• Login
    • Library Home
    View Item 
    •   BracU IR
    • School of Data and Sciences (SDS)
    • Department of Computer Science and Engineering (CSE)
    • Thesis & Report, BSc (Computer Science and Engineering)
    • View Item
    •   BracU IR
    • School of Data and Sciences (SDS)
    • Department of Computer Science and Engineering (CSE)
    • Thesis & Report, BSc (Computer Science and Engineering)
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Detection and classification of Mango Leaf diseases utilizing convolutional neural network models

    Thumbnail
    View/Open
    19101299,19301083,20301359,20301048,20101582_CSE.pdf (3.214Mb)
    Date
    2024-05
    Publisher
    Brac University
    Author
    Ferdous, Md.Rubaiat
    Amin, Abdulla Al
    Shaily, Senjuti Sarkar
    Ricky, Fabliha Tarannum
    Tabeeb, Tahmeed
    Metadata
    Show full item record
    URI
    http://hdl.handle.net/10361/24293
    Abstract
    Plant diseases, particularly those affecting crop plants, pose a significant danger to world food security by compromising the quality and yield of agricultural produce. Mango leaf disease is one such example. Mango leaf diseases are quite harmful since they can significantly lower crop yields of mangos, both in quantity and quality. Therefore, it is critical to identify leaf diseases in crops like mangos as soon as possible in order to take prompt preventive action. In large cultivated areas where mangoes are planted in significant quantities, the amount of manual inspection can be significantly reduced by mechanizing the process of disease recognition. With their exceptional ability to identify complex patterns in images, Convolutional Neural Networks (CNN) have great potential for automating the identification of illnesses affecting mango leaves. CNNs have been used in several studies with impressive accuracy rates, opening the door to improved crop management techniques and early disease diagnosis. In order to identify various illnesses, a research study using CNN—one of the most advanced deep learning algorithms—is presented in this work. CNN is used to segment and classify pictures of mango leaves. A number of CNN models, including Xception, MobileNetV2, ResNet50, DenseNet201, and ResNet50, have been used to accurately identify and categorize diseases, which will ultimately improve the production and health of mango crops. MangoLeafBD, a dataset that was acquired from Kaggle, has been used, and XAI (Explainable Artificial Intelligence), specifically Grad-CAM (Gradient-weighted Class Activation Mapping) and LIME (Local Interpretable Model-agnostic Explanations), were used to understand the rationale behind the decisions made by the applied models. Using these CNN models approach, it was observed that ResNet50 performs better than other models with 98% F1 score in identification and classification of mango leaf diseases.
    Keywords
    Mango leaf disease; Convolutional neural network; Deep learning; Artificial Intelligence; Machine learning
     
    LC Subject Headings
    Machine learning.; Data mining.; Artificial Intelligence.; Convolutional neural networks.; Plant diseases--Detection.
     
    Description
    This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2024.
     
    Cataloged from PDF version of thesis.
     
    Includes bibliographical references (pages 40-43).
    Department
    Department of Computer Science and Engineering, Brac University
    Type
    Thesis
    Collections
    • Thesis & Report, BSc (Computer Science and Engineering) [1600]

    Copyright © 2008-2024 Ayesha Abed Library, Brac University 
    Contact Us | Send Feedback
     

     

    Policy Guidelines

    • BracU Policy
    • Publisher Policy

    Browse

    All of BracU Institutional RepositoryCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

    My Account

    LoginRegister

    Copyright © 2008-2024 Ayesha Abed Library, Brac University 
    Contact Us | Send Feedback