View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Agricultural Technology
      • UT - Agricultural and Biosystem Engineering
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Agricultural Technology
      • UT - Agricultural and Biosystem Engineering
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Deteksi Penyakit Leaf Scorch, Leaf Blight, dan Leaf Spot pada Tanaman Stroberi (Fragaria sp.) Berbasis Deep Learning

      Thumbnail
      View/Open
      Cover (301.9Kb)
      Fulltext (1.394Mb)
      Lampiran (252.1Kb)
      Date
      2025
      Author
      Hazimah, Nabilah Nur
      Noorachmat, Bambang Pramudya
      Solahudin, Mohamad
      Metadata
      Show full item record
      Abstract
      Stroberi merupakan salah satu komoditas buah-buahan yang mendorong perekonomian di Indonesia. Tanaman stroberi rentan terkena penyakit, oleh karena itu mengidentifikasi penyakit pada tanaman sejak dini merupakan hal yang penting, namun bagi orang awam akan sulit untuk mengidentifikasi penyakit yang sedang menjangkiti tanaman tersebut, sehingga identifikasi penyakit menjadi lambat, efisiensinya rendah, dan bahkan bisa menjadi gagal panen. Berdasarkan hal tersebut, dibutuhkan metode deep learning yang dapat mengenali objek secara cepat, akurat, dan presisi, dengan bantuan kamera RGB. Penelitian ini bertujuan untuk membuat model deep learning yaitu You Only Look Once (YOLO) dengan akurat dan presisi dalam mengidentifikasi penyakit daun tanaman stroberi, di mana penelitian ini dapat bermanfaat untuk memudahkan petani maupun industri dalam pencegahan dan mendeteksi penyakit sejak dini pada daun stroberi. Penyakit pada daun tanaman stroberi dibagi menjadi 3 kelas: hawar daun, daun hangus, dan bercak daun. Tahapan awal pada penelitian ini dilakukan dengan pengumpulan data berupa gambar daun- daun stroberi yang terjangkit penyakit, kemudian data yang diperoleh akan dianotasikan menggunakan bounding box agar program dapat mengenali objek dengan baik. Dataset yang telah dianotasi selanjutnya dilakukan pelatihan dimana hasil keluaran dari pelatihan yaitu nilai bobot. Nilai bobot digunakan untuk melakukan evaluasi kinerja model dan hasil dari evaluasi didapatkan rata-rata nilai keseluruhan sebesar 93,15% accuracy, 87,67% precision, 91,64% recall, dan 89,43% F1-score. Berdasarkan hasil evaluasi tersebut dapat dinyatakan bahwa model memiliki perfoma yang baik. Model tersebut diimplementasikan dalam bentuk website menggunakan HTML, CSS, dan Flask, untuk mempermudah user dalam menggunakan aplikasi.
       
      Strawberries are one of the key fruit commodities that drive the economy in Indonesia. However, strawberry plants are highly susceptible to diseases. Early identification of plant diseases is essential, but for ordinary people, it is often challenging to accurately diagnose the disease affecting the plants. This difficulty can lead to delays in identification, low efficiency, and even crop failure. Based on this, a deep learning method is needed that can recognize objects quickly, accurately, and precisely, with the help of RGB cameras. This research aimed to develop a deep learning model, specifically You Only Look Once (YOLO), to achieve high accuracy and precision in identifying strawberry plant leaf diseases. The findings of this research could assist farmers and industries in the early detection and prevention of diseases in strawberry foliage. The diseases affecting strawberry plant leaves are categorized into 3 class: leaf blight, leaf scorch, and leaf spot. The initial stage of this research involved collecting data in the form of images of diseased strawberry leaves. The collected data were then annotated using bounding boxes, enabling the program to accurately recognize the objects. The annotated dataset was used for training, and the output of the training was the weight values. These weight values were subsequently used to evaluate the model’s performance. The evaluation results showed an overall accuracy of 93.15%, precision of 87.67%, recall of 91.64%, and F1-score of 89.43%. Based on the results obtained, the model exhibited strong performance. The model has been implemented as a website using HTML, CSS, and Flask, providing a user-friendly application for disease detection and prevention in strawberry plants.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/161413
      Collections
      • UT - Agricultural and Biosystem Engineering [3592]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository