf Diseases Detection Using Yolov5. Sukkur IBA Journal of Computing and Mathematical Sciences. http://dx.doi.org/10.30537/sjcms.v6i1.1009
    [15] H. Orchi, M. S. (2023). Real-Time Detection of Crop Leaf Diseases Using Enhanced YOLOv8 algorithm. 2023 International Wireless Communications and Mobile Computing (IWCMC), (pp. 1690-1696). Marrakesh, Morocco. https://doi.org/10.1109/IWCMC58020.2023.10182573
    [16] Soeb, M. J. (2023, April 13). Tea leaf disease detection and identification based on YOLOv7 (YOLO-T). https://doi.org/10.1038/s41598-023-33270-4
    [17] Zijia Yang, H. F. (2023, May 9). Tea Tree Pest Detection Algorithm Based on Improved Yolov7-Tiny. Agriculture, 13. https://doi.org/10.3390/agriculture13051031
    [18] Xing Gao, Z. T. (2023). HSSNet: A End-to-End Network for Detecting Tiny Targets of Apple Leaf Diseases in Complex Backgrounds . Switzerland https://doi.org/10.3390/plants12152806
    [19] N. A. H. A. Halim, S. S. (2023, April 30). Durian Tree Type Identification Based on Durian Leaves. Evolution in Electrical and Electronic Engineering, 4, 551-558. https://publisher.uthm.edu.my/periodicals/index.php/eeee/article/view/11273
    [20] A. L. Sabarre, A. S. (2021, April 29). Development of durian leaf disease detection on Android device. 11(No.6), 4962-4971. http://doi.org/10.11591/ijece.v11i6.pp4962-4971
    [21] Xiaoxiao Shi, J.-F. P. (2012). Learning from Heterogeneous Sources via Gradient Boosting Consensus. http://dx.doi.org/10.1137/1.9781611972825.20
    [22] K. Crammer, M. K. (2006). Learning from Multiple Sources. University of Pennsylvania Philadelphia, Canada. https://www.cis.upenn.edu/~mkearns/papers/multisource-jmlr.pdf
    [23] Gallenero, J. A., & Villaverde, J. (2023). Identification of Durian Leaf Disease Using Convolutional Neural Network. 2023 15th International Conference on Computer and Automation Engineering (ICCAE) (pp. 172-177). Sydney https://doi.org/10.1109/ICCAE56788.2023.10111159
    [24] N. Girard, Y. T. (2018). End-to-End Learning of Polygons for Remote Sensing Image Classification. IEEE International Geoscience and Remote Sensing Symposium (pp. 2083-2086). https://doi.org/10.1109/IGARSS.2018.8518116
    [25] Lluís Castrejón, K. K. (2017). Annotating Object Instances with a Polygon-RNN. Computer Vision and Pattern Recognition (pp. 4485-4493). https://doi.org/10.1109/CVPR.2017.477.
    [26] Wu, Q. &. (Nov, 2023). A classification method for soybean leaf diseases based on an improved ConvNeXt model. DOI:10.1038/s41598-023-46492-3.
    [27] Plantwise. (n.d.). PEST MANAGEMENT DECISION GUIDE. Retrieved from plantwiseplusknowledgebank: https://plantwiseplusknowledgebank.org/doi/pdf/10.1079/pwkb.20187800447
    [28] DoAE. (n.d.). ศูนย์วิทยบริการเพื่อส่งเสริมการเกษตร. Retrieved from https://esc.doae.go.th
    [29] Sirichewakesron, P.-o. (2016). สาเหตุโรคใบจุดทุเรียน. Home Agricultural Magazine, 212-214.
    [30] คูหาพิทักษ์ธรรม, ม. (2019). Characterization of morphology and pathogenicity of Phytophthora palmivora a causal agent of root rot and stem rot of durian in Thailand. Thailand: Burapha University.
    [31] Roboflow Doc. ( 2023). Retrieved from Roboflow Doc: https://docs.roboflow.com/annotate/annotation-tools
    [32] James F. Mullen Jr., F. R. (2019). Comparing the Effects of Annotation Type on Machine Learning Detection Performance. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), (pp. 855-861). Long Beach, CA, USA.https://doi.org/10.1109/CVPRW.2019.00114
    [33] Dvornik, N. M. (2018). On the Importance of Visual Context for Data Augmentation in Scene Understanding. IEEE Transactions on Pattern Analysis and Machine Intelligence (pp. 2014-2028). https://doi.org/10.1109/TPAMI.2019.2961896.
    [34] Kandel, I. C. (2022). Brightness as an Augmentation Technique for Image Classification. Emerging Science Journal. https://doi.org/10.28991/esj-2022-06-04-015.
    [35] Eunkyeong Kim, J. K. (2021). Adaptive Data Augmentation to Achieve Noise Robustness and Overcome Data Deficiency for Deep Learning. https://doi.org/10.3390/app11125586.
    [36] glenn-jocher, f. (2023, 11 12). Ultralytics YOLOv8 Docs. Retrieved from https://docs.ultralytics.com/models/yolov8/
    [37] Jocher, G. (2023, May 13). Paper of YOLOv8. Retrieved from https://github.com/ultralytics/ultralytics/issues/2572
    [38] Zhang, Z. (2023). Drone-YOLO: An Efficient Neural Network Method for Target Detection in Drone Images. DOI:10.3390/drones7080526.
    [39] Jocher, G. (2023, July 13). Retrieved from https://github.com/ultralytics/ultralytics/issues/3678#issuecomment-1633520959
    [40] RangeKing. (2023, Jan 10). Brief summary of YOLOv8 model structure. Retrieved from https://github.com/ultralytics/ultralytics/issues/189
    [41] Hidayatullah, D. P. (2023, Nov). YOLOv8 Architecture Detailed Explanation - A Complete Breakdown. Retrieved from https://youtu.be/HQXhDO7COj8?si=QwZa9uFK4ry7IYp3
    [42] Mulan Qiu, L. H.-H. (2022). ASFF-YOLOv5: Multielement Detection Method for Road Traffic in UAV Images Based on Multiscale Feature Fusion. MDPI (p. 3498). https://doi.org/10.3390/ rs14143498.
    [43] Yajnik, A. (2023, Oct 18). Computer Vision: Hands-on implementation onYOLOv8. Retrieved from https://medium.com/@ayushyajnik2/computer-vision-hands-on-implementation-onyolov8-65bf5d682c62
    [44] Rath, S. (2023, January 10). YOLOv8 : Comprehensive Guide to State Of The Art Object Detection. Retrieved from https://learnopencv.com/ultralytics-yolov8/
    [45] Tsung-Yi Lin, P. G. (2017). Focal Loss for Dense Object Detection. IEEE International Conference on Computer Vision (ICCV) (pp. 2980-2988). https://openaccess.thecvf.com/content_iccv_2017/html/Lin_Focal_Loss_for_ICCV_2017_paper.html.
    [46] Xiang Li1, 2. ,. (2020). Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection. 34th Conference on Neural Information Processing Systems . Vancouver, Canada.: https://proceedings.neurips.cc/paper/2020/file/f0bda020d2470f2e74990a07a607ebd9-Paper.pdf.
    [47] Xufei Wang, J. S. (2021). CIoU considers three geometric factors concurrently: overlap area, central point distance, and aspect ratio between the two bounding boxes of the ground truth and the prediction, with the aim of improving object detection performance. https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9497076.
    [48] Wang, P. &. (2021). DGANet: Dynamic Gradient Adjustment Anchor-Free Object Detection in Optical Remote Sensing Images. http://dx.doi.org/10.3390/rs13091642.
    [49] glenn-jocher. (2023, November 6). How does YOLOv8 anchor-free detection work? Retrieved from github: https://github.com/ultralytics/ultralytics/issues/3362
    [50] Nelson, J. (2020, Octorber 28). Train Test Split Guide and Overview. Retrieved from https://blog.roboflow.com/train-test-split-with-roboflow/
    [51] Jing Wang, Z. W. (2022). Model Lightweighting for Real-time Distraction Detection on Resource-Limited Devices. Computational Intelligence and Neuroscience. https://doi.org/10.1155/2022/7360170.
    [52] Dillon Reis, J. K. (2023). Real-Time Flying Object Detection with YOLOv8. Georgia Institute of Technology. https://arxiv.org/pdf/2305.09972.pdf.
    [53] GitBook. (2019). Convergence. Retrieved from https://machine-learning.paperspace.com/wiki/convergence
    [54] Apple. (n.d.). Xcode. Retrieved from https://developer.apple.com/xcode/
    [55] Inc, A. (n.d.). Swift. Retrieved from https://www.swift.org
    [56] developer, A. (n.d.). SwiftUI. Retrieved from https://developer.apple.com/xcode/swiftui/
    [57] glenn-jocher. (2023, November 12). Retrieved from https://docs.ultralytics.com/modes/export/
    [58] glenn-jocher. (2023, July 16). Retrived from https://github.com/ultralytics/ultralytics/blob/main/ultralytics/cfg/models/v8/yolov8.yaml
    [59] Rosebrock, Adrian (2016, November 7) Retrived from https://pyimagesearch.com/2016/11/07/intersection-over-union-iou-for-object-detection/

    QR CODE